Merge pull request #7313 from freqtrade/new_release

New release 2022.8
This commit is contained in:
Matthias 2022-08-30 23:01:19 +02:00 committed by GitHub
commit 958a4565db
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
156 changed files with 10546 additions and 1782 deletions

View File

@ -351,7 +351,7 @@ jobs:
python setup.py sdist bdist_wheel
- name: Publish to PyPI (Test)
uses: pypa/gh-action-pypi-publish@v1.5.0
uses: pypa/gh-action-pypi-publish@v1.5.1
if: (github.event_name == 'release')
with:
user: __token__
@ -359,7 +359,7 @@ jobs:
repository_url: https://test.pypi.org/legacy/
- name: Publish to PyPI
uses: pypa/gh-action-pypi-publish@v1.5.0
uses: pypa/gh-action-pypi-publish@v1.5.1
if: (github.event_name == 'release')
with:
user: __token__

6
.gitignore vendored
View File

@ -7,10 +7,15 @@ logfile.txt
user_data/*
!user_data/strategy/sample_strategy.py
!user_data/notebooks
!user_data/models
!user_data/freqaimodels
user_data/freqaimodels/*
user_data/models/*
user_data/notebooks/*
freqtrade-plot.html
freqtrade-profit-plot.html
freqtrade/rpc/api_server/ui/*
build_helpers/ta-lib/*
# Macos related
.DS_Store
@ -107,3 +112,4 @@ target/
!config_examples/config_ftx.example.json
!config_examples/config_full.example.json
!config_examples/config_kraken.example.json
!config_examples/config_freqai.example.json

View File

@ -15,7 +15,7 @@ repos:
additional_dependencies:
- types-cachetools==5.2.1
- types-filelock==3.2.7
- types-requests==2.28.3
- types-requests==2.28.9
- types-tabulate==0.8.11
- types-python-dateutil==2.8.19
# stages: [push]

View File

@ -1,4 +1,4 @@
FROM python:3.10.5-slim-bullseye as base
FROM python:3.10.6-slim-bullseye as base
# Setup env
ENV LANG C.UTF-8
@ -11,7 +11,7 @@ ENV FT_APP_ENV="docker"
# Prepare environment
RUN mkdir /freqtrade \
&& apt-get update \
&& apt-get -y install sudo libatlas3-base curl sqlite3 libhdf5-serial-dev \
&& apt-get -y install sudo libatlas3-base curl sqlite3 libhdf5-serial-dev libgomp1 \
&& apt-get clean \
&& useradd -u 1000 -G sudo -U -m -s /bin/bash ftuser \
&& chown ftuser:ftuser /freqtrade \

View File

@ -63,6 +63,7 @@ Please find the complete documentation on the [freqtrade website](https://www.fr
- [x] **Dry-run**: Run the bot without paying money.
- [x] **Backtesting**: Run a simulation of your buy/sell strategy.
- [x] **Strategy Optimization by machine learning**: Use machine learning to optimize your buy/sell strategy parameters with real exchange data.
- [X] **Adaptive prediction modeling**: Build a smart strategy with FreqAI that self-trains to the market via adaptive machine learning methods. [Learn more](https://www.freqtrade.io/en/stable/freqai/)
- [x] **Edge position sizing** Calculate your win rate, risk reward ratio, the best stoploss and adjust your position size before taking a position for each specific market. [Learn more](https://www.freqtrade.io/en/stable/edge/).
- [x] **Whitelist crypto-currencies**: Select which crypto-currency you want to trade or use dynamic whitelists.
- [x] **Blacklist crypto-currencies**: Select which crypto-currency you want to avoid.
@ -129,7 +130,7 @@ Telegram is not mandatory. However, this is a great way to control your bot. Mor
- `/start`: Starts the trader.
- `/stop`: Stops the trader.
- `/stopbuy`: Stop entering new trades.
- `/stopentry`: Stop entering new trades.
- `/status <trade_id>|[table]`: Lists all or specific open trades.
- `/profit [<n>]`: Lists cumulative profit from all finished trades, over the last n days.
- `/forceexit <trade_id>|all`: Instantly exits the given trade (Ignoring `minimum_roi`).
@ -193,7 +194,7 @@ Issues labeled [good first issue](https://github.com/freqtrade/freqtrade/labels/
The clock must be accurate, synchronized to a NTP server very frequently to avoid problems with communication to the exchanges.
### Min hardware required
### Minimum hardware required
To run this bot we recommend you a cloud instance with a minimum of:

View File

@ -4,7 +4,7 @@ else
INSTALL_LOC=${1}
fi
echo "Installing to ${INSTALL_LOC}"
if [ ! -f "${INSTALL_LOC}/lib/libta_lib.a" ]; then
if [ -n "$2" ] || [ ! -f "${INSTALL_LOC}/lib/libta_lib.a" ]; then
tar zxvf ta-lib-0.4.0-src.tar.gz
cd ta-lib \
&& sed -i.bak "s|0.00000001|0.000000000000000001 |g" src/ta_func/ta_utility.h \
@ -17,11 +17,17 @@ if [ ! -f "${INSTALL_LOC}/lib/libta_lib.a" ]; then
cd .. && rm -rf ./ta-lib/
exit 1
fi
if [ -z "$2" ]; then
which sudo && sudo make install || make install
if [ -x "$(command -v apt-get)" ]; then
echo "Updating library path using ldconfig"
sudo ldconfig
fi
else
# Don't install with sudo
make install
fi
cd .. && rm -rf ./ta-lib/
else
echo "TA-lib already installed, skipping installation"

View File

@ -6,10 +6,12 @@ export DOCKER_BUILDKIT=1
# Replace / with _ to create a valid tag
TAG=$(echo "${BRANCH_NAME}" | sed -e "s/\//_/g")
TAG_PLOT=${TAG}_plot
TAG_FREQAI=${TAG}_freqai
TAG_PI="${TAG}_pi"
TAG_ARM=${TAG}_arm
TAG_PLOT_ARM=${TAG_PLOT}_arm
TAG_FREQAI_ARM=${TAG_FREQAI}_arm
CACHE_IMAGE=freqtradeorg/freqtrade_cache
echo "Running for ${TAG}"
@ -38,8 +40,10 @@ fi
docker tag freqtrade:$TAG_ARM ${CACHE_IMAGE}:$TAG_ARM
docker build --cache-from freqtrade:${TAG_ARM} --build-arg sourceimage=${CACHE_IMAGE} --build-arg sourcetag=${TAG_ARM} -t freqtrade:${TAG_PLOT_ARM} -f docker/Dockerfile.plot .
docker build --cache-from freqtrade:${TAG_ARM} --build-arg sourceimage=${CACHE_IMAGE} --build-arg sourcetag=${TAG_ARM} -t freqtrade:${TAG_FREQAI_ARM} -f docker/Dockerfile.freqai .
docker tag freqtrade:$TAG_PLOT_ARM ${CACHE_IMAGE}:$TAG_PLOT_ARM
docker tag freqtrade:$TAG_FREQAI_ARM ${CACHE_IMAGE}:$TAG_FREQAI_ARM
# Run backtest
docker run --rm -v $(pwd)/config_examples/config_bittrex.example.json:/freqtrade/config.json:ro -v $(pwd)/tests:/tests freqtrade:${TAG_ARM} backtesting --datadir /tests/testdata --strategy-path /tests/strategy/strats/ --strategy StrategyTestV3
@ -53,6 +57,7 @@ docker images
# docker push ${IMAGE_NAME}
docker push ${CACHE_IMAGE}:$TAG_PLOT_ARM
docker push ${CACHE_IMAGE}:$TAG_FREQAI_ARM
docker push ${CACHE_IMAGE}:$TAG_ARM
# Create multi-arch image
@ -66,6 +71,9 @@ docker manifest push -p ${IMAGE_NAME}:${TAG}
docker manifest create ${IMAGE_NAME}:${TAG_PLOT} ${CACHE_IMAGE}:${TAG_PLOT_ARM} ${CACHE_IMAGE}:${TAG_PLOT}
docker manifest push -p ${IMAGE_NAME}:${TAG_PLOT}
docker manifest create ${IMAGE_NAME}:${TAG_FREQAI} ${CACHE_IMAGE}:${TAG_FREQAI_ARM} ${CACHE_IMAGE}:${TAG_FREQAI}
docker manifest push -p ${IMAGE_NAME}:${TAG_FREQAI}
# Tag as latest for develop builds
if [ "${TAG}" = "develop" ]; then
docker manifest create ${IMAGE_NAME}:latest ${CACHE_IMAGE}:${TAG_ARM} ${IMAGE_NAME}:${TAG_PI} ${CACHE_IMAGE}:${TAG}

View File

@ -5,6 +5,7 @@
# Replace / with _ to create a valid tag
TAG=$(echo "${BRANCH_NAME}" | sed -e "s/\//_/g")
TAG_PLOT=${TAG}_plot
TAG_FREQAI=${TAG}_freqai
TAG_PI="${TAG}_pi"
PI_PLATFORM="linux/arm/v7"
@ -49,8 +50,10 @@ fi
docker tag freqtrade:$TAG ${CACHE_IMAGE}:$TAG
docker build --cache-from freqtrade:${TAG} --build-arg sourceimage=${CACHE_IMAGE} --build-arg sourcetag=${TAG} -t freqtrade:${TAG_PLOT} -f docker/Dockerfile.plot .
docker build --cache-from freqtrade:${TAG} --build-arg sourceimage=${CACHE_IMAGE} --build-arg sourcetag=${TAG} -t freqtrade:${TAG_FREQAI} -f docker/Dockerfile.freqai .
docker tag freqtrade:$TAG_PLOT ${CACHE_IMAGE}:$TAG_PLOT
docker tag freqtrade:$TAG_FREQAI ${CACHE_IMAGE}:$TAG_FREQAI
# Run backtest
docker run --rm -v $(pwd)/config_examples/config_bittrex.example.json:/freqtrade/config.json:ro -v $(pwd)/tests:/tests freqtrade:${TAG} backtesting --datadir /tests/testdata --strategy-path /tests/strategy/strats/ --strategy StrategyTestV3
@ -64,6 +67,7 @@ docker images
docker push ${CACHE_IMAGE}
docker push ${CACHE_IMAGE}:$TAG_PLOT
docker push ${CACHE_IMAGE}:$TAG_FREQAI
docker push ${CACHE_IMAGE}:$TAG

View File

@ -0,0 +1,96 @@
{
"trading_mode": "futures",
"margin_mode": "isolated",
"max_open_trades": 5,
"stake_currency": "USDT",
"stake_amount": 200,
"tradable_balance_ratio": 1,
"fiat_display_currency": "USD",
"dry_run": true,
"timeframe": "3m",
"dry_run_wallet": 1000,
"cancel_open_orders_on_exit": true,
"unfilledtimeout": {
"entry": 10,
"exit": 30
},
"exchange": {
"name": "binance",
"key": "",
"secret": "",
"ccxt_config": {
"enableRateLimit": true
},
"ccxt_async_config": {
"enableRateLimit": true,
"rateLimit": 200
},
"pair_whitelist": [
"1INCH/USDT",
"ALGO/USDT"
],
"pair_blacklist": []
},
"entry_pricing": {
"price_side": "same",
"use_order_book": true,
"order_book_top": 1,
"price_last_balance": 0.0,
"check_depth_of_market": {
"enabled": false,
"bids_to_ask_delta": 1
}
},
"exit_pricing": {
"price_side": "other",
"use_order_book": true,
"order_book_top": 1
},
"pairlists": [
{
"method": "StaticPairList"
}
],
"freqai": {
"enabled": true,
"startup_candles": 10000,
"purge_old_models": true,
"train_period_days": 15,
"backtest_period_days": 7,
"live_retrain_hours": 0,
"identifier": "uniqe-id",
"feature_parameters": {
"include_timeframes": [
"3m",
"15m",
"1h"
],
"include_corr_pairlist": [
"BTC/USDT",
"ETH/USDT"
],
"label_period_candles": 20,
"include_shifted_candles": 2,
"DI_threshold": 0.9,
"weight_factor": 0.9,
"principal_component_analysis": false,
"use_SVM_to_remove_outliers": true,
"stratify_training_data": 0,
"indicator_max_period_candles": 20,
"indicator_periods_candles": [10, 20]
},
"data_split_parameters": {
"test_size": 0.33,
"random_state": 1
},
"model_training_parameters": {
"n_estimators": 1000
}
},
"bot_name": "",
"force_entry_enable": true,
"initial_state": "running",
"internals": {
"process_throttle_secs": 5
}
}

View File

@ -5,6 +5,7 @@
"tradable_balance_ratio": 0.99,
"fiat_display_currency": "USD",
"amount_reserve_percent": 0.05,
"available_capital": 1000,
"amend_last_stake_amount": false,
"last_stake_amount_min_ratio": 0.5,
"dry_run": true,
@ -92,6 +93,7 @@
"secret": "your_exchange_secret",
"password": "",
"log_responses": false,
// "unknown_fee_rate": 1,
"ccxt_config": {},
"ccxt_async_config": {},
"pair_whitelist": [

9
docker/Dockerfile.freqai Normal file
View File

@ -0,0 +1,9 @@
ARG sourceimage=freqtradeorg/freqtrade
ARG sourcetag=develop
FROM ${sourceimage}:${sourcetag}
# Install dependencies
COPY requirements-freqai.txt /freqtrade/
RUN pip install -r requirements-freqai.txt --user --no-cache-dir

BIN
docs/assets/freqai_DI.jpg Normal file

Binary file not shown.

After

Width:  |  Height:  |  Size: 307 KiB

BIN
docs/assets/freqai_algo.jpg Normal file

Binary file not shown.

After

Width:  |  Height:  |  Size: 345 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 66 KiB

File diff suppressed because one or more lines are too long

After

Width:  |  Height:  |  Size: 2.0 MiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 270 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 191 KiB

View File

@ -514,6 +514,7 @@ You can then load the trades to perform further analysis as shown in the [data a
Since backtesting lacks some detailed information about what happens within a candle, it needs to take a few assumptions:
- Exchange [trading limits](#trading-limits-in-backtesting) are respected
- Buys happen at open-price
- All orders are filled at the requested price (no slippage, no unfilled orders)
- Exit-signal exits happen at open-price of the consecutive candle
@ -543,7 +544,32 @@ Also, keep in mind that past results don't guarantee future success.
In addition to the above assumptions, strategy authors should carefully read the [Common Mistakes](strategy-customization.md#common-mistakes-when-developing-strategies) section, to avoid using data in backtesting which is not available in real market conditions.
### Improved backtest accuracy
### Trading limits in backtesting
Exchanges have certain trading limits, like minimum base currency, or minimum stake (quote) currency.
These limits are usually listed in the exchange documentation as "trading rules" or similar.
Backtesting (as well as live and dry-run) does honor these limits, and will ensure that a stoploss can be placed below this value - so the value will be slightly higher than what the exchange specifies.
Freqtrade has however no information about historic limits.
This can lead to situations where trading-limits are inflated by using a historic price, resulting in minimum amounts > 50$.
For example:
BTC minimum tradable amount is 0.001.
BTC trades at 22.000\$ today (0.001 BTC is related to this) - but the backtesting period includes prices as high as 50.000\$.
Today's minimum would be `0.001 * 22_000` - or 22\$.
However the limit could also be 50$ - based on `0.001 * 50_000` in some historic setting.
#### Trading precision limits
Most exchanges pose precision limits on both price and amounts, so you cannot buy 1.0020401 of a pair, or at a price of 1.24567123123.
Instead, these prices and amounts will be rounded or truncated (based on the exchange definition) to the defined trading precision.
The above values may for example be rounded to an amount of 1.002, and a price of 1.24567.
These precision values are based on current exchange limits (as described in the [above section](#trading-limits-in-backtesting)), as historic precision limits are not available.
## Improved backtest accuracy
One big limitation of backtesting is it's inability to know how prices moved intra-candle (was high before close, or viceversa?).
So assuming you run backtesting with a 1h timeframe, there will be 4 prices for that candle (Open, High, Low, Close).

View File

@ -70,7 +70,7 @@ This loop will be repeated again and again until the bot is stopped.
* Determine stake size by calling the `custom_stake_amount()` callback.
* Check position adjustments for open trades if enabled and call `adjust_trade_position()` to determine if an additional order is requested.
* Call `custom_stoploss()` and `custom_exit()` to find custom exit points.
* For exits based on exit-signal and custom-exit: Call `custom_exit_price()` to determine exit price (Prices are moved to be within the closing candle).
* For exits based on exit-signal, custom-exit and partial exits: Call `custom_exit_price()` to determine exit price (Prices are moved to be within the closing candle).
* Generate backtest report output
!!! Note

View File

@ -105,7 +105,7 @@ This is similar to using multiple `--config` parameters, but simpler in usage as
``` json title="Result"
{
"max_open_trades": 10,
"max_open_trades": 3,
"stake_currency": "USDT",
"stake_amount": "unlimited"
}

View File

@ -63,7 +63,7 @@ optional arguments:
`jsongz`).
--trading-mode {spot,margin,futures}
Select Trading mode
--prepend Allow data prepending.
--prepend Allow data prepending. (Data-appending is disabled)
Common arguments:
-v, --verbose Verbose mode (-vv for more, -vvv to get all messages).
@ -186,7 +186,7 @@ Freqtrade currently supports 3 data-formats for both OHLCV and trades data:
By default, OHLCV data is stored as `json` data, while trades data is stored as `jsongz` data.
This can be changed via the `--data-format-ohlcv` and `--data-format-trades` command line arguments respectively.
To persist this change, you can should also add the following snippet to your configuration, so you don't have to insert the above arguments each time:
To persist this change, you should also add the following snippet to your configuration, so you don't have to insert the above arguments each time:
``` jsonc
// ...
@ -374,6 +374,7 @@ usage: freqtrade list-data [-h] [-v] [--logfile FILE] [-V] [-c PATH] [-d PATH]
[--data-format-ohlcv {json,jsongz,hdf5}]
[-p PAIRS [PAIRS ...]]
[--trading-mode {spot,margin,futures}]
[--show-timerange]
optional arguments:
-h, --help show this help message and exit
@ -387,6 +388,8 @@ optional arguments:
separated.
--trading-mode {spot,margin,futures}
Select Trading mode
--show-timerange Show timerange available for available data. (May take
a while to calculate).
Common arguments:
-v, --verbose Verbose mode (-vv for more, -vvv to get all messages).

View File

@ -68,6 +68,36 @@ def test_method_to_test(caplog):
```
### Debug configuration
To debug freqtrade, we recommend VSCode with the following launch configuration (located in `.vscode/launch.json`).
Details will obviously vary between setups - but this should work to get you started.
``` json
{
"name": "freqtrade trade",
"type": "python",
"request": "launch",
"module": "freqtrade",
"console": "integratedTerminal",
"args": [
"trade",
// Optional:
// "--userdir", "user_data",
"--strategy",
"MyAwesomeStrategy",
]
},
```
Command line arguments can be added in the `"args"` array.
This method can also be used to debug a strategy, by setting the breakpoints within the strategy.
A similar setup can also be taken for Pycharm - using `freqtrade` as module name, and setting the command line arguments as "parameters".
!!! Note "Startup directory"
This assumes that you have the repository checked out, and the editor is started at the repository root level (so setup.py is at the top level of your repository).
## ErrorHandling
Freqtrade Exceptions all inherit from `FreqtradeException`.

View File

@ -77,9 +77,9 @@ Freqtrade will not provide incomplete candles to strategies. Using incomplete ca
You can use "current" market data by using the [dataprovider](strategy-customization.md#orderbookpair-maximum)'s orderbook or ticker methods - which however cannot be used during backtesting.
### Is there a setting to only SELL the coins being held and not perform anymore BUYS?
### Is there a setting to only Exit the trades being held and not perform any new Entries?
You can use the `/stopbuy` command in Telegram to prevent future buys, followed by `/forceexit all` (sell all open trades).
You can use the `/stopentry` command in Telegram to prevent future trade entry, followed by `/forceexit all` (sell all open trades).
### I want to run multiple bots on the same machine

759
docs/freqai.md Normal file
View File

@ -0,0 +1,759 @@
![freqai-logo](assets/freqai_doc_logo.svg)
# FreqAI
FreqAI is a module designed to automate a variety of tasks associated with training a predictive machine learning model to generate market forecasts given a set of input features.
Features include:
* **Self-adaptive retraining**: retrain models during [live deployments](#running-the-model-live) to self-adapt to the market in an unsupervised manner.
* **Rapid feature engineering**: create large rich [feature sets](#feature-engineering) (10k+ features) based on simple user-created strategies.
* **High performance**: adaptive retraining occurs on a separate thread (or on GPU if available) from inferencing and bot trade operations. Newest models and data are kept in memory for rapid inferencing.
* **Realistic backtesting**: emulate self-adaptive retraining with a [backtesting module](#backtesting) that automates past retraining.
* **Modifiability**: use the generalized and robust architecture for incorporating any [machine learning library/method](#building-a-custom-prediction-model) available in Python. Eight examples are currently available, including classifiers, regressors, and a convolutional neural network.
* **Smart outlier removal**: remove outliers from training and prediction data sets using a variety of [outlier detection techniques](#outlier-removal).
* **Crash resilience**: store model to disk to make reloading from a crash fast and easy, and [purge obsolete files](#purging-old-model-data) for sustained dry/live runs.
* **Automatic data normalization**: [normalize the data](#feature-normalization) in a smart and statistically safe way.
* **Automatic data download**: compute the data download timerange and update historic data (in live deployments).
* **Cleaning of incoming data**: handle NaNs safely before training and prediction.
* **Dimensionality reduction**: reduce the size of the training data via [Principal Component Analysis](#reducing-data-dimensionality-with-principal-component-analysis).
* **Deploying bot fleets**: set one bot to train models while a fleet of [follower bots](#setting-up-a-follower) inference the models and handle trades.
## Quick start
The easiest way to quickly test FreqAI is to run it in dry mode with the following command
```bash
freqtrade trade --config config_examples/config_freqai.example.json --strategy FreqaiExampleStrategy --freqaimodel LightGBMRegressor --strategy-path freqtrade/templates
```
The user will see the boot-up process of automatic data downloading, followed by simultaneous training and trading.
The example strategy, example prediction model, and example config can be found in
`freqtrade/templates/FreqaiExampleStrategy.py`, `freqtrade/freqai/prediction_models/LightGBMRegressor.py`, and
`config_examples/config_freqai.example.json`, respectively.
## General approach
The user provides FreqAI with a set of custom *base* indicators (the same way as in a typical Freqtrade strategy) as well as target values (*labels*).
FreqAI trains a model to predict the target values based on the input of custom indicators, for each pair in the whitelist. These models are consistently retrained to adapt to market conditions. FreqAI offers the ability to both backtest strategies (emulating reality with periodic retraining) and deploy dry/live runs. In dry/live conditions, FreqAI can be set to constant retraining in a background thread in an effort to keep models as up to date as possible.
An overview of the algorithm is shown below, explaining the data processing pipeline and the model usage.
![freqai-algo](assets/freqai_algo.jpg)
### Important machine learning vocabulary
**Features** - the quantities with which a model is trained. All features for a single candle is stored as a vector. In FreqAI, the user
builds the feature sets from anything they can construct in the strategy.
**Labels** - the target values that a model is trained
toward. Each set of features is associated with a single label that is
defined by the user within the strategy. These labels intentionally look into the
future, and are not available to the model during dry/live/backtesting.
**Training** - the process of feeding individual feature sets, composed of historic data, with associated labels into the
model with the goal of matching input feature sets to associated labels.
**Train data** - a subset of the historic data that is fed to the model during
training. This data directly influences weight connections in the model.
**Test data** - a subset of the historic data that is used to evaluate the performance of the model after training. This data does not influence nodal weights within the model.
## Install prerequisites
The normal Freqtrade install process will ask the user if they wish to install FreqAI dependencies. The user should reply "yes" to this question if they wish to use FreqAI. If the user did not reply yes, they can manually install these dependencies after the install with:
``` bash
pip install -r requirements-freqai.txt
```
!!! Note
Catboost will not be installed on arm devices (raspberry, Mac M1, ARM based VPS, ...), since Catboost does not provide wheels for this platform.
### Usage with docker
For docker users, a dedicated tag with freqAI dependencies is available as `:freqai`.
As such - you can replace the image line in your docker-compose file with `image: freqtradeorg/freqtrade:develop_freqai`.
This image contains the regular freqAI dependencies. Similar to native installs, Catboost will not be available on ARM based devices.
## Setting up FreqAI
### Parameter table
The table below will list all configuration parameters available for FreqAI, presented in the same order as `config_examples/config_freqai.example.json`.
Mandatory parameters are marked as **Required**, which means that they are required to be set in one of the possible ways.
| Parameter | Description |
|------------|-------------|
| | **General configuration parameters**
| `freqai` | **Required.** <br> The parent dictionary containing all the parameters for controlling FreqAI. <br> **Datatype:** Dictionary.
| `startup_candles` | Number of candles needed for *backtesting only* to ensure all indicators are non NaNs at the start of the first train period. <br> **Datatype:** Positive integer.
| `purge_old_models` | Delete obsolete models (otherwise, all historic models will remain on disk). <br> **Datatype:** Boolean. Default: `False`.
| `train_period_days` | **Required.** <br> Number of days to use for the training data (width of the sliding window). <br> **Datatype:** Positive integer.
| `backtest_period_days` | **Required.** <br> Number of days to inference from the trained model before sliding the window defined above, and retraining the model. This can be fractional days, but beware that the user-provided `timerange` will be divided by this number to yield the number of trainings necessary to complete the backtest. <br> **Datatype:** Float.
| `identifier` | **Required.** <br> A unique name for the current model. This can be reused to reload pre-trained models/data. <br> **Datatype:** String.
| `live_retrain_hours` | Frequency of retraining during dry/live runs. <br> Default set to 0, which means the model will retrain as often as possible. <br> **Datatype:** Float > 0.
| `expiration_hours` | Avoid making predictions if a model is more than `expiration_hours` old. <br> Defaults set to 0, which means models never expire. <br> **Datatype:** Positive integer.
| `fit_live_predictions_candles` | Number of historical candles to use for computing target (label) statistics from prediction data, instead of from the training data set. <br> **Datatype:** Positive integer.
| `follow_mode` | If true, this instance of FreqAI will look for models associated with `identifier` and load those for inferencing. A `follower` will **not** train new models. <br> **Datatype:** Boolean. Default: `False`.
| | **Feature parameters**
| `feature_parameters` | A dictionary containing the parameters used to engineer the feature set. Details and examples are shown [here](#feature-engineering). <br> **Datatype:** Dictionary.
| `include_timeframes` | A list of timeframes that all indicators in `populate_any_indicators` will be created for. The list is added as features to the base asset feature set. <br> **Datatype:** List of timeframes (strings).
| `include_corr_pairlist` | A list of correlated coins that FreqAI will add as additional features to all `pair_whitelist` coins. All indicators set in `populate_any_indicators` during feature engineering (see details [here](#feature-engineering)) will be created for each coin in this list, and that set of features is added to the base asset feature set. <br> **Datatype:** List of assets (strings).
| `label_period_candles` | Number of candles into the future that the labels are created for. This is used in `populate_any_indicators` (see `templates/FreqaiExampleStrategy.py` for detailed usage). The user can create custom labels, making use of this parameter or not. <br> **Datatype:** Positive integer.
| `include_shifted_candles` | Add features from previous candles to subsequent candles to add historical information. FreqAI takes all features from the `include_shifted_candles` previous candles, duplicates and shifts them so that the information is available for the subsequent candle. <br> **Datatype:** Positive integer.
| `weight_factor` | Used to set weights for training data points according to their recency. See details about how it works [here](#controlling-the-model-learning-process). <br> **Datatype:** Positive float (typically < 1).
| `indicator_max_period_candles` | The maximum period used in `populate_any_indicators()` for indicator creation. FreqAI uses this information in combination with the maximum timeframe to calculate how many data points that should be downloaded so that the first data point does not have a NaN. <br> **Datatype:** Positive integer.
| `indicator_periods_candles` | Calculate indicators for `indicator_periods_candles` time periods and add them to the feature set. <br> **Datatype:** List of positive integers.
| `stratify_training_data` | This value is used to indicate the grouping of the data. For example, 2 would set every 2nd data point into a separate dataset to be pulled from during training/testing. See details about how it works [here](#stratifying-the-data-for-training-and-testing-the-model) <br> **Datatype:** Positive integer.
| `principal_component_analysis` | Automatically reduce the dimensionality of the data set using Principal Component Analysis. See details about how it works [here](#reducing-data-dimensionality-with-principal-component-analysis) <br> **Datatype:** Boolean.
| `DI_threshold` | Activates the Dissimilarity Index for outlier detection when > 0. See details about how it works [here](#removing-outliers-with-the-dissimilarity-index). <br> **Datatype:** Positive float (typically < 1).
| `use_SVM_to_remove_outliers` | Train a support vector machine to detect and remove outliers from the training data set, as well as from incoming data points. See details about how it works [here](#removing-outliers-using-a-support-vector-machine-svm). <br> **Datatype:** Boolean.
| `svm_params` | All parameters available in Sklearn's `SGDOneClassSVM()`. See details about some select parameters [here](#removing-outliers-using-a-support-vector-machine-svm). <br> **Datatype:** Dictionary.
| `use_DBSCAN_to_remove_outliers` | Cluster data using DBSCAN to identify and remove outliers from training and prediction data. See details about how it works [here](#removing-outliers-with-dbscan). <br> **Datatype:** Boolean.
| `outlier_protection_percentage` | If more than `outlier_protection_percentage` fraction of points are removed as outliers, FreqAI will log a warning message and ignore outlier detection while keeping the original dataset intact. <br> **Datatype:** float. Default: `30`
| | **Data split parameters**
| `data_split_parameters` | Include any additional parameters available from Scikit-learn `test_train_split()`, which are shown [here](https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.train_test_split.html) (external website). <br> **Datatype:** Dictionary.
| `test_size` | Fraction of data that should be used for testing instead of training. <br> **Datatype:** Positive float < 1.
| `shuffle` | Shuffle the training data points during training. Typically, for time-series forecasting, this is set to `False`. <br>
| | **Model training parameters**
| `model_training_parameters` | A flexible dictionary that includes all parameters available by the user selected model library. For example, if the user uses `LightGBMRegressor`, this dictionary can contain any parameter available by the `LightGBMRegressor` [here](https://lightgbm.readthedocs.io/en/latest/pythonapi/lightgbm.LGBMRegressor.html) (external website). If the user selects a different model, this dictionary can contain any parameter from that model. <br> **Datatype:** Dictionary.**Datatype:** Boolean.
| `n_estimators` | The number of boosted trees to fit in regression. <br> **Datatype:** Integer.
| `learning_rate` | Boosting learning rate during regression. <br> **Datatype:** Float.
| `n_jobs`, `thread_count`, `task_type` | Set the number of threads for parallel processing and the `task_type` (`gpu` or `cpu`). Different model libraries use different parameter names. <br> **Datatype:** Float.
| | **Extraneous parameters**
| `keras` | If your model makes use of Keras (typical for Tensorflow-based prediction models), activate this flag so that the model save/loading follows Keras standards. <br> **Datatype:** Boolean. Default: `False`.
| `conv_width` | The width of a convolutional neural network input tensor. This replaces the need for shifting candles (`include_shifted_candles`) by feeding in historical data points as the second dimension of the tensor. Technically, this parameter can also be used for regressors, but it only adds computational overhead and does not change the model training/prediction. <br> **Datatype:** Integer. Default: 2.
### Important dataframe key patterns
Below are the values the user can expect to include/use inside a typical strategy dataframe (`df[]`):
| DataFrame Key | Description |
|------------|-------------|
| `df['&*']` | Any dataframe column prepended with `&` in `populate_any_indicators()` is treated as a training target (label) inside FreqAI (typically following the naming convention `&-s*`). The names of these dataframe columns are fed back to the user as the predictions. For example, if the user wishes to predict the price change in the next 40 candles (similar to `templates/FreqaiExampleStrategy.py`), they set `df['&-s_close']`. FreqAI makes the predictions and gives them back under the same key (`df['&-s_close']`) to be used in `populate_entry/exit_trend()`. <br> **Datatype:** Depends on the output of the model.
| `df['&*_std/mean']` | Standard deviation and mean values of the user-defined labels during training (or live tracking with `fit_live_predictions_candles`). Commonly used to understand the rarity of a prediction (use the z-score as shown in `templates/FreqaiExampleStrategy.py` to evaluate how often a particular prediction was observed during training or historically with `fit_live_predictions_candles`). <br> **Datatype:** Float.
| `df['do_predict']` | Indication of an outlier data point. The return value is integer between -1 and 2, which lets the user know if the prediction is trustworthy or not. `do_predict==1` means the prediction is trustworthy. If the Dissimilarity Index (DI, see details [here](#removing-outliers-with-the-dissimilarity-index)) of the input data point is above the user-defined threshold, FreqAI will subtract 1 from `do_predict`, resulting in `do_predict==0`. If `use_SVM_to_remove_outliers()` is active, the Support Vector Machine (SVM) may also detect outliers in training and prediction data. In this case, the SVM will also subtract 1 from `do_predict`. If the input data point was considered an outlier by the SVM but not by the DI, the result will be `do_predict==0`. If both the DI and the SVM considers the input data point to be an outlier, the result will be `do_predict==-1`. A particular case is when `do_predict == 2`, which means that the model has expired due to exceeding `expired_hours`. <br> **Datatype:** Integer between -1 and 2.
| `df['DI_values']` | Dissimilarity Index values are proxies to the level of confidence FreqAI has in the prediction. A lower DI means the prediction is close to the training data, i.e., higher prediction confidence. <br> **Datatype:** Float.
| `df['%*']` | Any dataframe column prepended with `%` in `populate_any_indicators()` is treated as a training feature. For example, the user can include the RSI in the training feature set (similar to in `templates/FreqaiExampleStrategy.py`) by setting `df['%-rsi']`. See more details on how this is done [here](#feature-engineering). <br> **Note**: Since the number of features prepended with `%` can multiply very quickly (10s of thousands of features is easily engineered using the multiplictative functionality described in the `feature_parameters` table shown above), these features are removed from the dataframe upon return from FreqAI. If the user wishes to keep a particular type of feature for plotting purposes, they can prepend it with `%%`. <br> **Datatype:** Depends on the output of the model.
### File structure
`user_data_dir/models/` contains all the data associated with the trainings and backtests.
This file structure is heavily controlled and inferenced by the `FreqaiDataKitchen()`
and should therefore not be modified.
### Example config file
The user interface is isolated to the typical Freqtrade config file. A FreqAI config should include:
```json
"freqai": {
"enabled": true,
"startup_candles": 10000,
"purge_old_models": true,
"train_period_days": 30,
"backtest_period_days": 7,
"identifier" : "unique-id",
"feature_parameters" : {
"include_timeframes": ["5m","15m","4h"],
"include_corr_pairlist": [
"ETH/USD",
"LINK/USD",
"BNB/USD"
],
"label_period_candles": 24,
"include_shifted_candles": 2,
"indicator_max_period_candles": 20,
"indicator_periods_candles": [10, 20]
},
"data_split_parameters" : {
"test_size": 0.25
},
"model_training_parameters" : {
"n_estimators": 100
},
}
```
## Building a FreqAI strategy
The FreqAI strategy requires the user to include the following lines of code in the standard Freqtrade strategy:
```python
def informative_pairs(self):
whitelist_pairs = self.dp.current_whitelist()
corr_pairs = self.config["freqai"]["feature_parameters"]["include_corr_pairlist"]
informative_pairs = []
for tf in self.config["freqai"]["feature_parameters"]["include_timeframes"]:
for pair in whitelist_pairs:
informative_pairs.append((pair, tf))
for pair in corr_pairs:
if pair in whitelist_pairs:
continue # avoid duplication
informative_pairs.append((pair, tf))
return informative_pairs
def populate_indicators(self, dataframe: DataFrame, metadata: dict) -> DataFrame:
# the model will return all labels created by user in `populate_any_indicators`
# (& appended targets), an indication of whether or not the prediction should be accepted,
# the target mean/std values for each of the labels created by user in
# `populate_any_indicators()` for each training period.
dataframe = self.freqai.start(dataframe, metadata, self)
return dataframe
def populate_any_indicators(
self, pair, df, tf, informative=None, set_generalized_indicators=False
):
"""
Function designed to automatically generate, name and merge features
from user indicated timeframes in the configuration file. User controls the indicators
passed to the training/prediction by prepending indicators with `'%-' + coin `
(see convention below). I.e. user should not prepend any supporting metrics
(e.g. bb_lowerband below) with % unless they explicitly want to pass that metric to the
model.
:param pair: pair to be used as informative
:param df: strategy dataframe which will receive merges from informatives
:param tf: timeframe of the dataframe which will modify the feature names
:param informative: the dataframe associated with the informative pair
:param coin: the name of the coin which will modify the feature names.
"""
coin = pair.split('/')[0]
if informative is None:
informative = self.dp.get_pair_dataframe(pair, tf)
# first loop is automatically duplicating indicators for time periods
for t in self.freqai_info["feature_parameters"]["indicator_periods_candles"]:
t = int(t)
informative[f"%-{coin}rsi-period_{t}"] = ta.RSI(informative, timeperiod=t)
informative[f"%-{coin}mfi-period_{t}"] = ta.MFI(informative, timeperiod=t)
informative[f"%-{coin}adx-period_{t}"] = ta.ADX(informative, window=t)
indicators = [col for col in informative if col.startswith("%")]
# This loop duplicates and shifts all indicators to add a sense of recency to data
for n in range(self.freqai_info["feature_parameters"]["include_shifted_candles"] + 1):
if n == 0:
continue
informative_shift = informative[indicators].shift(n)
informative_shift = informative_shift.add_suffix("_shift-" + str(n))
informative = pd.concat((informative, informative_shift), axis=1)
df = merge_informative_pair(df, informative, self.config["timeframe"], tf, ffill=True)
skip_columns = [
(s + "_" + tf) for s in ["date", "open", "high", "low", "close", "volume"]
]
df = df.drop(columns=skip_columns)
# Add generalized indicators here (because in live, it will call this
# function to populate indicators during training). Notice how we ensure not to
# add them multiple times
if set_generalized_indicators:
# user adds targets here by prepending them with &- (see convention below)
# If user wishes to use multiple targets, a multioutput prediction model
# needs to be used such as templates/CatboostPredictionMultiModel.py
df["&-s_close"] = (
df["close"]
.shift(-self.freqai_info["feature_parameters"]["label_period_candles"])
.rolling(self.freqai_info["feature_parameters"]["label_period_candles"])
.mean()
/ df["close"]
- 1
)
return df
```
Notice how the `populate_any_indicators()` is where the user adds their own features ([more information](#feature-engineering)) and labels ([more information](#setting-classifier-targets)). See a full example at `templates/FreqaiExampleStrategy.py`.
## Creating a dynamic target
The `&*_std/mean` return values describe the statistical fit of the user defined label *during the most recent training*. This value allows the user to know the rarity of a given prediction. For example, `templates/FreqaiExampleStrategy.py`, creates a `target_roi` which is based on filtering out predictions that are below a given z-score of 1.25.
```python
dataframe["target_roi"] = dataframe["&-s_close_mean"] + dataframe["&-s_close_std"] * 1.25
dataframe["sell_roi"] = dataframe["&-s_close_mean"] - dataframe["&-s_close_std"] * 1.25
```
If the user wishes to consider the population
of *historical predictions* for creating the dynamic target instead of the trained labels, (as discussed above) the user
can do so by setting `fit_live_prediction_candles` in the config to the number of historical prediction candles
the user wishes to use to generate target statistics.
```json
"freqai": {
"fit_live_prediction_candles": 300,
}
```
If the user sets this value, FreqAI will initially use the predictions from the training data
and subsequently begin introducing real prediction data as it is generated. FreqAI will save
this historical data to be reloaded if the user stops and restarts a model with the same `identifier`.
## Building a custom prediction model
FreqAI has multiple example prediction model libraries, such as `Catboost` regression (`freqai/prediction_models/CatboostRegressor.py`) and `LightGBM` regression.
However, the user can customize and create their own prediction models using the `IFreqaiModel` class.
The user is encouraged to inherit `train()` and `predict()` to let them customize various aspects of their training procedures.
## Feature engineering
Features are added by the user inside the `populate_any_indicators()` method of the strategy
by prepending indicators with `%`, and labels with `&`.
There are some important components/structures that the user *must* include when building their feature set; the use of these is shown below:
```python
def populate_any_indicators(
self, pair, df, tf, informative=None, set_generalized_indicators=False
):
"""
Function designed to automatically generate, name, and merge features
from user-indicated timeframes in the configuration file. The user controls the indicators
passed to the training/prediction by prepending indicators with `'%-' + coin `
(see convention below). I.e., the user should not prepend any supporting metrics
(e.g., bb_lowerband below) with % unless they explicitly want to pass that metric to the
model.
:param pair: pair to be used as informative
:param df: strategy dataframe which will receive merges from informatives
:param tf: timeframe of the dataframe which will modify the feature names
:param informative: the dataframe associated with the informative pair
:param coin: the name of the coin which will modify the feature names.
"""
coin = pair.split('/')[0]
if informative is None:
informative = self.dp.get_pair_dataframe(pair, tf)
# first loop is automatically duplicating indicators for time periods
for t in self.freqai_info["feature_parameters"]["indicator_periods_candles"]:
t = int(t)
informative[f"%-{coin}rsi-period_{t}"] = ta.RSI(informative, timeperiod=t)
informative[f"%-{coin}mfi-period_{t}"] = ta.MFI(informative, timeperiod=t)
informative[f"%-{coin}adx-period_{t}"] = ta.ADX(informative, window=t)
bollinger = qtpylib.bollinger_bands(
qtpylib.typical_price(informative), window=t, stds=2.2
)
informative[f"{coin}bb_lowerband-period_{t}"] = bollinger["lower"]
informative[f"{coin}bb_middleband-period_{t}"] = bollinger["mid"]
informative[f"{coin}bb_upperband-period_{t}"] = bollinger["upper"]
informative[f"%-{coin}bb_width-period_{t}"] = (
informative[f"{coin}bb_upperband-period_{t}"]
- informative[f"{coin}bb_lowerband-period_{t}"]
) / informative[f"{coin}bb_middleband-period_{t}"]
informative[f"%-{coin}close-bb_lower-period_{t}"] = (
informative["close"] / informative[f"{coin}bb_lowerband-period_{t}"]
)
informative[f"%-{coin}relative_volume-period_{t}"] = (
informative["volume"] / informative["volume"].rolling(t).mean()
)
indicators = [col for col in informative if col.startswith("%")]
# This loop duplicates and shifts all indicators to add a sense of recency to data
for n in range(self.freqai_info["feature_parameters"]["include_shifted_candles"] + 1):
if n == 0:
continue
informative_shift = informative[indicators].shift(n)
informative_shift = informative_shift.add_suffix("_shift-" + str(n))
informative = pd.concat((informative, informative_shift), axis=1)
df = merge_informative_pair(df, informative, self.config["timeframe"], tf, ffill=True)
skip_columns = [
(s + "_" + tf) for s in ["date", "open", "high", "low", "close", "volume"]
]
df = df.drop(columns=skip_columns)
# Add generalized indicators here (because in live, it will call this
# function to populate indicators during training). Notice how we ensure not to
# add them multiple times
if set_generalized_indicators:
df["%-day_of_week"] = (df["date"].dt.dayofweek + 1) / 7
df["%-hour_of_day"] = (df["date"].dt.hour + 1) / 25
# user adds targets here by prepending them with &- (see convention below)
# If user wishes to use multiple targets, a multioutput prediction model
# needs to be used such as templates/CatboostPredictionMultiModel.py
df["&-s_close"] = (
df["close"]
.shift(-self.freqai_info["feature_parameters"]["label_period_candles"])
.rolling(self.freqai_info["feature_parameters"]["label_period_candles"])
.mean()
/ df["close"]
- 1
)
return df
```
In the presented example strategy, the user does not wish to pass the `bb_lowerband` as a feature to the model,
and has therefore not prepended it with `%`. The user does, however, wish to pass `bb_width` to the
model for training/prediction and has therefore prepended it with `%`.
The `include_timeframes` in the example config above are the timeframes (`tf`) of each call to `populate_any_indicators()` in the strategy. In the present case, the user is asking for the
`5m`, `15m`, and `4h` timeframes of the `rsi`, `mfi`, `roc`, and `bb_width` to be included in the feature set.
The user can ask for each of the defined features to be included also from
informative pairs using the `include_corr_pairlist`. This means that the feature
set will include all the features from `populate_any_indicators` on all the `include_timeframes` for each of the correlated pairs defined in the config (`ETH/USD`, `LINK/USD`, and `BNB/USD`).
`include_shifted_candles` indicates the number of previous
candles to include in the feature set. For example, `include_shifted_candles: 2` tells
FreqAI to include the past 2 candles for each of the features in the feature set.
In total, the number of features the user of the presented example strat has created is:
length of `include_timeframes` * no. features in `populate_any_indicators()` * length of `include_corr_pairlist` * no. `include_shifted_candles` * length of `indicator_periods_candles`
$= 3 * 3 * 3 * 2 * 2 = 108$.
Another structure to consider is the location of the labels at the bottom of the example function (below `if set_generalized_indicators:`).
This is where the user will add single features and labels to their feature set to avoid duplication of them from
various configuration parameters that multiply the feature set, such as `include_timeframes`.
!!! Note
Features **must** be defined in `populate_any_indicators()`. Definining FreqAI features in `populate_indicators()`
will cause the algorithm to fail in live/dry mode. If the user wishes to add generalized features that are not associated with
a specific pair or timeframe, they should use the following structure inside `populate_any_indicators()`
(as exemplified in `freqtrade/templates/FreqaiExampleStrategy.py`):
```python
def populate_any_indicators(self, metadata, pair, df, tf, informative=None, coin="", set_generalized_indicators=False):
...
# Add generalized indicators here (because in live, it will call only this function to populate
# indicators for retraining). Notice how we ensure not to add them multiple times by associating
# these generalized indicators to the basepair/timeframe
if set_generalized_indicators:
df['%-day_of_week'] = (df["date"].dt.dayofweek + 1) / 7
df['%-hour_of_day'] = (df['date'].dt.hour + 1) / 25
# user adds targets here by prepending them with &- (see convention below)
# If user wishes to use multiple targets, a multioutput prediction model
# needs to be used such as templates/CatboostPredictionMultiModel.py
df["&-s_close"] = (
df["close"]
.shift(-self.freqai_info["feature_parameters"]["label_period_candles"])
.rolling(self.freqai_info["feature_parameters"]["label_period_candles"])
.mean()
/ df["close"]
- 1
)
```
(Please see the example script located in `freqtrade/templates/FreqaiExampleStrategy.py` for a full example of `populate_any_indicators()`.)
## Setting classifier targets
FreqAI includes the `CatboostClassifier` via the flag `--freqaimodel CatboostClassifier`. The user should take care to set the classes using strings:
```python
df['&s-up_or_down'] = np.where( df["close"].shift(-100) > df["close"], 'up', 'down')
```
Additionally, the example classifier models do not accommodate multiple labels, but they do allow multi-class classification within a single label column.
## Running FreqAI
There are two ways to train and deploy an adaptive machine learning model. FreqAI enables live deployment as well as backtesting analyses. In both cases, a model is trained periodically, as shown in the following figure.
![freqai-window](assets/freqai_moving-window.jpg)
### Running the model live
FreqAI can be run dry/live using the following command:
```bash
freqtrade trade --strategy FreqaiExampleStrategy --config config_freqai.example.json --freqaimodel LightGBMRegressor
```
By default, FreqAI will not find any existing models and will start by training a new one
based on the user's configuration settings. Following training, the model will be used to make predictions on incoming candles until a new model is available. New models are typically generated as often as possible, with FreqAI managing an internal queue of the coin pairs to try to keep all models equally up to date. FreqAI will always use the most recently trained model to make predictions on incoming live data. If the user does not want FreqAI to retrain new models as often as possible, they can set `live_retrain_hours` to tell FreqAI to wait at least that number of hours before training a new model. Additionally, the user can set `expired_hours` to tell FreqAI to avoid making predictions on models that are older than that number of hours.
If the user wishes to start a dry/live run from a saved backtest model (or from a previously crashed dry/live session), the user only needs to reuse
the same `identifier` parameter:
```json
"freqai": {
"identifier": "example",
"live_retrain_hours": 0.5
}
```
In this case, although FreqAI will initiate with a
pre-trained model, it will still check to see how much time has elapsed since the model was trained,
and if a full `live_retrain_hours` has elapsed since the end of the loaded model, FreqAI will retrain.
### Backtesting
The FreqAI backtesting module can be executed with the following command:
```bash
freqtrade backtesting --strategy FreqaiExampleStrategy --config config_freqai.example.json --freqaimodel LightGBMRegressor --timerange 20210501-20210701
```
Backtesting mode requires the user to have the data pre-downloaded (unlike in dry/live mode where FreqAI automatically downloads the necessary data). The user should be careful to consider that the time range of the downloaded data is more than the backtesting time range. This is because FreqAI needs data prior to the desired backtesting time range in order to train a model to be ready to make predictions on the first candle of the user-set backtesting time range. More details on how to calculate the data to download can be found [here](#deciding-the-sliding-training-window-and-backtesting-duration).
If this command has never been executed with the existing config file, it will train a new model
for each pair, for each backtesting window within the expanded `--timerange`.
!!! Note "Model reuse"
Once the training is completed, the user can execute the backtesting again with the same config file and
FreqAI will find the trained models and load them instead of spending time training. This is useful
if the user wants to tweak (or even hyperopt) buy and sell criteria inside the strategy. If the user
*wants* to retrain a new model with the same config file, then they should simply change the `identifier`.
This way, the user can return to using any model they wish by simply specifying the `identifier`.
---
### Deciding the size of the sliding training window and backtesting duration
The user defines the backtesting timerange with the typical `--timerange` parameter in the
configuration file. The duration of the sliding training window is set by `train_period_days`, whilst
`backtest_period_days` is the sliding backtesting window, both in number of days (`backtest_period_days` can be
a float to indicate sub-daily retraining in live/dry mode). In the presented example config,
the user is asking FreqAI to use a training period of 30 days and backtest on the subsequent 7 days.
This means that if the user sets `--timerange 20210501-20210701`,
FreqAI will train have trained 8 separate models at the end of `--timerange` (because the full range comprises 8 weeks). After the training of the model, FreqAI will backtest the subsequent 7 days. The "sliding window" then moves one week forward (emulating FreqAI retraining once per week in live mode) and the new model uses the previous 30 days (including the 7 days used for backtesting by the previous model) to train. This is repeated until the end of `--timerange`.
In live mode, the required training data is automatically computed and downloaded. However, in backtesting mode,
the user must manually enter the required number of `startup_candles` in the config. This value
is used to increase the data to FreqAI, which should be sufficient to enable all indicators
to be NaN free at the beginning of the first training. This is done by identifying the
longest timeframe (`4h` in presented example config) and the longest indicator period (`20` days in presented example config)
and adding this to the `train_period_days`. The units need to be in the base candle time frame:
`startup_candles` = ( 4 hours * 20 max period * 60 minutes/hour + 30 day train_period_days * 1440 minutes per day ) / 5 min (base time frame) = 9360.
!!! Note
In dry/live mode, this is all precomputed and handled automatically. Thus, `startup_candle` has no influence on dry/live mode.
!!! Note
Although fractional `backtest_period_days` is allowed, the user should be aware that the `--timerange` is divided by this value to determine the number of models that FreqAI will need to train in order to backtest the full range. For example, if the user wants to set a `--timerange` of 10 days, and asks for a `backtest_period_days` of 0.1, FreqAI will need to train 100 models per pair to complete the full backtest. Because of this, a true backtest of FreqAI adaptive training would take a *very* long time. The best way to fully test a model is to run it dry and let it constantly train. In this case, backtesting would take the exact same amount of time as a dry run.
### Defining model expirations
During dry/live mode, FreqAI trains each coin pair sequentially (on separate threads/GPU from the main Freqtrade bot). This means that there is always an age discrepancy between models. If a user is training on 50 pairs, and each pair requires 5 minutes to train, the oldest model will be over 4 hours old. This may be undesirable if the characteristic time scale (the trade duration target) for a strategy is less than 4 hours. The user can decide to only make trade entries if the model is less than
a certain number of hours old by setting the `expiration_hours` in the config file:
```json
"freqai": {
"expiration_hours": 0.5,
}
```
In the presented example config, the user will only allow predictions on models that are less than 1/2 hours old.
### Purging old model data
FreqAI stores new model files each time it retrains. These files become obsolete as new models are trained and FreqAI adapts to new market conditions. Users planning to leave FreqAI running for extended periods of time with high frequency retraining should enable `purge_old_models` in their config:
```json
"freqai": {
"purge_old_models": true,
}
```
This will automatically purge all models older than the two most recently trained ones.
### Returning additional info from training
The user may find that there are some important metrics that they'd like to return to the strategy at the end of each model training.
The user can include these metrics by assigning them to `dk.data['extra_returns_per_train']['my_new_value'] = XYZ` inside their custom prediction model class. FreqAI takes the `my_new_value` assigned in this dictionary and expands it to fit the return dataframe to the strategy.
The user can then use the value in the strategy with `dataframe['my_new_value']`. An example of how this is already used in FreqAI is
the `&*_mean` and `&*_std` values, which indicate the mean and standard deviation of the particular target (label) during the most recent training.
An example, where the user wants to use live metrics from the trade database, is shown below:
```json
"freqai": {
"extra_returns_per_train": {"total_profit": 4}
}
```
The user needs to set the standard dictionary in the config so that FreqAI can return proper dataframe shapes. These values will likely be overridden by the prediction model, but in the case where the model has yet to set them, or needs a default initial value, this is the value that will be returned.
### Setting up a follower
The user can define:
```json
"freqai": {
"follow_mode": true,
"identifier": "example"
}
```
to indicate to the bot that it should not train models, but instead should look for models trained by a leader with the same `identifier`. In this example, the user has a leader bot with the `identifier: "example"`. The leader bot is already running or launching simultaneously as the follower.
The follower will load models created by the leader and inference them to obtain predictions.
## Data manipulation techniques
### Feature normalization
The feature set created by the user is automatically normalized to the training data. This includes all test data and unseen prediction data (dry/live/backtest).
### Reducing data dimensionality with Principal Component Analysis
Users can reduce the dimensionality of their features by activating the `principal_component_analysis` in the config:
```json
"freqai": {
"feature_parameters" : {
"principal_component_analysis": true
}
}
```
This will perform PCA on the features and reduce the dimensionality of the data so that the explained variance of the data set is >= 0.999.
### Stratifying the data for training and testing the model
The user can stratify (group) the training/testing data using:
```json
"freqai": {
"feature_parameters" : {
"stratify_training_data": 3
}
}
```
This will split the data chronologically so that every Xth data point is used to test the model after training. In the
example above, the user is asking for every third data point in the dataframe to be used for
testing; the other points are used for training.
The test data is used to evaluate the performance of the model after training. If the test score is high, the model is able to capture the behavior of the data well. If the test score is low, either the model either does not capture the complexity of the data, the test data is significantly different from the train data, or a different model should be used.
### Controlling the model learning process
Model training parameters are unique to the machine learning library selected by the user. FreqAI allows the user to set any parameter for any library using the `model_training_parameters` dictionary in the user configuration file. The example configuration file (found in `config_examples/config_freqai.example.json`) show some of the example parameters associated with `Catboost` and `LightGBM`, but the user can add any parameters available in those libraries.
Data split parameters are defined in `data_split_parameters` which can be any parameters associated with `Sklearn`'s `train_test_split()` function.
FreqAI includes some additional parameters such as `weight_factor`, which allows the user to weight more recent data more strongly
than past data via an exponential function:
$$ W_i = \exp(\frac{-i}{\alpha*n}) $$
where $W_i$ is the weight of data point $i$ in a total set of $n$ data points. Below is a figure showing the effect of different weight factors on the data points (candles) in a feature set.
![weight-factor](assets/freqai_weight-factor.jpg)
`train_test_split()` has a parameters called `shuffle` that allows the user to keep the data unshuffled. This is particularly useful to avoid biasing training with temporally auto-correlated data.
Finally, `label_period_candles` defines the offset (number of candles into the future) used for the `labels`. In the presented example config,
the user is asking for `labels` that are 24 candles in the future.
### Outlier removal
#### Removing outliers with the Dissimilarity Index
The user can tell FreqAI to remove outlier data points from the training/test data sets using a Dissimilarity Index by including the following statement in the config:
```json
"freqai": {
"feature_parameters" : {
"DI_threshold": 1
}
}
```
Equity and crypto markets suffer from a high level of non-patterned noise in the form of outlier data points. The Dissimilarity Index (DI) aims to quantify the uncertainty associated with each prediction made by the model. The DI allows predictions which are outliers (not existent in the model feature space) to be thrown out due to low levels of certainty.
To do so, FreqAI measures the distance between each training data point (feature vector), $X_{a}$, and all other training data points:
$$ d_{ab} = \sqrt{\sum_{j=1}^p(X_{a,j}-X_{b,j})^2} $$
where $d_{ab}$ is the distance between the normalized points $a$ and $b$. $p$ is the number of features, i.e., the length of the vector $X$. The characteristic distance, $\overline{d}$ for a set of training data points is simply the mean of the average distances:
$$ \overline{d} = \sum_{a=1}^n(\sum_{b=1}^n(d_{ab}/n)/n) $$
$\overline{d}$ quantifies the spread of the training data, which is compared to the distance between a new prediction feature vectors, $X_k$ and all the training data:
$$ d_k = \arg \min d_{k,i} $$
which enables the estimation of the Dissimilarity Index as:
$$ DI_k = d_k/\overline{d} $$
The user can tweak the DI through the `DI_threshold` to increase or decrease the extrapolation of the trained model.
Below is a figure that describes the DI for a 3D data set.
![DI](assets/freqai_DI.jpg)
#### Removing outliers using a Support Vector Machine (SVM)
The user can tell FreqAI to remove outlier data points from the training/test data sets using a SVM by setting:
```json
"freqai": {
"feature_parameters" : {
"use_SVM_to_remove_outliers": true
}
}
```
FreqAI will train an SVM on the training data (or components of it if the user activated
`principal_component_analysis`) and remove any data point that the SVM deems to be beyond the feature space.
The parameter `shuffle` is by default set to `False` to ensure consistent results. If it is set to `True`, running the SVM multiple times on the same data set might result in different outcomes due to `max_iter` being to low for the algorithm to reach the demanded `tol`. Increasing `max_iter` solves this issue but causes the procedure to take longer time.
The parameter `nu`, *very* broadly, is the amount of data points that should be considered outliers.
#### Removing outliers with DBSCAN
The user can configure FreqAI to use DBSCAN to cluster and remove outliers from the training/test data set or incoming outliers from predictions, by activating `use_DBSCAN_to_remove_outliers` in the config:
```json
"freqai": {
"feature_parameters" : {
"use_DBSCAN_to_remove_outliers": true
}
}
```
DBSCAN is an unsupervised machine learning algorithm that clusters data without needing to know how many clusters there should be.
Given a number of data points $N$, and a distance $\varepsilon$, DBSCAN clusters the data set by setting all data points that have $N-1$ other data points within a distance of $\varepsilon$ as *core points*. A data point that is within a distance of $\varepsilon$ from a *core point* but that does not have $N-1$ other data points within a distance of $\varepsilon$ from itself is considered an *edge point*. A cluster is then the collection of *core points* and *edge points*. Data points that have no other data points at a distance $<\varepsilon$ are considered outliers. The figure below shows a cluster with $N = 3$.
![dbscan](assets/freqai_dbscan.jpg)
FreqAI uses `sklearn.cluster.DBSCAN` (details are available on scikit-learn's webpage [here](#https://scikit-learn.org/stable/modules/generated/sklearn.cluster.DBSCAN.html)) with `min_samples` ($N$) taken as double the no. of user-defined features, and `eps` ($\varepsilon$) taken as the longest distance in the *k-distance graph* computed from the nearest neighbors in the pairwise distances of all data points in the feature set.
## Additional information
### Common pitfalls
FreqAI cannot be combined with dynamic `VolumePairlists` (or any pairlist filter that adds and removes pairs dynamically).
This is for performance reasons - FreqAI relies on making quick predictions/retrains. To do this effectively,
it needs to download all the training data at the beginning of a dry/live instance. FreqAI stores and appends
new candles automatically for future retrains. This means that if new pairs arrive later in the dry run due to a volume pairlist, it will not have the data ready. However, FreqAI does work with the `ShufflePairlist` or a `VolumePairlist` which keeps the total pairlist constant (but reorders the pairs according to volume).
## Credits
FreqAI was developed by a group of individuals who all contributed specific skillsets to the project.
Conception and software development:
Robert Caulk @robcaulk
Theoretical brainstorming, data analysis:
Elin Törnquist @th0rntwig
Code review, software architecture brainstorming:
@xmatthias
Beta testing and bug reporting:
@bloodhunter4rc, Salah Lamkadem @ikonx, @ken11o2, @longyu, @paranoidandy, @smidelis, @smarm
Juha Nykänen @suikula, Wagner Costa @wagnercosta

View File

@ -40,7 +40,8 @@ pip install -r requirements-hyperopt.txt
```
usage: freqtrade hyperopt [-h] [-v] [--logfile FILE] [-V] [-c PATH] [-d PATH]
[--userdir PATH] [-s NAME] [--strategy-path PATH]
[--recursive-strategy-search] [-i TIMEFRAME]
[--recursive-strategy-search] [--freqaimodel NAME]
[--freqaimodel-path PATH] [-i TIMEFRAME]
[--timerange TIMERANGE]
[--data-format-ohlcv {json,jsongz,hdf5}]
[--max-open-trades INT]
@ -53,7 +54,7 @@ usage: freqtrade hyperopt [-h] [-v] [--logfile FILE] [-V] [-c PATH] [-d PATH]
[--print-all] [--no-color] [--print-json] [-j JOBS]
[--random-state INT] [--min-trades INT]
[--hyperopt-loss NAME] [--disable-param-export]
[--ignore-missing-spaces]
[--ignore-missing-spaces] [--analyze-per-epoch]
optional arguments:
-h, --help show this help message and exit
@ -129,6 +130,7 @@ optional arguments:
--ignore-missing-spaces, --ignore-unparameterized-spaces
Suppress errors for any requested Hyperopt spaces that
do not contain any parameters.
--analyze-per-epoch Run populate_indicators once per epoch.
Common arguments:
-v, --verbose Verbose mode (-vv for more, -vvv to get all messages).
@ -154,6 +156,10 @@ Strategy arguments:
--recursive-strategy-search
Recursively search for a strategy in the strategies
folder.
--freqaimodel NAME Specify a custom freqaimodels.
--freqaimodel-path PATH
Specify additional lookup path for freqaimodels.
```
### Hyperopt checklist
@ -185,7 +191,7 @@ Rarely you may also need to create a [nested class](advanced-hyperopt.md#overrid
### Hyperopt execution logic
Hyperopt will first load your data into memory and will then run `populate_indicators()` once per Pair to generate all indicators.
Hyperopt will first load your data into memory and will then run `populate_indicators()` once per Pair to generate all indicators, unless `--analyze-per-epoch` is specified.
Hyperopt will then spawn into different processes (number of processors, or `-j <n>`), and run backtesting over and over again, changing the parameters that are part of the `--spaces` defined.
@ -426,9 +432,10 @@ While this strategy is most likely too simple to provide consistent profit, it s
`range` property may also be used with `DecimalParameter` and `CategoricalParameter`. `RealParameter` does not provide this property due to infinite search space.
??? Hint "Performance tip"
By doing the calculation of all possible indicators in `populate_indicators()`, the calculation of the indicator happens only once for every parameter.
While this may slow down the hyperopt startup speed, the overall performance will increase as the Hyperopt execution itself may pick the same value for multiple epochs (changing other values).
You should however try to use space ranges as small as possible. Every new column will require more memory, and every possibility hyperopt can try will increase the search space.
During normal hyperopting, indicators are calculated once and supplied to each epoch, linearly increasing RAM usage as a factor of increasing cores. As this also has performance implications, hyperopt provides `--analyze-per-epoch` which will move the execution of `populate_indicators()` to the epoch process, calculating a single value per parameter per epoch instead of using the `.range` functionality. In this case, `.range` functionality will only return the actually used value. This will reduce RAM usage, but increase CPU usage. However, your hyperopting run will be less likely to fail due to Out Of Memory (OOM) issues.
In either case, you should try to use space ranges as small as possible this will improve CPU/RAM usage in both scenarios.
## Optimizing protections
@ -879,6 +886,7 @@ To combat these, you have multiple options:
* Avoid using `--timeframe-detail` (this loads a lot of additional data into memory).
* Reduce the number of parallel processes (`-j <n>`).
* Increase the memory of your machine.
* Use `--analyze-per-epoch` if you're using a lot of parameters with `.range` functionality.
## The objective has been evaluated at this point before.

View File

@ -326,6 +326,16 @@ python3 -m pip install --upgrade pip
python3 -m pip install -e .
```
Patch conda libta-lib (Linux only)
```bash
# Ensure that the environment is active!
conda activate freqtrade-conda
cd build_helpers
bash install_ta-lib.sh ${CONDA_PREFIX} nosudo
```
### Congratulations
[You are ready](#you-are-ready), and run the bot

View File

@ -1,6 +1,6 @@
markdown==3.3.7
mkdocs==1.3.1
mkdocs-material==8.3.9
mkdocs-material==8.4.1
mdx_truly_sane_lists==1.3
pymdown-extensions==9.5
jinja2==3.1.2

View File

@ -163,6 +163,8 @@ python3 scripts/rest_client.py --config rest_config.json <command> [optional par
| `strategy <strategy>` | Get specific Strategy content. **Alpha**
| `available_pairs` | List available backtest data. **Alpha**
| `version` | Show version.
| `sysinfo` | Show informations about the system load.
| `health` | Show bot health (last bot loop).
!!! Warning "Alpha status"
Endpoints labeled with *Alpha status* above may change at any time without notice.
@ -227,6 +229,11 @@ forceexit
Force-exit a trade.
:param tradeid: Id of the trade (can be received via status command)
:param ordertype: Order type to use (must be market or limit)
:param amount: Amount to sell. Full sell if not given
health
Provides a quick health check of the running bot.
locks
Return current locks
@ -312,12 +319,13 @@ version
whitelist
Show the current whitelist.
```
### OpenAPI interface
To enable the builtin openAPI interface (Swagger UI), specify `"enable_openapi": true` in the api_server configuration.
This will enable the Swagger UI at the `/docs` endpoint. By default, that's running at http://localhost:8080/docs/ - but it'll depend on your settings.
This will enable the Swagger UI at the `/docs` endpoint. By default, that's running at http://localhost:8080/docs - but it'll depend on your settings.
### Advanced API usage using JWT tokens

View File

@ -75,7 +75,7 @@ class AwesomeStrategy(IStrategy):
```
### Stake size management
## Stake size management
Called before entering a trade, makes it possible to manage your position size when placing a new trade.
@ -423,7 +423,7 @@ class AwesomeStrategy(IStrategy):
!!! Warning "Backtesting"
Custom prices are supported in backtesting (starting with 2021.12), and orders will fill if the price falls within the candle's low/high range.
Orders that don't fill immediately are subject to regular timeout handling, which happens once per (detail) candle.
`custom_exit_price()` is only called for sells of type exit_signal and Custom exit. All other exit-types will use regular backtesting prices.
`custom_exit_price()` is only called for sells of type exit_signal, Custom exit and partial exits. All other exit-types will use regular backtesting prices.
## Custom order timeout rules
@ -623,12 +623,13 @@ class AwesomeStrategy(IStrategy):
!!! Warning
`confirm_trade_exit()` can prevent stoploss exits, causing significant losses as this would ignore stoploss exits.
`confirm_trade_exit()` will not be called for Liquidations - as liquidations are forced by the exchange, and therefore cannot be rejected.
## Adjust trade position
The `position_adjustment_enable` strategy property enables the usage of `adjust_trade_position()` callback in the strategy.
For performance reasons, it's disabled by default and freqtrade will show a warning message on startup if enabled.
`adjust_trade_position()` can be used to perform additional orders, for example to manage risk with DCA (Dollar Cost Averaging).
`adjust_trade_position()` can be used to perform additional orders, for example to manage risk with DCA (Dollar Cost Averaging) or to increase or decrease positions.
`max_entry_position_adjustment` property is used to limit the number of additional buys per trade (on top of the first buy) that the bot can execute. By default, the value is -1 which means the bot have no limit on number of adjustment buys.
@ -636,10 +637,13 @@ The strategy is expected to return a stake_amount (in stake currency) between `m
If there are not enough funds in the wallet (the return value is above `max_stake`) then the signal will be ignored.
Additional orders also result in additional fees and those orders don't count towards `max_open_trades`.
This callback is **not** called when there is an open order (either buy or sell) waiting for execution, or when you have reached the maximum amount of extra buys that you have set on `max_entry_position_adjustment`.
This callback is **not** called when there is an open order (either buy or sell) waiting for execution.
`adjust_trade_position()` is called very frequently for the duration of a trade, so you must keep your implementation as performant as possible.
Position adjustments will always be applied in the direction of the trade, so a positive value will always increase your position, no matter if it's a long or short trade. Modifications to leverage are not possible.
Additional Buys are ignored once you have reached the maximum amount of extra buys that you have set on `max_entry_position_adjustment`, but the callback is called anyway looking for partial exits.
Position adjustments will always be applied in the direction of the trade, so a positive value will always increase your position (negative values will decrease your position), no matter if it's a long or short trade. Modifications to leverage are not possible.
!!! Note "About stake size"
Using fixed stake size means it will be the amount used for the first order, just like without position adjustment.
@ -648,12 +652,12 @@ Position adjustments will always be applied in the direction of the trade, so a
!!! Warning
Stoploss is still calculated from the initial opening price, not averaged price.
Regular stoploss rules still apply (cannot move down).
!!! Warning "/stopbuy"
While `/stopbuy` command stops the bot from entering new trades, the position adjustment feature will continue buying new orders on existing trades.
While `/stopentry` command stops the bot from entering new trades, the position adjustment feature will continue buying new orders on existing trades.
!!! Warning "Backtesting"
During backtesting this callback is called for each candle in `timeframe` or `timeframe_detail`, so performance will be affected.
During backtesting this callback is called for each candle in `timeframe` or `timeframe_detail`, so run-time performance will be affected.
``` python
from freqtrade.persistence import Trade
@ -684,22 +688,41 @@ def custom_stake_amount(self, pair: str, current_time: datetime, current_rate: f
return proposed_stake / self.max_dca_multiplier
def adjust_trade_position(self, trade: Trade, current_time: datetime,
current_rate: float, current_profit: float, min_stake: Optional[float],
max_stake: float, **kwargs):
current_rate: float, current_profit: float,
min_stake: Optional[float], max_stake: float,
current_entry_rate: float, current_exit_rate: float,
current_entry_profit: float, current_exit_profit: float,
**kwargs) -> Optional[float]:
"""
Custom trade adjustment logic, returning the stake amount that a trade should be increased.
This means extra buy orders with additional fees.
Custom trade adjustment logic, returning the stake amount that a trade should be
increased or decreased.
This means extra buy or sell orders with additional fees.
Only called when `position_adjustment_enable` is set to True.
For full documentation please go to https://www.freqtrade.io/en/latest/strategy-advanced/
When not implemented by a strategy, returns None
:param trade: trade object.
:param current_time: datetime object, containing the current datetime
:param current_rate: Current buy rate.
:param current_profit: Current profit (as ratio), calculated based on current_rate.
:param min_stake: Minimal stake size allowed by exchange.
:param max_stake: Balance available for trading.
:param min_stake: Minimal stake size allowed by exchange (for both entries and exits)
:param max_stake: Maximum stake allowed (either through balance, or by exchange limits).
:param current_entry_rate: Current rate using entry pricing.
:param current_exit_rate: Current rate using exit pricing.
:param current_entry_profit: Current profit using entry pricing.
:param current_exit_profit: Current profit using exit pricing.
:param **kwargs: Ensure to keep this here so updates to this won't break your strategy.
:return float: Stake amount to adjust your trade
:return float: Stake amount to adjust your trade,
Positive values to increase position, Negative values to decrease position.
Return None for no action.
"""
if current_profit > 0.05 and trade.nr_of_successful_exits == 0:
# Take half of the profit at +5%
return -(trade.stake_amount / 2)
if current_profit > -0.05:
return None
@ -734,6 +757,25 @@ def custom_stake_amount(self, pair: str, current_time: datetime, current_rate: f
```
### Position adjust calculations
* Entry rates are calculated using weighted averages.
* Exits will not influence the average entry rate.
* Partial exit relative profit is relative to the average entry price at this point.
* Final exit relative profit is calculated based on the total invested capital. (See example below)
??? example "Calculation example"
*This example assumes 0 fees for simplicity, and a long position on an imaginary coin.*
* Buy 100@8\$
* Buy 100@9\$ -> Avg price: 8.5\$
* Sell 100@10\$ -> Avg price: 8.5\$, realized profit 150\$, 17.65%
* Buy 150@11\$ -> Avg price: 10\$, realized profit 150\$, 17.65%
* Sell 100@12\$ -> Avg price: 10\$, total realized profit 350\$, 20%
* Sell 150@14\$ -> Avg price: 10\$, total realized profit 950\$, 40%
The total profit for this trade was 950$ on a 3350$ investment (`100@8$ + 100@9$ + 150@11$`). As such - the final relative profit is 28.35% (`950 / 3350`).
## Adjust Entry Price
The `adjust_entry_price()` callback may be used by strategy developer to refresh/replace limit orders upon arrival of new candles.

View File

@ -617,7 +617,6 @@ Please always check the mode of operation to select the correct method to get da
### *available_pairs*
``` python
if self.dp:
for pair, timeframe in self.dp.available_pairs:
print(f"available {pair}, {timeframe}")
```
@ -630,7 +629,7 @@ The strategy might look something like this:
*Scan through the top 10 pairs by volume using the `VolumePairList` every 5 minutes and use a 14 day RSI to buy and sell.*
Due to the limited available data, it's very difficult to resample `5m` candles into daily candles for use in a 14 day RSI. Most exchanges limit us to just 500 candles which effectively gives us around 1.74 daily candles. We need 14 days at least!
Due to the limited available data, it's very difficult to resample `5m` candles into daily candles for use in a 14 day RSI. Most exchanges limit us to just 500-1000 candles which effectively gives us around 1.74 daily candles. We need 14 days at least!
Since we can't resample the data we will have to use an informative pair; and since the whitelist will be dynamic we don't know which pair(s) to use.
@ -653,7 +652,6 @@ This is where calling `self.dp.current_whitelist()` comes in handy.
``` python
# fetch live / historical candle (OHLCV) data for the first informative pair
if self.dp:
inf_pair, inf_timeframe = self.informative_pairs()[0]
informative = self.dp.get_pair_dataframe(pair=inf_pair,
timeframe=inf_timeframe)
@ -671,7 +669,6 @@ It can also be used in specific callbacks to get the signal that caused the acti
``` python
# fetch current dataframe
if self.dp:
if self.dp.runmode.value in ('live', 'dry_run'):
dataframe, last_updated = self.dp.get_analyzed_dataframe(pair=metadata['pair'],
timeframe=self.timeframe)
@ -684,7 +681,6 @@ if self.dp:
### *orderbook(pair, maximum)*
``` python
if self.dp:
if self.dp.runmode.value in ('live', 'dry_run'):
ob = self.dp.orderbook(metadata['pair'], 1)
dataframe['best_bid'] = ob['bids'][0][0]
@ -717,7 +713,6 @@ Therefore, using `ob['bids'][0][0]` as demonstrated above will result in using t
### *ticker(pair)*
``` python
if self.dp:
if self.dp.runmode.value in ('live', 'dry_run'):
ticker = self.dp.ticker(metadata['pair'])
dataframe['last_price'] = ticker['last']
@ -732,7 +727,24 @@ if self.dp:
data returned from the exchange and add appropriate error handling / defaults.
!!! Warning "Warning about backtesting"
This method will always return up-to-date values - so usage during backtesting / hyperopt will lead to wrong results.
This method will always return up-to-date values - so usage during backtesting / hyperopt without runmode checks will lead to wrong results.
### Send Notification
The dataprovider `.send_msg()` function allows you to send custom notifications from your strategy.
Identical notifications will only be sent once per candle, unless the 2nd argument (`always_send`) is set to True.
``` python
self.dp.send_msg(f"{metadata['pair']} just got hot!")
# Force send this notification, avoid caching (Please read warning below!)
self.dp.send_msg(f"{metadata['pair']} just got hot!", always_send=True)
```
Notifications will only be sent in trading modes (Live/Dry-run) - so this method can be called without conditions for backtesting.
!!! Warning "Spamming"
You can spam yourself pretty good by setting `always_send=True` in this method. Use this with great care and only in conditions you know will not happen throughout a candle to avoid a message every 5 seconds.
### Complete Data-provider sample

View File

@ -14,7 +14,7 @@ from freqtrade.configuration import Configuration
# Initialize empty configuration object
config = Configuration.from_files([])
# Optionally, use existing configuration file
# Optionally (recommended), use existing configuration file
# config = Configuration.from_files(["config.json"])
# Define some constants
@ -22,7 +22,7 @@ config["timeframe"] = "5m"
# Name of the strategy class
config["strategy"] = "SampleStrategy"
# Location of the data
data_location = Path(config['user_data_dir'], 'data', 'binance')
data_location = config['datadir']
# Pair to analyze - Only use one pair here
pair = "BTC/USDT"
```

View File

@ -98,6 +98,7 @@ Example configuration showing the different settings:
"exit_fill": "off",
"protection_trigger": "off",
"protection_trigger_global": "on",
"strategy_msg": "off",
"show_candle": "off"
},
"reload": true,
@ -109,7 +110,8 @@ Example configuration showing the different settings:
`exit` notifications are sent when the order is placed, while `exit_fill` notifications are sent when the order is filled on the exchange.
`*_fill` notifications are off by default and must be explicitly enabled.
`protection_trigger` notifications are sent when a protection triggers and `protection_trigger_global` notifications trigger when global protections are triggered.
`show_candle` - show candle values as part of entry/exit messages. Only possible value is "ohlc".
`strategy_msg` - Receive notifications from the strategy, sent via `self.dp.send_msg()` from the strategy [more details](strategy-customization.md#send-notification).
`show_candle` - show candle values as part of entry/exit messages. Only possible values are `"ohlc"` or `"off"`.
`balance_dust_level` will define what the `/balance` command takes as "dust" - Currencies with a balance below this will be shown.
`reload` allows you to disable reload-buttons on selected messages.
@ -147,7 +149,7 @@ You can create your own keyboard in `config.json`:
!!! Note "Supported Commands"
Only the following commands are allowed. Command arguments are not supported!
`/start`, `/stop`, `/status`, `/status table`, `/trades`, `/profit`, `/performance`, `/daily`, `/stats`, `/count`, `/locks`, `/balance`, `/stopbuy`, `/reload_config`, `/show_config`, `/logs`, `/whitelist`, `/blacklist`, `/edge`, `/help`, `/version`
`/start`, `/stop`, `/status`, `/status table`, `/trades`, `/profit`, `/performance`, `/daily`, `/stats`, `/count`, `/locks`, `/balance`, `/stopentry`, `/reload_config`, `/show_config`, `/logs`, `/whitelist`, `/blacklist`, `/edge`, `/help`, `/version`
## Telegram commands
@ -159,7 +161,7 @@ official commands. You can ask at any moment for help with `/help`.
|----------|-------------|
| `/start` | Starts the trader
| `/stop` | Stops the trader
| `/stopbuy` | Stops the trader from opening new trades. Gracefully closes open trades according to their rules.
| `/stopbuy | /stopentry` | Stops the trader from opening new trades. Gracefully closes open trades according to their rules.
| `/reload_config` | Reloads the configuration file
| `/show_config` | Shows part of the current configuration with relevant settings to operation
| `/logs [limit]` | Show last log messages.
@ -185,7 +187,7 @@ official commands. You can ask at any moment for help with `/help`.
| `/stats` | Shows Wins / losses by Exit reason as well as Avg. holding durations for buys and sells
| `/exits` | Shows Wins / losses by Exit reason as well as Avg. holding durations for buys and sells
| `/entries` | Shows Wins / losses by Exit reason as well as Avg. holding durations for buys and sells
| `/whitelist` | Show the current whitelist
| `/whitelist [sorted] [baseonly]` | Show the current whitelist. Optionally display in alphabetical order and/or with just the base currency of each pairing.
| `/blacklist [pair]` | Show the current blacklist, or adds a pair to the blacklist.
| `/edge` | Show validated pairs by Edge if it is enabled.
| `/help` | Show help message

View File

@ -611,6 +611,26 @@ Common arguments:
```
### Webserver mode - docker
You can also use webserver mode via docker.
Starting a one-off container requires the configuration of the port explicitly, as ports are not exposed by default.
You can use `docker-compose run --rm -p 127.0.0.1:8080:8080 freqtrade webserver` to start a one-off container that'll be removed once you stop it. This assumes that port 8080 is still available and no other bot is running on that port.
Alternatively, you can reconfigure the docker-compose file to have the command updated:
``` yml
command: >
webserver
--config /freqtrade/user_data/config.json
```
You can now use `docker-compose up` to start the webserver.
This assumes that the configuration has a webserver enabled and configured for docker (listening port = `0.0.0.0`).
!!! Tip
Don't forget to reset the command back to the trade command if you want to start a live or dry-run bot.
## Show previous Backtest results
Allows you to show previous backtest results.

View File

@ -9,6 +9,7 @@ dependencies:
- pandas
- pip
- py-find-1st
- aiohttp
- SQLAlchemy
- python-telegram-bot
@ -64,7 +65,7 @@ dependencies:
- pip:
- pycoingecko
- py_find_1st
# - py_find_1st
- tables
- pytest-random-order
- ccxt

View File

@ -1,5 +1,5 @@
""" Freqtrade bot """
__version__ = '2022.7'
__version__ = '2022.8'
if 'dev' in __version__:
try:

View File

@ -12,7 +12,8 @@ from freqtrade.constants import DEFAULT_CONFIG
ARGS_COMMON = ["verbosity", "logfile", "version", "config", "datadir", "user_data_dir"]
ARGS_STRATEGY = ["strategy", "strategy_path", "recursive_strategy_search"]
ARGS_STRATEGY = ["strategy", "strategy_path", "recursive_strategy_search", "freqaimodel",
"freqaimodel_path"]
ARGS_TRADE = ["db_url", "sd_notify", "dry_run", "dry_run_wallet", "fee"]
@ -33,7 +34,7 @@ ARGS_HYPEROPT = ARGS_COMMON_OPTIMIZE + ["hyperopt", "hyperopt_path",
"print_colorized", "print_json", "hyperopt_jobs",
"hyperopt_random_state", "hyperopt_min_trades",
"hyperopt_loss", "disableparamexport",
"hyperopt_ignore_missing_space"]
"hyperopt_ignore_missing_space", "analyze_per_epoch"]
ARGS_EDGE = ARGS_COMMON_OPTIMIZE + ["stoploss_range"]
@ -68,7 +69,7 @@ ARGS_CONVERT_DATA_OHLCV = ARGS_CONVERT_DATA + ["timeframes", "exchange", "tradin
ARGS_CONVERT_TRADES = ["pairs", "timeframes", "exchange", "dataformat_ohlcv", "dataformat_trades"]
ARGS_LIST_DATA = ["exchange", "dataformat_ohlcv", "pairs", "trading_mode"]
ARGS_LIST_DATA = ["exchange", "dataformat_ohlcv", "pairs", "trading_mode", "show_timerange"]
ARGS_DOWNLOAD_DATA = ["pairs", "pairs_file", "days", "new_pairs_days", "include_inactive",
"timerange", "download_trades", "exchange", "timeframes",

View File

@ -67,7 +67,7 @@ def ask_user_config() -> Dict[str, Any]:
"type": "text",
"name": "stake_amount",
"message": f"Please insert your stake amount (Number or '{UNLIMITED_STAKE_AMOUNT}'):",
"default": "100",
"default": "unlimited",
"validate": lambda val: val == UNLIMITED_STAKE_AMOUNT or validate_is_float(val),
"filter": lambda val: '"' + UNLIMITED_STAKE_AMOUNT + '"'
if val == UNLIMITED_STAKE_AMOUNT
@ -164,7 +164,7 @@ def ask_user_config() -> Dict[str, Any]:
"when": lambda x: x['telegram']
},
{
"type": "text",
"type": "password",
"name": "telegram_chat_id",
"message": "Insert Telegram chat id",
"when": lambda x: x['telegram']
@ -191,7 +191,7 @@ def ask_user_config() -> Dict[str, Any]:
"when": lambda x: x['api_server']
},
{
"type": "text",
"type": "password",
"name": "api_server_password",
"message": "Insert api-server password",
"when": lambda x: x['api_server']

View File

@ -255,6 +255,13 @@ AVAILABLE_CLI_OPTIONS = {
nargs='+',
default='default',
),
"analyze_per_epoch": Arg(
'--analyze-per-epoch',
help='Run populate_indicators once per epoch.',
action='store_true',
default=False,
),
"print_all": Arg(
'--print-all',
help='Print all results, not only the best ones.',
@ -367,7 +374,7 @@ AVAILABLE_CLI_OPTIONS = {
metavar='BASE_CURRENCY',
),
"trading_mode": Arg(
'--trading-mode',
'--trading-mode', '--tradingmode',
help='Select Trading mode',
choices=constants.TRADING_MODES,
),
@ -434,6 +441,11 @@ AVAILABLE_CLI_OPTIONS = {
help='Storage format for downloaded trades data. (default: `jsongz`).',
choices=constants.AVAILABLE_DATAHANDLERS,
),
"show_timerange": Arg(
'--show-timerange',
help='Show timerange available for available data. (May take a while to calculate).',
action='store_true',
),
"exchange": Arg(
'--exchange',
help=f'Exchange name (default: `{constants.DEFAULT_EXCHANGE}`). '
@ -450,7 +462,7 @@ AVAILABLE_CLI_OPTIONS = {
),
"prepend_data": Arg(
'--prepend',
help='Allow data prepending.',
help='Allow data prepending. (Data-appending is disabled)',
action='store_true',
),
"erase": Arg(
@ -647,4 +659,14 @@ AVAILABLE_CLI_OPTIONS = {
nargs='+',
default=[],
),
"freqaimodel": Arg(
'--freqaimodel',
help='Specify a custom freqaimodels.',
metavar='NAME',
),
"freqaimodel_path": Arg(
'--freqaimodel-path',
help='Specify additional lookup path for freqaimodels.',
metavar='PATH',
),
}

View File

@ -5,14 +5,14 @@ from datetime import datetime, timedelta
from typing import Any, Dict, List
from freqtrade.configuration import TimeRange, setup_utils_configuration
from freqtrade.constants import DATETIME_PRINT_FORMAT
from freqtrade.data.converter import convert_ohlcv_format, convert_trades_format
from freqtrade.data.history import (convert_trades_to_ohlcv, refresh_backtest_ohlcv_data,
refresh_backtest_trades_data)
from freqtrade.enums import CandleType, RunMode, TradingMode
from freqtrade.exceptions import OperationalException
from freqtrade.exchange import timeframe_to_minutes
from freqtrade.exchange.exchange import market_is_active
from freqtrade.plugins.pairlist.pairlist_helpers import expand_pairlist
from freqtrade.exchange import market_is_active, timeframe_to_minutes
from freqtrade.plugins.pairlist.pairlist_helpers import dynamic_expand_pairlist, expand_pairlist
from freqtrade.resolvers import ExchangeResolver
@ -50,7 +50,8 @@ def start_download_data(args: Dict[str, Any]) -> None:
exchange = ExchangeResolver.load_exchange(config['exchange']['name'], config, validate=False)
markets = [p for p, m in exchange.markets.items() if market_is_active(m)
or config.get('include_inactive')]
expanded_pairs = expand_pairlist(config['pairs'], markets)
expanded_pairs = dynamic_expand_pairlist(config, markets)
# Manual validations of relevant settings
if not config['exchange'].get('skip_pair_validation', False):
@ -79,7 +80,7 @@ def start_download_data(args: Dict[str, Any]) -> None:
data_format_trades=config['dataformat_trades'],
)
else:
if not exchange._ft_has.get('ohlcv_has_history', True):
if not exchange.get_option('ohlcv_has_history', True):
raise OperationalException(
f"Historic klines not available for {exchange.name}. "
"Please use `--dl-trades` instead for this exchange "
@ -176,6 +177,7 @@ def start_list_data(args: Dict[str, Any]) -> None:
paircombs = [comb for comb in paircombs if comb[0] in args['pairs']]
print(f"Found {len(paircombs)} pair / timeframe combinations.")
if not config.get('show_timerange'):
groupedpair = defaultdict(list)
for pair, timeframe, candle_type in sorted(
paircombs,
@ -190,3 +192,16 @@ def start_list_data(args: Dict[str, Any]) -> None:
],
headers=("Pair", "Timeframe", "Type"),
tablefmt='psql', stralign='right'))
else:
paircombs1 = [(
pair, timeframe, candle_type,
*dhc.ohlcv_data_min_max(pair, timeframe, candle_type)
) for pair, timeframe, candle_type in paircombs]
print(tabulate([
(pair, timeframe, candle_type,
start.strftime(DATETIME_PRINT_FORMAT),
end.strftime(DATETIME_PRINT_FORMAT))
for pair, timeframe, candle_type, start, end in paircombs1
],
headers=("Pair", "Timeframe", "Type", 'From', 'To'),
tablefmt='psql', stralign='right'))

View File

@ -4,5 +4,4 @@ from freqtrade.configuration.check_exchange import check_exchange
from freqtrade.configuration.config_setup import setup_utils_configuration
from freqtrade.configuration.config_validation import validate_config_consistency
from freqtrade.configuration.configuration import Configuration
from freqtrade.configuration.PeriodicCache import PeriodicCache
from freqtrade.configuration.timerange import TimeRange

View File

@ -97,6 +97,8 @@ class Configuration:
self._process_analyze_options(config)
self._process_freqai_options(config)
# Check if the exchange set by the user is supported
check_exchange(config, config.get('experimental', {}).get('block_bad_exchanges', True))
@ -300,6 +302,9 @@ class Configuration:
self._args_to_config(config, argname='spaces',
logstring='Parameter -s/--spaces detected: {}')
self._args_to_config(config, argname='analyze_per_epoch',
logstring='Parameter --analyze-per-epoch detected.')
self._args_to_config(config, argname='print_all',
logstring='Parameter --print-all detected ...')
@ -424,6 +429,9 @@ class Configuration:
self._args_to_config(config, argname='dataformat_trades',
logstring='Using "{}" to store trades data.')
self._args_to_config(config, argname='show_timerange',
logstring='Detected --show-timerange')
def _process_data_options(self, config: Dict[str, Any]) -> None:
self._args_to_config(config, argname='new_pairs_days',
logstring='Detected --new-pairs-days: {}')
@ -461,6 +469,16 @@ class Configuration:
config.update({'runmode': self.runmode})
def _process_freqai_options(self, config: Dict[str, Any]) -> None:
self._args_to_config(config, argname='freqaimodel',
logstring='Using freqaimodel class name: {}')
self._args_to_config(config, argname='freqaimodel_path',
logstring='Using freqaimodel path: {}')
return
def _args_to_config(self, config: Dict[str, Any], argname: str,
logstring: str, logfun: Optional[Callable] = None,
deprecated_msg: Optional[str] = None) -> None:

View File

@ -55,6 +55,7 @@ FTHYPT_FILEVERSION = 'fthypt_fileversion'
USERPATH_HYPEROPTS = 'hyperopts'
USERPATH_STRATEGIES = 'strategies'
USERPATH_NOTEBOOKS = 'notebooks'
USERPATH_FREQAIMODELS = 'freqaimodels'
TELEGRAM_SETTING_OPTIONS = ['on', 'off', 'silent']
WEBHOOK_FORMAT_OPTIONS = ['form', 'json', 'raw']
@ -240,6 +241,7 @@ CONF_SCHEMA = {
},
'exchange': {'$ref': '#/definitions/exchange'},
'edge': {'$ref': '#/definitions/edge'},
'freqai': {'$ref': '#/definitions/freqai'},
'experimental': {
'type': 'object',
'properties': {
@ -317,6 +319,10 @@ CONF_SCHEMA = {
'type': 'string',
'enum': ['off', 'ohlc'],
},
'strategy_msg': {
'type': 'string',
'enum': TELEGRAM_SETTING_OPTIONS,
},
}
},
'reload': {'type': 'boolean'},
@ -476,8 +482,61 @@ CONF_SCHEMA = {
'remove_pumps': {'type': 'boolean'}
},
'required': ['process_throttle_secs', 'allowed_risk']
},
"freqai": {
"type": "object",
"properties": {
"enabled": {"type": "boolean", "default": False},
"keras": {"type": "boolean", "default": False},
"conv_width": {"type": "integer", "default": 2},
"train_period_days": {"type": "integer", "default": 0},
"backtest_period_days": {"type": "number", "default": 7},
"identifier": {"type": "string", "default": "example"},
"feature_parameters": {
"type": "object",
"properties": {
"include_corr_pairlist": {"type": "array"},
"include_timeframes": {"type": "array"},
"label_period_candles": {"type": "integer"},
"include_shifted_candles": {"type": "integer", "default": 0},
"DI_threshold": {"type": "number", "default": 0},
"weight_factor": {"type": "number", "default": 0},
"principal_component_analysis": {"type": "boolean", "default": False},
"use_SVM_to_remove_outliers": {"type": "boolean", "default": False},
"svm_params": {"type": "object",
"properties": {
"shuffle": {"type": "boolean", "default": False},
"nu": {"type": "number", "default": 0.1}
},
}
},
"required": ["include_timeframes", "include_corr_pairlist", ]
},
"data_split_parameters": {
"type": "object",
"properties": {
"test_size": {"type": "number"},
"random_state": {"type": "integer"},
},
},
"model_training_parameters": {
"type": "object",
"properties": {
"n_estimators": {"type": "integer", "default": 1000}
},
},
},
"required": [
"enabled",
"train_period_days",
"backtest_period_days",
"identifier",
"feature_parameters",
"data_split_parameters",
"model_training_parameters"
]
},
},
}
SCHEMA_TRADE_REQUIRED = [

View File

@ -5,6 +5,7 @@ including ticker and orderbook data, live and historical candle (OHLCV) data
Common Interface for bot and strategy to access data.
"""
import logging
from collections import deque
from datetime import datetime, timezone
from typing import Any, Dict, List, Optional, Tuple
@ -16,6 +17,7 @@ from freqtrade.data.history import load_pair_history
from freqtrade.enums import CandleType, RunMode
from freqtrade.exceptions import ExchangeError, OperationalException
from freqtrade.exchange import Exchange, timeframe_to_seconds
from freqtrade.util import PeriodicCache
logger = logging.getLogger(__name__)
@ -33,6 +35,10 @@ class DataProvider:
self.__cached_pairs: Dict[PairWithTimeframe, Tuple[DataFrame, datetime]] = {}
self.__slice_index: Optional[int] = None
self.__cached_pairs_backtesting: Dict[PairWithTimeframe, DataFrame] = {}
self._msg_queue: deque = deque()
self.__msg_cache = PeriodicCache(
maxsize=1000, ttl=timeframe_to_seconds(self._config.get('timeframe', '1h')))
def _set_dataframe_max_index(self, limit_index: int):
"""
@ -265,3 +271,20 @@ class DataProvider:
if self._exchange is None:
raise OperationalException(NO_EXCHANGE_EXCEPTION)
return self._exchange.fetch_l2_order_book(pair, maximum)
def send_msg(self, message: str, *, always_send: bool = False) -> None:
"""
Send custom RPC Notifications from your bot.
Will not send any bot in modes other than Dry-run or Live.
:param message: Message to be sent. Must be below 4096.
:param always_send: If False, will send the message only once per candle, and surpress
identical messages.
Careful as this can end up spaming your chat.
Defaults to False
"""
if self.runmode not in (RunMode.DRY_RUN, RunMode.LIVE):
return
if always_send or message not in self.__msg_cache:
self._msg_queue.append(message)
self.__msg_cache[message] = True

View File

@ -7,9 +7,8 @@ import numpy as np
import pandas as pd
from freqtrade.configuration import TimeRange
from freqtrade.constants import (DEFAULT_DATAFRAME_COLUMNS, DEFAULT_TRADES_COLUMNS,
ListPairsWithTimeframes, TradeList)
from freqtrade.enums import CandleType, TradingMode
from freqtrade.constants import DEFAULT_DATAFRAME_COLUMNS, DEFAULT_TRADES_COLUMNS, TradeList
from freqtrade.enums import CandleType
from .idatahandler import IDataHandler
@ -21,29 +20,6 @@ class HDF5DataHandler(IDataHandler):
_columns = DEFAULT_DATAFRAME_COLUMNS
@classmethod
def ohlcv_get_available_data(
cls, datadir: Path, trading_mode: TradingMode) -> ListPairsWithTimeframes:
"""
Returns a list of all pairs with ohlcv data available in this datadir
:param datadir: Directory to search for ohlcv files
:param trading_mode: trading-mode to be used
:return: List of Tuples of (pair, timeframe)
"""
if trading_mode == TradingMode.FUTURES:
datadir = datadir.joinpath('futures')
_tmp = [
re.search(
cls._OHLCV_REGEX, p.name
) for p in datadir.glob("*.h5")
]
return [
(
cls.rebuild_pair_from_filename(match[1]),
cls.rebuild_timeframe_from_filename(match[2]),
CandleType.from_string(match[3])
) for match in _tmp if match and len(match.groups()) > 1]
@classmethod
def ohlcv_get_pairs(cls, datadir: Path, timeframe: str, candle_type: CandleType) -> List[str]:
"""

View File

@ -56,7 +56,7 @@ def load_pair_history(pair: str,
fill_missing=fill_up_missing,
drop_incomplete=drop_incomplete,
startup_candles=startup_candles,
candle_type=candle_type
candle_type=candle_type,
)
@ -97,14 +97,15 @@ def load_data(datadir: Path,
fill_up_missing=fill_up_missing,
startup_candles=startup_candles,
data_handler=data_handler,
candle_type=candle_type
candle_type=candle_type,
)
if not hist.empty:
result[pair] = hist
else:
if candle_type is CandleType.FUNDING_RATE and user_futures_funding_rate is not None:
logger.warn(f"{pair} using user specified [{user_futures_funding_rate}]")
result[pair] = DataFrame(columns=["open", "close", "high", "low", "volume"])
elif candle_type not in (CandleType.SPOT, CandleType.FUTURES):
result[pair] = DataFrame(columns=["date", "open", "close", "high", "low", "volume"])
if fail_without_data and not result:
raise OperationalException("No data found. Terminating.")
@ -301,8 +302,8 @@ def refresh_backtest_ohlcv_data(exchange: Exchange, pairs: List[str], timeframes
if trading_mode == 'futures':
# Predefined candletype (and timeframe) depending on exchange
# Downloads what is necessary to backtest based on futures data.
tf_mark = exchange._ft_has['mark_ohlcv_timeframe']
fr_candle_type = CandleType.from_string(exchange._ft_has['mark_ohlcv_price'])
tf_mark = exchange.get_option('mark_ohlcv_timeframe')
fr_candle_type = CandleType.from_string(exchange.get_option('mark_ohlcv_price'))
# All exchanges need FundingRate for futures trading.
# The timeframe is aligned to the mark-price timeframe.
for funding_candle_type in (CandleType.FUNDING_RATE, fr_candle_type):
@ -329,13 +330,12 @@ def _download_trades_history(exchange: Exchange,
try:
until = None
since = 0
if timerange:
if timerange.starttype == 'date':
since = timerange.startts * 1000
if timerange.stoptype == 'date':
until = timerange.stopts * 1000
else:
since = arrow.utcnow().shift(days=-new_pairs_days).int_timestamp * 1000
trades = data_handler.trades_load(pair)
@ -348,6 +348,9 @@ def _download_trades_history(exchange: Exchange,
logger.info(f"Start earlier than available data. Redownloading trades for {pair}...")
trades = []
if not since:
since = arrow.utcnow().shift(days=-new_pairs_days).int_timestamp * 1000
from_id = trades[-1][1] if trades else None
if trades and since < trades[-1][0]:
# Reset since to the last available point

View File

@ -9,7 +9,7 @@ from abc import ABC, abstractmethod
from copy import deepcopy
from datetime import datetime, timezone
from pathlib import Path
from typing import List, Optional, Type
from typing import List, Optional, Tuple, Type
from pandas import DataFrame
@ -39,15 +39,26 @@ class IDataHandler(ABC):
raise NotImplementedError()
@classmethod
@abstractmethod
def ohlcv_get_available_data(
cls, datadir: Path, trading_mode: TradingMode) -> ListPairsWithTimeframes:
"""
Returns a list of all pairs with ohlcv data available in this datadir
:param datadir: Directory to search for ohlcv files
:param trading_mode: trading-mode to be used
:return: List of Tuples of (pair, timeframe)
:return: List of Tuples of (pair, timeframe, CandleType)
"""
if trading_mode == TradingMode.FUTURES:
datadir = datadir.joinpath('futures')
_tmp = [
re.search(
cls._OHLCV_REGEX, p.name
) for p in datadir.glob(f"*.{cls._get_file_extension()}")]
return [
(
cls.rebuild_pair_from_filename(match[1]),
cls.rebuild_timeframe_from_filename(match[2]),
CandleType.from_string(match[3])
) for match in _tmp if match and len(match.groups()) > 1]
@classmethod
@abstractmethod
@ -73,6 +84,18 @@ class IDataHandler(ABC):
:return: None
"""
def ohlcv_data_min_max(self, pair: str, timeframe: str,
candle_type: CandleType) -> Tuple[datetime, datetime]:
"""
Returns the min and max timestamp for the given pair and timeframe.
:param pair: Pair to get min/max for
:param timeframe: Timeframe to get min/max for
:param candle_type: Any of the enum CandleType (must match trading mode!)
:return: (min, max)
"""
data = self._ohlcv_load(pair, timeframe, None, candle_type)
return data.iloc[0]['date'].to_pydatetime(), data.iloc[-1]['date'].to_pydatetime()
@abstractmethod
def _ohlcv_load(self, pair: str, timeframe: str, timerange: Optional[TimeRange],
candle_type: CandleType

View File

@ -8,9 +8,9 @@ from pandas import DataFrame, read_json, to_datetime
from freqtrade import misc
from freqtrade.configuration import TimeRange
from freqtrade.constants import DEFAULT_DATAFRAME_COLUMNS, ListPairsWithTimeframes, TradeList
from freqtrade.constants import DEFAULT_DATAFRAME_COLUMNS, TradeList
from freqtrade.data.converter import trades_dict_to_list
from freqtrade.enums import CandleType, TradingMode
from freqtrade.enums import CandleType
from .idatahandler import IDataHandler
@ -23,28 +23,6 @@ class JsonDataHandler(IDataHandler):
_use_zip = False
_columns = DEFAULT_DATAFRAME_COLUMNS
@classmethod
def ohlcv_get_available_data(
cls, datadir: Path, trading_mode: TradingMode) -> ListPairsWithTimeframes:
"""
Returns a list of all pairs with ohlcv data available in this datadir
:param datadir: Directory to search for ohlcv files
:param trading_mode: trading-mode to be used
:return: List of Tuples of (pair, timeframe)
"""
if trading_mode == 'futures':
datadir = datadir.joinpath('futures')
_tmp = [
re.search(
cls._OHLCV_REGEX, p.name
) for p in datadir.glob(f"*.{cls._get_file_extension()}")]
return [
(
cls.rebuild_pair_from_filename(match[1]),
cls.rebuild_timeframe_from_filename(match[2]),
CandleType.from_string(match[3])
) for match in _tmp if match and len(match.groups()) > 1]
@classmethod
def ohlcv_get_pairs(cls, datadir: Path, timeframe: str, candle_type: CandleType) -> List[str]:
"""

View File

@ -15,7 +15,7 @@ from freqtrade.constants import DATETIME_PRINT_FORMAT, UNLIMITED_STAKE_AMOUNT
from freqtrade.data.history import get_timerange, load_data, refresh_data
from freqtrade.enums import CandleType, ExitType, RunMode
from freqtrade.exceptions import OperationalException
from freqtrade.exchange.exchange import timeframe_to_seconds
from freqtrade.exchange import timeframe_to_seconds
from freqtrade.plugins.pairlist.pairlist_helpers import expand_pairlist
from freqtrade.strategy.interface import IStrategy

View File

@ -3,6 +3,7 @@ from freqtrade.enums.backteststate import BacktestState
from freqtrade.enums.candletype import CandleType
from freqtrade.enums.exitchecktuple import ExitCheckTuple
from freqtrade.enums.exittype import ExitType
from freqtrade.enums.hyperoptstate import HyperoptState
from freqtrade.enums.marginmode import MarginMode
from freqtrade.enums.ordertypevalue import OrderTypeValues
from freqtrade.enums.rpcmessagetype import RPCMessageType

View File

@ -9,10 +9,12 @@ class ExitType(Enum):
STOP_LOSS = "stop_loss"
STOPLOSS_ON_EXCHANGE = "stoploss_on_exchange"
TRAILING_STOP_LOSS = "trailing_stop_loss"
LIQUIDATION = "liquidation"
EXIT_SIGNAL = "exit_signal"
FORCE_EXIT = "force_exit"
EMERGENCY_EXIT = "emergency_exit"
CUSTOM_EXIT = "custom_exit"
PARTIAL_EXIT = "partial_exit"
NONE = ""
def __str__(self):

View File

@ -0,0 +1,12 @@
from enum import Enum
class HyperoptState(Enum):
""" Hyperopt states """
STARTUP = 1
DATALOAD = 2
INDICATORS = 3
OPTIMIZE = 4
def __str__(self):
return f"{self.name.lower()}"

View File

@ -17,6 +17,8 @@ class RPCMessageType(Enum):
PROTECTION_TRIGGER = 'protection_trigger'
PROTECTION_TRIGGER_GLOBAL = 'protection_trigger_global'
STRATEGY_MSG = 'strategy_msg'
def __repr__(self):
return self.value

View File

@ -9,12 +9,14 @@ from freqtrade.exchange.bitpanda import Bitpanda
from freqtrade.exchange.bittrex import Bittrex
from freqtrade.exchange.bybit import Bybit
from freqtrade.exchange.coinbasepro import Coinbasepro
from freqtrade.exchange.exchange import (available_exchanges, ccxt_exchanges,
from freqtrade.exchange.exchange import (amount_to_contract_precision, amount_to_contracts,
amount_to_precision, available_exchanges, ccxt_exchanges,
contracts_to_amount, date_minus_candles,
is_exchange_known_ccxt, is_exchange_officially_supported,
market_is_active, timeframe_to_minutes, timeframe_to_msecs,
timeframe_to_next_date, timeframe_to_prev_date,
timeframe_to_seconds, validate_exchange,
validate_exchanges)
market_is_active, price_to_precision, timeframe_to_minutes,
timeframe_to_msecs, timeframe_to_next_date,
timeframe_to_prev_date, timeframe_to_seconds,
validate_exchange, validate_exchanges)
from freqtrade.exchange.ftx import Ftx
from freqtrade.exchange.gateio import Gateio
from freqtrade.exchange.hitbtc import Hitbtc

View File

@ -137,23 +137,27 @@ class Binance(Exchange):
pair: str,
open_rate: float, # Entry price of position
is_short: bool,
position: float, # Absolute value of position size
amount: float,
stake_amount: float,
wallet_balance: float, # Or margin balance
mm_ex_1: float = 0.0, # (Binance) Cross only
upnl_ex_1: float = 0.0, # (Binance) Cross only
) -> Optional[float]:
"""
Important: Must be fetching data from cached values as this is used by backtesting!
MARGIN: https://www.binance.com/en/support/faq/f6b010588e55413aa58b7d63ee0125ed
PERPETUAL: https://www.binance.com/en/support/faq/b3c689c1f50a44cabb3a84e663b81d93
:param exchange_name:
:param open_rate: (EP1) Entry price of position
:param open_rate: Entry price of position
:param is_short: True if the trade is a short, false otherwise
:param position: Absolute value of position size (in base currency)
:param wallet_balance: (WB)
:param amount: Absolute value of position size incl. leverage (in base currency)
:param stake_amount: Stake amount - Collateral in settle currency.
:param trading_mode: SPOT, MARGIN, FUTURES, etc.
:param margin_mode: Either ISOLATED or CROSS
:param wallet_balance: Amount of margin_mode in the wallet being used to trade
Cross-Margin Mode: crossWalletBalance
Isolated-Margin Mode: isolatedWalletBalance
:param maintenance_amt:
# * Only required for Cross
:param mm_ex_1: (TMM)
@ -165,12 +169,11 @@ class Binance(Exchange):
"""
side_1 = -1 if is_short else 1
position = abs(position)
cross_vars = upnl_ex_1 - mm_ex_1 if self.margin_mode == MarginMode.CROSS else 0.0
# mm_ratio: Binance's formula specifies maintenance margin rate which is mm_ratio * 100%
# maintenance_amt: (CUM) Maintenance Amount of position
mm_ratio, maintenance_amt = self.get_maintenance_ratio_and_amt(pair, position)
mm_ratio, maintenance_amt = self.get_maintenance_ratio_and_amt(pair, stake_amount)
if (maintenance_amt is None):
raise OperationalException(
@ -182,9 +185,9 @@ class Binance(Exchange):
return (
(
(wallet_balance + cross_vars + maintenance_amt) -
(side_1 * position * open_rate)
(side_1 * amount * open_rate)
) / (
(position * mm_ratio) - (side_1 * position)
(amount * mm_ratio) - (side_1 * amount)
)
)
else:

File diff suppressed because it is too large Load Diff

View File

@ -16,7 +16,8 @@ import arrow
import ccxt
import ccxt.async_support as ccxt_async
from cachetools import TTLCache
from ccxt import ROUND_DOWN, ROUND_UP, TICK_SIZE, TRUNCATE, Precise, decimal_to_precision
from ccxt import ROUND_DOWN, ROUND_UP, TICK_SIZE, TRUNCATE, decimal_to_precision
from dateutil import parser
from pandas import DataFrame
from freqtrade.constants import (DEFAULT_AMOUNT_RESERVE_PERCENT, NON_OPEN_EXCHANGE_STATES, BuySell,
@ -30,8 +31,10 @@ from freqtrade.exchange.common import (API_FETCH_ORDER_RETRY_COUNT, BAD_EXCHANGE
EXCHANGE_HAS_OPTIONAL, EXCHANGE_HAS_REQUIRED,
SUPPORTED_EXCHANGES, remove_credentials, retrier,
retrier_async)
from freqtrade.misc import chunks, deep_merge_dicts, safe_value_fallback2
from freqtrade.misc import (chunks, deep_merge_dicts, file_dump_json, file_load_json,
safe_value_fallback2)
from freqtrade.plugins.pairlist.pairlist_helpers import expand_pairlist
from freqtrade.util import FtPrecise
CcxtModuleType = Any
@ -51,8 +54,8 @@ class Exchange:
# Parameters to add directly to buy/sell calls (like agreeing to trading agreement)
_params: Dict = {}
# Additional headers - added to the ccxt object
_headers: Dict = {}
# Additional parameters - added to the ccxt object
_ccxt_params: Dict = {}
# Dict to specify which options each exchange implements
# This defines defaults, which can be selectively overridden by subclasses using _ft_has
@ -115,6 +118,7 @@ class Exchange:
self._last_markets_refresh: int = 0
# Cache for 10 minutes ...
self._cache_lock = Lock()
self._fetch_tickers_cache: TTLCache = TTLCache(maxsize=2, ttl=60 * 10)
# Cache values for 1800 to avoid frequent polling of the exchange for prices
# Caching only applies to RPC methods, so prices for open trades are still
@ -238,9 +242,9 @@ class Exchange:
}
if ccxt_kwargs:
logger.info('Applying additional ccxt config: %s', ccxt_kwargs)
if self._headers:
# Inject static headers after the above output to not confuse users.
ccxt_kwargs = deep_merge_dicts({'headers': self._headers}, ccxt_kwargs)
if self._ccxt_params:
# Inject static options after the above output to not confuse users.
ccxt_kwargs = deep_merge_dicts(self._ccxt_params, ccxt_kwargs)
if ccxt_kwargs:
ex_config.update(ccxt_kwargs)
try:
@ -404,7 +408,7 @@ class Exchange:
else:
return DataFrame()
def _get_contract_size(self, pair: str) -> float:
def get_contract_size(self, pair: str) -> float:
if self.trading_mode == TradingMode.FUTURES:
market = self.markets[pair]
contract_size: float = 1.0
@ -417,7 +421,7 @@ class Exchange:
def _trades_contracts_to_amount(self, trades: List) -> List:
if len(trades) > 0 and 'symbol' in trades[0]:
contract_size = self._get_contract_size(trades[0]['symbol'])
contract_size = self.get_contract_size(trades[0]['symbol'])
if contract_size != 1:
for trade in trades:
trade['amount'] = trade['amount'] * contract_size
@ -425,7 +429,7 @@ class Exchange:
def _order_contracts_to_amount(self, order: Dict) -> Dict:
if 'symbol' in order and order['symbol'] is not None:
contract_size = self._get_contract_size(order['symbol'])
contract_size = self.get_contract_size(order['symbol'])
if contract_size != 1:
for prop in self._ft_has.get('order_props_in_contracts', []):
if prop in order and order[prop] is not None:
@ -434,19 +438,13 @@ class Exchange:
def _amount_to_contracts(self, pair: str, amount: float) -> float:
contract_size = self._get_contract_size(pair)
if contract_size and contract_size != 1:
return amount / contract_size
else:
return amount
contract_size = self.get_contract_size(pair)
return amount_to_contracts(amount, contract_size)
def _contracts_to_amount(self, pair: str, num_contracts: float) -> float:
contract_size = self._get_contract_size(pair)
if contract_size and contract_size != 1:
return num_contracts * contract_size
else:
return num_contracts
contract_size = self.get_contract_size(pair)
return contracts_to_amount(num_contracts, contract_size)
def set_sandbox(self, api: ccxt.Exchange, exchange_config: dict, name: str) -> None:
if exchange_config.get('sandbox'):
@ -670,6 +668,12 @@ class Exchange:
f"Freqtrade does not support {mm_value} {trading_mode.value} on {self.name}"
)
def get_option(self, param: str, default: Any = None) -> Any:
"""
Get parameter value from _ft_has
"""
return self._ft_has.get(param, default)
def exchange_has(self, endpoint: str) -> bool:
"""
Checks if exchange implements a specific API endpoint.
@ -679,45 +683,35 @@ class Exchange:
"""
return endpoint in self._api.has and self._api.has[endpoint]
def get_precision_amount(self, pair: str) -> Optional[float]:
"""
Returns the amount precision of the exchange.
:param pair: Pair to get precision for
:return: precision for amount or None. Must be used in combination with precisionMode
"""
return self.markets.get(pair, {}).get('precision', {}).get('amount', None)
def get_precision_price(self, pair: str) -> Optional[float]:
"""
Returns the price precision of the exchange.
:param pair: Pair to get precision for
:return: precision for price or None. Must be used in combination with precisionMode
"""
return self.markets.get(pair, {}).get('precision', {}).get('price', None)
def amount_to_precision(self, pair: str, amount: float) -> float:
"""
Returns the amount to buy or sell to a precision the Exchange accepts
Re-implementation of ccxt internal methods - ensuring we can test the result is correct
based on our definitions.
"""
if self.markets[pair]['precision']['amount'] is not None:
amount = float(decimal_to_precision(amount, rounding_mode=TRUNCATE,
precision=self.markets[pair]['precision']['amount'],
counting_mode=self.precisionMode,
))
return amount
"""
return amount_to_precision(amount, self.get_precision_amount(pair), self.precisionMode)
def price_to_precision(self, pair: str, price: float) -> float:
"""
Returns the price rounded up to the precision the Exchange accepts.
Partial Re-implementation of ccxt internal method decimal_to_precision(),
which does not support rounding up
TODO: If ccxt supports ROUND_UP for decimal_to_precision(), we could remove this and
align with amount_to_precision().
Rounds up
"""
if self.markets[pair]['precision']['price']:
# price = float(decimal_to_precision(price, rounding_mode=ROUND,
# precision=self.markets[pair]['precision']['price'],
# counting_mode=self.precisionMode,
# ))
if self.precisionMode == TICK_SIZE:
precision = Precise(str(self.markets[pair]['precision']['price']))
price_str = Precise(str(price))
missing = price_str % precision
if not missing == Precise("0"):
price = round(float(str(price_str - missing + precision)), 14)
else:
symbol_prec = self.markets[pair]['precision']['price']
big_price = price * pow(10, symbol_prec)
price = ceil(big_price) / pow(10, symbol_prec)
return price
return price_to_precision(price, self.get_precision_price(pair), self.precisionMode)
def price_get_one_pip(self, pair: str, price: float) -> float:
"""
@ -849,6 +843,7 @@ class Exchange:
dry_order.update({
'average': average,
'filled': _amount,
'remaining': 0.0,
'cost': (dry_order['amount'] * average) / leverage
})
# market orders will always incurr taker fees
@ -1017,7 +1012,8 @@ class Exchange:
time_in_force: str = 'gtc',
) -> Dict:
if self._config['dry_run']:
dry_order = self.create_dry_run_order(pair, ordertype, side, amount, rate, leverage)
dry_order = self.create_dry_run_order(
pair, ordertype, side, amount, self.price_to_precision(pair, rate), leverage)
return dry_order
params = self._get_params(side, ordertype, leverage, reduceOnly, time_in_force)
@ -1332,11 +1328,19 @@ class Exchange:
raise OperationalException(e) from e
@retrier
def fetch_positions(self) -> List[Dict]:
def fetch_positions(self, pair: str = None) -> List[Dict]:
"""
Fetch positions from the exchange.
If no pair is given, all positions are returned.
:param pair: Pair for the query
"""
if self._config['dry_run'] or self.trading_mode != TradingMode.FUTURES:
return []
try:
positions: List[Dict] = self._api.fetch_positions()
symbols = []
if pair:
symbols.append(pair)
positions: List[Dict] = self._api.fetch_positions(symbols)
self._log_exchange_response('fetch_positions', positions)
return positions
except ccxt.DDoSProtection as e:
@ -1377,11 +1381,13 @@ class Exchange:
if not self.exchange_has('fetchBidsAsks'):
return {}
if cached:
with self._cache_lock:
tickers = self._fetch_tickers_cache.get('fetch_bids_asks')
if tickers:
return tickers
try:
tickers = self._api.fetch_bids_asks(symbols)
with self._cache_lock:
self._fetch_tickers_cache['fetch_bids_asks'] = tickers
return tickers
except ccxt.NotSupported as e:
@ -1403,11 +1409,13 @@ class Exchange:
:return: fetch_tickers result
"""
if cached:
with self._cache_lock:
tickers = self._fetch_tickers_cache.get('fetch_tickers')
if tickers:
return tickers
try:
tickers = self._api.fetch_tickers(symbols)
with self._cache_lock:
self._fetch_tickers_cache['fetch_tickers'] = tickers
return tickers
except ccxt.NotSupported as e:
@ -1499,7 +1507,8 @@ class Exchange:
return price_side
def get_rate(self, pair: str, refresh: bool,
side: EntryExit, is_short: bool) -> float:
side: EntryExit, is_short: bool,
order_book: Optional[dict] = None, ticker: Optional[dict] = None) -> float:
"""
Calculates bid/ask target
bid rate - between current ask price and last price
@ -1516,6 +1525,7 @@ class Exchange:
cache_rate: TTLCache = self._entry_rate_cache if side == "entry" else self._exit_rate_cache
if not refresh:
with self._cache_lock:
rate = cache_rate.get(pair)
# Check if cache has been invalidated
if rate:
@ -1531,6 +1541,7 @@ class Exchange:
if conf_strategy.get('use_order_book', False):
order_book_top = conf_strategy.get('order_book_top', 1)
if order_book is None:
order_book = self.fetch_l2_order_book(pair, order_book_top)
logger.debug('order_book %s', order_book)
# top 1 = index 0
@ -1538,14 +1549,15 @@ class Exchange:
rate = order_book[f"{price_side}s"][order_book_top - 1][0]
except (IndexError, KeyError) as e:
logger.warning(
f"{name} Price at location {order_book_top} from orderbook could not be "
f"determined. Orderbook: {order_book}"
f"{pair} - {name} Price at location {order_book_top} from orderbook "
f"could not be determined. Orderbook: {order_book}"
)
raise PricingError from e
logger.debug(f"{name} price from orderbook {price_side_word}"
logger.debug(f"{pair} - {name} price from orderbook {price_side_word}"
f"side - top {order_book_top} order book {side} rate {rate:.8f}")
else:
logger.debug(f"Using Last {price_side_word} / Last Price")
if ticker is None:
ticker = self.fetch_ticker(pair)
ticker_rate = ticker[price_side]
if ticker['last'] and ticker_rate:
@ -1559,10 +1571,39 @@ class Exchange:
if rate is None:
raise PricingError(f"{name}-Rate for {pair} was empty.")
with self._cache_lock:
cache_rate[pair] = rate
return rate
def get_rates(self, pair: str, refresh: bool, is_short: bool) -> Tuple[float, float]:
entry_rate = None
exit_rate = None
if not refresh:
with self._cache_lock:
entry_rate = self._entry_rate_cache.get(pair)
exit_rate = self._exit_rate_cache.get(pair)
if entry_rate:
logger.debug(f"Using cached buy rate for {pair}.")
if exit_rate:
logger.debug(f"Using cached sell rate for {pair}.")
entry_pricing = self._config.get('entry_pricing', {})
exit_pricing = self._config.get('exit_pricing', {})
order_book = ticker = None
if not entry_rate and entry_pricing.get('use_order_book', False):
order_book_top = max(entry_pricing.get('order_book_top', 1),
exit_pricing.get('order_book_top', 1))
order_book = self.fetch_l2_order_book(pair, order_book_top)
entry_rate = self.get_rate(pair, refresh, 'entry', is_short, order_book=order_book)
elif not entry_rate:
ticker = self.fetch_ticker(pair)
entry_rate = self.get_rate(pair, refresh, 'entry', is_short, ticker=ticker)
if not exit_rate:
exit_rate = self.get_rate(pair, refresh, 'exit',
is_short, order_book=order_book, ticker=ticker)
return entry_rate, exit_rate
# Fee handling
@retrier
@ -2168,6 +2209,7 @@ class Exchange:
@retrier_async
async def get_market_leverage_tiers(self, symbol: str) -> Tuple[str, List[Dict]]:
""" Leverage tiers per symbol """
try:
tier = await self._api_async.fetch_market_leverage_tiers(symbol)
return symbol, tier
@ -2199,20 +2241,34 @@ class Exchange:
tiers: Dict[str, List[Dict]] = {}
tiers_cached = self.load_cached_leverage_tiers(self._config['stake_currency'])
if tiers_cached:
tiers = tiers_cached
coros = [
self.get_market_leverage_tiers(symbol)
for symbol in sorted(symbols) if symbol not in tiers]
# Be verbose here, as this delays startup by ~1 minute.
if coros:
logger.info(
f"Initializing leverage_tiers for {len(symbols)} markets. "
"This will take about a minute.")
else:
logger.info("Using cached leverage_tiers.")
coros = [self.get_market_leverage_tiers(symbol) for symbol in sorted(symbols)]
async def gather_results():
return await asyncio.gather(*input_coro, return_exceptions=True)
for input_coro in chunks(coros, 100):
results = self.loop.run_until_complete(
asyncio.gather(*input_coro, return_exceptions=True))
with self._loop_lock:
results = self.loop.run_until_complete(gather_results())
for symbol, res in results:
tiers[symbol] = res
if len(coros) > 0:
self.cache_leverage_tiers(tiers, self._config['stake_currency'])
logger.info(f"Done initializing {len(symbols)} markets.")
return tiers
@ -2221,6 +2277,30 @@ class Exchange:
else:
return {}
def cache_leverage_tiers(self, tiers: Dict[str, List[Dict]], stake_currency: str) -> None:
filename = self._config['datadir'] / "futures" / f"leverage_tiers_{stake_currency}.json"
if not filename.parent.is_dir():
filename.parent.mkdir(parents=True)
data = {
"updated": datetime.now(timezone.utc),
"data": tiers,
}
file_dump_json(filename, data)
def load_cached_leverage_tiers(self, stake_currency: str) -> Optional[Dict[str, List[Dict]]]:
filename = self._config['datadir'] / "futures" / f"leverage_tiers_{stake_currency}.json"
if filename.is_file():
tiers = file_load_json(filename)
updated = tiers.get('updated')
if updated:
updated_dt = parser.parse(updated)
if updated_dt < datetime.now(timezone.utc) - timedelta(days=1):
logger.info("Cached leverage tiers are outdated. Will update.")
return None
return tiers['data']
return None
def fill_leverage_tiers(self) -> None:
"""
Assigns property _leverage_tiers to a dictionary of information about the leverage
@ -2236,10 +2316,10 @@ class Exchange:
def parse_leverage_tier(self, tier) -> Dict:
info = tier.get('info', {})
return {
'min': tier['minNotional'],
'max': tier['maxNotional'],
'mmr': tier['maintenanceMarginRate'],
'lev': tier['maxLeverage'],
'minNotional': tier['minNotional'],
'maxNotional': tier['maxNotional'],
'maintenanceMarginRate': tier['maintenanceMarginRate'],
'maxLeverage': tier['maxLeverage'],
'maintAmt': float(info['cum']) if 'cum' in info else None,
}
@ -2268,18 +2348,18 @@ class Exchange:
pair_tiers = self._leverage_tiers[pair]
if stake_amount == 0:
return self._leverage_tiers[pair][0]['lev'] # Max lev for lowest amount
return self._leverage_tiers[pair][0]['maxLeverage'] # Max lev for lowest amount
for tier_index in range(len(pair_tiers)):
tier = pair_tiers[tier_index]
lev = tier['lev']
lev = tier['maxLeverage']
if tier_index < len(pair_tiers) - 1:
next_tier = pair_tiers[tier_index + 1]
next_floor = next_tier['min'] / next_tier['lev']
next_floor = next_tier['minNotional'] / next_tier['maxLeverage']
if next_floor > stake_amount: # Next tier min too high for stake amount
return min((tier['max'] / stake_amount), lev)
return min((tier['maxNotional'] / stake_amount), lev)
#
# With the two leverage tiers below,
# - a stake amount of 150 would mean a max leverage of (10000 / 150) = 66.66
@ -2300,10 +2380,11 @@ class Exchange:
#
else: # if on the last tier
if stake_amount > tier['max']: # If stake is > than max tradeable amount
if stake_amount > tier['maxNotional']:
# If stake is > than max tradeable amount
raise InvalidOrderException(f'Amount {stake_amount} too high for {pair}')
else:
return tier['lev']
return tier['maxLeverage']
raise OperationalException(
'Looped through all tiers without finding a max leverage. Should never be reached'
@ -2334,7 +2415,8 @@ class Exchange:
return
try:
self._api.set_leverage(symbol=pair, leverage=leverage)
res = self._api.set_leverage(symbol=pair, leverage=leverage)
self._log_exchange_response('set_leverage', res)
except ccxt.DDoSProtection as e:
raise DDosProtection(e) from e
except (ccxt.NetworkError, ccxt.ExchangeError) as e:
@ -2355,6 +2437,7 @@ class Exchange:
pair: str,
open_rate: float,
amount: float, # quote currency, includes leverage
stake_amount: float,
leverage: float,
is_short: bool
) -> Optional[float]:
@ -2362,23 +2445,22 @@ class Exchange:
if self.trading_mode in TradingMode.SPOT:
return None
elif (
self.margin_mode == MarginMode.ISOLATED and
self.trading_mode == TradingMode.FUTURES
):
wallet_balance = (amount * open_rate) / leverage
isolated_liq = self.get_or_calculate_liquidation_price(
pair=pair,
open_rate=open_rate,
is_short=is_short,
position=amount,
wallet_balance=wallet_balance,
amount=amount,
stake_amount=stake_amount,
wallet_balance=stake_amount, # In isolated mode, stake-amount = wallet size
mm_ex_1=0.0,
upnl_ex_1=0.0,
)
return isolated_liq
else:
raise OperationalException(
"Freqtrade only supports isolated futures for leverage trading")
"Freqtrade currently only supports futures for leverage trading.")
def funding_fee_cutoff(self, open_date: datetime):
"""
@ -2398,7 +2480,8 @@ class Exchange:
return
try:
self._api.set_margin_mode(margin_mode.value, pair, params)
res = self._api.set_margin_mode(margin_mode.value, pair, params)
self._log_exchange_response('set_margin_mode', res)
except ccxt.DDoSProtection as e:
raise DDosProtection(e) from e
except (ccxt.NetworkError, ccxt.ExchangeError) as e:
@ -2539,25 +2622,24 @@ class Exchange:
else:
return 0.0
@retrier
def get_or_calculate_liquidation_price(
self,
pair: str,
# Dry-run
open_rate: float, # Entry price of position
is_short: bool,
position: float, # Absolute value of position size
amount: float, # Absolute value of position size
stake_amount: float,
wallet_balance: float, # Or margin balance
mm_ex_1: float = 0.0, # (Binance) Cross only
upnl_ex_1: float = 0.0, # (Binance) Cross only
) -> Optional[float]:
"""
Set's the margin mode on the exchange to cross or isolated for a specific pair
:param pair: base/quote currency pair (e.g. "ADA/USDT")
"""
if self.trading_mode == TradingMode.SPOT:
return None
elif (self.trading_mode != TradingMode.FUTURES and self.margin_mode != MarginMode.ISOLATED):
elif (self.trading_mode != TradingMode.FUTURES):
raise OperationalException(
f"{self.name} does not support {self.margin_mode.value} {self.trading_mode.value}")
@ -2567,26 +2649,19 @@ class Exchange:
pair=pair,
open_rate=open_rate,
is_short=is_short,
position=position,
amount=amount,
stake_amount=stake_amount,
wallet_balance=wallet_balance,
mm_ex_1=mm_ex_1,
upnl_ex_1=upnl_ex_1
)
else:
try:
positions = self._api.fetch_positions([pair])
positions = self.fetch_positions(pair)
if len(positions) > 0:
pos = positions[0]
isolated_liq = pos['liquidationPrice']
else:
return None
except ccxt.DDoSProtection as e:
raise DDosProtection(e) from e
except (ccxt.NetworkError, ccxt.ExchangeError) as e:
raise TemporaryError(
f'Could not set margin mode due to {e.__class__.__name__}. Message: {e}') from e
except ccxt.BaseError as e:
raise OperationalException(e) from e
if isolated_liq:
buffer_amount = abs(open_rate - isolated_liq) * self.liquidation_buffer
@ -2604,22 +2679,24 @@ class Exchange:
pair: str,
open_rate: float, # Entry price of position
is_short: bool,
position: float, # Absolute value of position size
amount: float,
stake_amount: float,
wallet_balance: float, # Or margin balance
mm_ex_1: float = 0.0, # (Binance) Cross only
upnl_ex_1: float = 0.0, # (Binance) Cross only
) -> Optional[float]:
"""
Important: Must be fetching data from cached values as this is used by backtesting!
PERPETUAL:
gateio: https://www.gate.io/help/futures/perpetual/22160/calculation-of-liquidation-price
okex: https://www.okex.com/support/hc/en-us/articles/
360053909592-VI-Introduction-to-the-isolated-mode-of-Single-Multi-currency-Portfolio-margin
Important: Must be fetching data from cached values as this is used by backtesting!
:param exchange_name:
:param open_rate: Entry price of position
:param is_short: True if the trade is a short, false otherwise
:param position: Absolute value of position size incl. leverage (in base currency)
:param amount: Absolute value of position size incl. leverage (in base currency)
:param stake_amount: Stake amount - Collateral in settle currency.
:param trading_mode: SPOT, MARGIN, FUTURES, etc.
:param margin_mode: Either ISOLATED or CROSS
:param wallet_balance: Amount of margin_mode in the wallet being used to trade
@ -2633,7 +2710,7 @@ class Exchange:
market = self.markets[pair]
taker_fee_rate = market['taker']
mm_ratio, _ = self.get_maintenance_ratio_and_amt(pair, position)
mm_ratio, _ = self.get_maintenance_ratio_and_amt(pair, stake_amount)
if self.trading_mode == TradingMode.FUTURES and self.margin_mode == MarginMode.ISOLATED:
@ -2641,7 +2718,7 @@ class Exchange:
raise OperationalException(
"Freqtrade does not yet support inverse contracts")
value = wallet_balance / position
value = wallet_balance / amount
mm_ratio_taker = (mm_ratio + taker_fee_rate)
if is_short:
@ -2677,8 +2754,8 @@ class Exchange:
pair_tiers = self._leverage_tiers[pair]
for tier in reversed(pair_tiers):
if nominal_value >= tier['min']:
return (tier['mmr'], tier['maintAmt'])
if nominal_value >= tier['minNotional']:
return (tier['maintenanceMarginRate'], tier['maintAmt'])
raise OperationalException("nominal value can not be lower than 0")
# The lowest notional_floor for any pair in fetch_leverage_tiers is always 0 because it
@ -2818,3 +2895,111 @@ def market_is_active(market: Dict) -> bool:
# See https://github.com/ccxt/ccxt/issues/4874,
# https://github.com/ccxt/ccxt/issues/4075#issuecomment-434760520
return market.get('active', True) is not False
def amount_to_contracts(amount: float, contract_size: Optional[float]) -> float:
"""
Convert amount to contracts.
:param amount: amount to convert
:param contract_size: contract size - taken from exchange.get_contract_size(pair)
:return: num-contracts
"""
if contract_size and contract_size != 1:
return amount / contract_size
else:
return amount
def contracts_to_amount(num_contracts: float, contract_size: Optional[float]) -> float:
"""
Takes num-contracts and converts it to contract size
:param num_contracts: number of contracts
:param contract_size: contract size - taken from exchange.get_contract_size(pair)
:return: Amount
"""
if contract_size and contract_size != 1:
return num_contracts * contract_size
else:
return num_contracts
def amount_to_precision(amount: float, amount_precision: Optional[float],
precisionMode: Optional[int]) -> float:
"""
Returns the amount to buy or sell to a precision the Exchange accepts
Re-implementation of ccxt internal methods - ensuring we can test the result is correct
based on our definitions.
:param amount: amount to truncate
:param amount_precision: amount precision to use.
should be retrieved from markets[pair]['precision']['amount']
:param precisionMode: precision mode to use. Should be used from precisionMode
one of ccxt's DECIMAL_PLACES, SIGNIFICANT_DIGITS, or TICK_SIZE
:return: truncated amount
"""
if amount_precision is not None and precisionMode is not None:
precision = int(amount_precision) if precisionMode != TICK_SIZE else amount_precision
# precision must be an int for non-ticksize inputs.
amount = float(decimal_to_precision(amount, rounding_mode=TRUNCATE,
precision=precision,
counting_mode=precisionMode,
))
return amount
def amount_to_contract_precision(
amount, amount_precision: Optional[float], precisionMode: Optional[int],
contract_size: Optional[float]) -> float:
"""
Returns the amount to buy or sell to a precision the Exchange accepts
including calculation to and from contracts.
Re-implementation of ccxt internal methods - ensuring we can test the result is correct
based on our definitions.
:param amount: amount to truncate
:param amount_precision: amount precision to use.
should be retrieved from markets[pair]['precision']['amount']
:param precisionMode: precision mode to use. Should be used from precisionMode
one of ccxt's DECIMAL_PLACES, SIGNIFICANT_DIGITS, or TICK_SIZE
:param contract_size: contract size - taken from exchange.get_contract_size(pair)
:return: truncated amount
"""
if amount_precision is not None and precisionMode is not None:
contracts = amount_to_contracts(amount, contract_size)
amount_p = amount_to_precision(contracts, amount_precision, precisionMode)
return contracts_to_amount(amount_p, contract_size)
return amount
def price_to_precision(price: float, price_precision: Optional[float],
precisionMode: Optional[int]) -> float:
"""
Returns the price rounded up to the precision the Exchange accepts.
Partial Re-implementation of ccxt internal method decimal_to_precision(),
which does not support rounding up
TODO: If ccxt supports ROUND_UP for decimal_to_precision(), we could remove this and
align with amount_to_precision().
!!! Rounds up
:param price: price to convert
:param price_precision: price precision to use. Used from markets[pair]['precision']['price']
:param precisionMode: precision mode to use. Should be used from precisionMode
one of ccxt's DECIMAL_PLACES, SIGNIFICANT_DIGITS, or TICK_SIZE
:return: price rounded up to the precision the Exchange accepts
"""
if price_precision is not None and precisionMode is not None:
# price = float(decimal_to_precision(price, rounding_mode=ROUND,
# precision=price_precision,
# counting_mode=self.precisionMode,
# ))
if precisionMode == TICK_SIZE:
precision = FtPrecise(price_precision)
price_str = FtPrecise(price)
missing = price_str % precision
if not missing == FtPrecise("0"):
price = round(float(str(price_str - missing + precision)), 14)
else:
symbol_prec = price_precision
big_price = price * pow(10, symbol_prec)
price = ceil(big_price) / pow(10, symbol_prec)
return price

View File

@ -1,6 +1,6 @@
""" FTX exchange subclass """
import logging
from typing import Any, Dict, List, Tuple
from typing import Any, Dict, List, Optional, Tuple
import ccxt
@ -116,9 +116,17 @@ class Ftx(Exchange):
if len(order) == 1:
if order[0].get('status') == 'closed':
# Trigger order was triggered ...
real_order_id = order[0].get('info', {}).get('orderId')
real_order_id: Optional[str] = order[0].get('info', {}).get('orderId')
# OrderId may be None for stoploss-market orders
# But contains "average" in these cases.
# So we need to get it through the endpoint
# /conditional_orders/{conditional_order_id}/triggers
if not real_order_id:
res = self._api.privateGetConditionalOrdersConditionalOrderIdTriggers(
params={'conditional_order_id': order_id})
self._log_exchange_response('fetch_stoploss_order2', res)
real_order_id = res['result'][0]['orderId'] if res.get(
'result', []) else None
if real_order_id:
order1 = self._api.fetch_order(real_order_id, pair)
self._log_exchange_response('fetch_stoploss_order1', order1)

View File

@ -25,7 +25,6 @@ class Gateio(Exchange):
_ft_has: Dict = {
"ohlcv_candle_limit": 1000,
"ohlcv_volume_currency": "quote",
"time_in_force_parameter": "timeInForce",
"order_time_in_force": ['gtc', 'ioc'],
"stoploss_order_types": {"limit": "limit"},

View File

@ -7,9 +7,8 @@ from freqtrade.constants import BuySell
from freqtrade.enums import MarginMode, TradingMode
from freqtrade.enums.candletype import CandleType
from freqtrade.exceptions import DDosProtection, OperationalException, TemporaryError
from freqtrade.exchange import Exchange
from freqtrade.exchange import Exchange, date_minus_candles
from freqtrade.exchange.common import retrier
from freqtrade.exchange.exchange import date_minus_candles
logger = logging.getLogger(__name__)
@ -40,6 +39,8 @@ class Okx(Exchange):
net_only = True
_ccxt_params: Dict = {'options': {'brokerId': 'ffb5405ad327SUDE'}}
def ohlcv_candle_limit(
self, timeframe: str, candle_type: CandleType, since_ms: Optional[int] = None) -> int:
"""
@ -145,4 +146,4 @@ class Okx(Exchange):
return float('inf')
pair_tiers = self._leverage_tiers[pair]
return pair_tiers[-1]['max'] / leverage
return pair_tiers[-1]['maxNotional'] / leverage

View File

View File

@ -0,0 +1,608 @@
import collections
import json
import logging
import re
import shutil
import threading
from pathlib import Path
from typing import Any, Dict, Tuple, TypedDict
import numpy as np
import pandas as pd
import rapidjson
from joblib import dump, load
from joblib.externals import cloudpickle
from numpy.typing import NDArray
from pandas import DataFrame
from freqtrade.configuration import TimeRange
from freqtrade.data.history import load_pair_history
from freqtrade.exceptions import OperationalException
from freqtrade.freqai.data_kitchen import FreqaiDataKitchen
from freqtrade.strategy.interface import IStrategy
logger = logging.getLogger(__name__)
class pair_info(TypedDict):
model_filename: str
first: bool
trained_timestamp: int
priority: int
data_path: str
extras: dict
class FreqaiDataDrawer:
"""
Class aimed at holding all pair models/info in memory for better inferencing/retrainig/saving
/loading to/from disk.
This object remains persistent throughout live/dry.
Record of contribution:
FreqAI was developed by a group of individuals who all contributed specific skillsets to the
project.
Conception and software development:
Robert Caulk @robcaulk
Theoretical brainstorming:
Elin Törnquist @th0rntwig
Code review, software architecture brainstorming:
@xmatthias
Beta testing and bug reporting:
@bloodhunter4rc, Salah Lamkadem @ikonx, @ken11o2, @longyu, @paranoidandy, @smidelis, @smarm
Juha Nykänen @suikula, Wagner Costa @wagnercosta, Johan Vlugt @Jooopieeert
"""
def __init__(self, full_path: Path, config: dict, follow_mode: bool = False):
self.config = config
self.freqai_info = config.get("freqai", {})
# dictionary holding all pair metadata necessary to load in from disk
self.pair_dict: Dict[str, pair_info] = {}
# dictionary holding all actively inferenced models in memory given a model filename
self.model_dictionary: Dict[str, Any] = {}
self.model_return_values: Dict[str, DataFrame] = {}
self.historic_data: Dict[str, Dict[str, DataFrame]] = {}
self.historic_predictions: Dict[str, DataFrame] = {}
self.follower_dict: Dict[str, pair_info] = {}
self.full_path = full_path
self.follower_name: str = self.config.get("bot_name", "follower1")
self.follower_dict_path = Path(
self.full_path / f"follower_dictionary-{self.follower_name}.json"
)
self.historic_predictions_path = Path(self.full_path / "historic_predictions.pkl")
self.pair_dictionary_path = Path(self.full_path / "pair_dictionary.json")
self.follow_mode = follow_mode
if follow_mode:
self.create_follower_dict()
self.load_drawer_from_disk()
self.load_historic_predictions_from_disk()
self.training_queue: Dict[str, int] = {}
self.history_lock = threading.Lock()
self.save_lock = threading.Lock()
self.pair_dict_lock = threading.Lock()
self.old_DBSCAN_eps: Dict[str, float] = {}
self.empty_pair_dict: pair_info = {
"model_filename": "", "trained_timestamp": 0,
"priority": 1, "first": True, "data_path": "", "extras": {}}
def load_drawer_from_disk(self):
"""
Locate and load a previously saved data drawer full of all pair model metadata in
present model folder.
:return: bool - whether or not the drawer was located
"""
exists = self.pair_dictionary_path.is_file()
if exists:
with open(self.pair_dictionary_path, "r") as fp:
self.pair_dict = json.load(fp)
elif not self.follow_mode:
logger.info("Could not find existing datadrawer, starting from scratch")
else:
logger.warning(
f"Follower could not find pair_dictionary at {self.full_path} "
"sending null values back to strategy"
)
return exists
def load_historic_predictions_from_disk(self):
"""
Locate and load a previously saved historic predictions.
:return: bool - whether or not the drawer was located
"""
exists = self.historic_predictions_path.is_file()
if exists:
with open(self.historic_predictions_path, "rb") as fp:
self.historic_predictions = cloudpickle.load(fp)
logger.info(
f"Found existing historic predictions at {self.full_path}, but beware "
"that statistics may be inaccurate if the bot has been offline for "
"an extended period of time."
)
elif not self.follow_mode:
logger.info("Could not find existing historic_predictions, starting from scratch")
else:
logger.warning(
f"Follower could not find historic predictions at {self.full_path} "
"sending null values back to strategy"
)
return exists
def save_historic_predictions_to_disk(self):
"""
Save data drawer full of all pair model metadata in present model folder.
"""
with open(self.historic_predictions_path, "wb") as fp:
cloudpickle.dump(self.historic_predictions, fp, protocol=cloudpickle.DEFAULT_PROTOCOL)
def save_drawer_to_disk(self):
"""
Save data drawer full of all pair model metadata in present model folder.
"""
with self.save_lock:
with open(self.pair_dictionary_path, 'w') as fp:
rapidjson.dump(self.pair_dict, fp, default=self.np_encoder,
number_mode=rapidjson.NM_NATIVE)
def save_follower_dict_to_disk(self):
"""
Save follower dictionary to disk (used by strategy for persistent prediction targets)
"""
with open(self.follower_dict_path, "w") as fp:
rapidjson.dump(self.follower_dict, fp, default=self.np_encoder,
number_mode=rapidjson.NM_NATIVE)
def create_follower_dict(self):
"""
Create or dictionary for each follower to maintain unique persistent prediction targets
"""
whitelist_pairs = self.config.get("exchange", {}).get("pair_whitelist")
exists = self.follower_dict_path.is_file()
if exists:
logger.info("Found an existing follower dictionary")
for pair in whitelist_pairs:
self.follower_dict[pair] = {}
self.save_follower_dict_to_disk()
def np_encoder(self, object):
if isinstance(object, np.generic):
return object.item()
def get_pair_dict_info(self, pair: str) -> Tuple[str, int, bool]:
"""
Locate and load existing model metadata from persistent storage. If not located,
create a new one and append the current pair to it and prepare it for its first
training
:param pair: str: pair to lookup
:return:
model_filename: str = unique filename used for loading persistent objects from disk
trained_timestamp: int = the last time the coin was trained
return_null_array: bool = Follower could not find pair metadata
"""
pair_dict = self.pair_dict.get(pair)
data_path_set = self.pair_dict.get(pair, self.empty_pair_dict).get("data_path", "")
return_null_array = False
if pair_dict:
model_filename = pair_dict["model_filename"]
trained_timestamp = pair_dict["trained_timestamp"]
elif not self.follow_mode:
self.pair_dict[pair] = self.empty_pair_dict.copy()
model_filename = ""
trained_timestamp = 0
self.pair_dict[pair]["priority"] = len(self.pair_dict)
if not data_path_set and self.follow_mode:
logger.warning(
f"Follower could not find current pair {pair} in "
f"pair_dictionary at path {self.full_path}, sending null values "
"back to strategy."
)
trained_timestamp = 0
model_filename = ''
return_null_array = True
return model_filename, trained_timestamp, return_null_array
def set_pair_dict_info(self, metadata: dict) -> None:
pair_in_dict = self.pair_dict.get(metadata["pair"])
if pair_in_dict:
return
else:
self.pair_dict[metadata["pair"]] = self.empty_pair_dict.copy()
self.pair_dict[metadata["pair"]]["priority"] = len(self.pair_dict)
return
def pair_to_end_of_training_queue(self, pair: str) -> None:
# march all pairs up in the queue
with self.pair_dict_lock:
for p in self.pair_dict:
self.pair_dict[p]["priority"] -= 1
# send pair to end of queue
self.pair_dict[pair]["priority"] = len(self.pair_dict)
def set_initial_return_values(self, pair: str, pred_df: DataFrame) -> None:
"""
Set the initial return values to the historical predictions dataframe. This avoids needing
to repredict on historical candles, and also stores historical predictions despite
retrainings (so stored predictions are true predictions, not just inferencing on trained
data)
"""
hist_df = self.historic_predictions
len_diff = len(hist_df[pair].index) - len(pred_df.index)
if len_diff < 0:
df_concat = pd.concat([pred_df.iloc[:abs(len_diff)], hist_df[pair]],
ignore_index=True, keys=hist_df[pair].keys())
else:
df_concat = hist_df[pair].tail(len(pred_df.index)).reset_index(drop=True)
df_concat = df_concat.fillna(0)
self.model_return_values[pair] = df_concat
def append_model_predictions(self, pair: str, predictions: DataFrame,
do_preds: NDArray[np.int_],
dk: FreqaiDataKitchen, len_df: int) -> None:
"""
Append model predictions to historic predictions dataframe, then set the
strategy return dataframe to the tail of the historic predictions. The length of
the tail is equivalent to the length of the dataframe that entered FreqAI from
the strategy originally. Doing this allows FreqUI to always display the correct
historic predictions.
"""
index = self.historic_predictions[pair].index[-1:]
columns = self.historic_predictions[pair].columns
nan_df = pd.DataFrame(np.nan, index=index, columns=columns)
self.historic_predictions[pair] = pd.concat(
[self.historic_predictions[pair], nan_df], ignore_index=True, axis=0)
df = self.historic_predictions[pair]
# model outputs and associated statistics
for label in predictions.columns:
df[label].iloc[-1] = predictions[label].iloc[-1]
if df[label].dtype == object:
continue
df[f"{label}_mean"].iloc[-1] = dk.data["labels_mean"][label]
df[f"{label}_std"].iloc[-1] = dk.data["labels_std"][label]
# outlier indicators
df["do_predict"].iloc[-1] = do_preds[-1]
if self.freqai_info["feature_parameters"].get("DI_threshold", 0) > 0:
df["DI_values"].iloc[-1] = dk.DI_values[-1]
# extra values the user added within custom prediction model
if dk.data['extra_returns_per_train']:
rets = dk.data['extra_returns_per_train']
for return_str in rets:
df[return_str].iloc[-1] = rets[return_str]
self.model_return_values[pair] = df.tail(len_df).reset_index(drop=True)
def attach_return_values_to_return_dataframe(
self, pair: str, dataframe: DataFrame) -> DataFrame:
"""
Attach the return values to the strat dataframe
:param dataframe: DataFrame = strategy dataframe
:return: DataFrame = strat dataframe with return values attached
"""
df = self.model_return_values[pair]
to_keep = [col for col in dataframe.columns if not col.startswith("&")]
dataframe = pd.concat([dataframe[to_keep], df], axis=1)
return dataframe
def return_null_values_to_strategy(self, dataframe: DataFrame, dk: FreqaiDataKitchen) -> None:
"""
Build 0 filled dataframe to return to strategy
"""
dk.find_features(dataframe)
full_labels = dk.label_list + dk.unique_class_list
for label in full_labels:
dataframe[label] = 0
dataframe[f"{label}_mean"] = 0
dataframe[f"{label}_std"] = 0
dataframe["do_predict"] = 0
if self.freqai_info["feature_parameters"].get("DI_threshold", 0) > 0:
dataframe["DI_values"] = 0
if dk.data['extra_returns_per_train']:
rets = dk.data['extra_returns_per_train']
for return_str in rets:
dataframe[return_str] = 0
dk.return_dataframe = dataframe
def purge_old_models(self) -> None:
model_folders = [x for x in self.full_path.iterdir() if x.is_dir()]
pattern = re.compile(r"sub-train-(\w+)_(\d{10})")
delete_dict: Dict[str, Any] = {}
for dir in model_folders:
result = pattern.match(str(dir.name))
if result is None:
break
coin = result.group(1)
timestamp = result.group(2)
if coin not in delete_dict:
delete_dict[coin] = {}
delete_dict[coin]["num_folders"] = 1
delete_dict[coin]["timestamps"] = {int(timestamp): dir}
else:
delete_dict[coin]["num_folders"] += 1
delete_dict[coin]["timestamps"][int(timestamp)] = dir
for coin in delete_dict:
if delete_dict[coin]["num_folders"] > 2:
sorted_dict = collections.OrderedDict(
sorted(delete_dict[coin]["timestamps"].items())
)
num_delete = len(sorted_dict) - 2
deleted = 0
for k, v in sorted_dict.items():
if deleted >= num_delete:
break
logger.info(f"Freqai purging old model file {v}")
shutil.rmtree(v)
deleted += 1
def update_follower_metadata(self):
# follower needs to load from disk to get any changes made by leader to pair_dict
self.load_drawer_from_disk()
if self.config.get("freqai", {}).get("purge_old_models", False):
self.purge_old_models()
# Functions pulled back from FreqaiDataKitchen because they relied on DataDrawer
def save_data(self, model: Any, coin: str, dk: FreqaiDataKitchen) -> None:
"""
Saves all data associated with a model for a single sub-train time range
:params:
:model: User trained model which can be reused for inferencing to generate
predictions
"""
if not dk.data_path.is_dir():
dk.data_path.mkdir(parents=True, exist_ok=True)
save_path = Path(dk.data_path)
# Save the trained model
if not dk.keras:
dump(model, save_path / f"{dk.model_filename}_model.joblib")
else:
model.save(save_path / f"{dk.model_filename}_model.h5")
if dk.svm_model is not None:
dump(dk.svm_model, save_path / f"{dk.model_filename}_svm_model.joblib")
dk.data["data_path"] = str(dk.data_path)
dk.data["model_filename"] = str(dk.model_filename)
dk.data["training_features_list"] = list(dk.data_dictionary["train_features"].columns)
dk.data["label_list"] = dk.label_list
# store the metadata
with open(save_path / f"{dk.model_filename}_metadata.json", "w") as fp:
rapidjson.dump(dk.data, fp, default=self.np_encoder, number_mode=rapidjson.NM_NATIVE)
# save the train data to file so we can check preds for area of applicability later
dk.data_dictionary["train_features"].to_pickle(
save_path / f"{dk.model_filename}_trained_df.pkl"
)
dk.data_dictionary["train_dates"].to_pickle(
save_path / f"{dk.model_filename}_trained_dates_df.pkl"
)
if self.freqai_info["feature_parameters"].get("principal_component_analysis"):
cloudpickle.dump(
dk.pca, open(dk.data_path / f"{dk.model_filename}_pca_object.pkl", "wb")
)
# if self.live:
self.model_dictionary[coin] = model
self.pair_dict[coin]["model_filename"] = dk.model_filename
self.pair_dict[coin]["data_path"] = str(dk.data_path)
self.save_drawer_to_disk()
return
def load_data(self, coin: str, dk: FreqaiDataKitchen) -> Any:
"""
loads all data required to make a prediction on a sub-train time range
:returns:
:model: User trained model which can be inferenced for new predictions
"""
if not self.pair_dict[coin]["model_filename"]:
return None
if dk.live:
dk.model_filename = self.pair_dict[coin]["model_filename"]
dk.data_path = Path(self.pair_dict[coin]["data_path"])
if self.freqai_info.get("follow_mode", False):
# follower can be on a different system which is rsynced from the leader:
dk.data_path = Path(
self.config["user_data_dir"]
/ "models"
/ dk.data_path.parts[-2]
/ dk.data_path.parts[-1]
)
with open(dk.data_path / f"{dk.model_filename}_metadata.json", "r") as fp:
dk.data = json.load(fp)
dk.training_features_list = dk.data["training_features_list"]
dk.label_list = dk.data["label_list"]
dk.data_dictionary["train_features"] = pd.read_pickle(
dk.data_path / f"{dk.model_filename}_trained_df.pkl"
)
# try to access model in memory instead of loading object from disk to save time
if dk.live and coin in self.model_dictionary:
model = self.model_dictionary[coin]
elif not dk.keras:
model = load(dk.data_path / f"{dk.model_filename}_model.joblib")
else:
from tensorflow import keras
model = keras.models.load_model(dk.data_path / f"{dk.model_filename}_model.h5")
if Path(dk.data_path / f"{dk.model_filename}_svm_model.joblib").is_file():
dk.svm_model = load(dk.data_path / f"{dk.model_filename}_svm_model.joblib")
if not model:
raise OperationalException(
f"Unable to load model, ensure model exists at " f"{dk.data_path} "
)
if self.config["freqai"]["feature_parameters"]["principal_component_analysis"]:
dk.pca = cloudpickle.load(
open(dk.data_path / f"{dk.model_filename}_pca_object.pkl", "rb")
)
return model
def update_historic_data(self, strategy: IStrategy, dk: FreqaiDataKitchen) -> None:
"""
Append new candles to our stores historic data (in memory) so that
we do not need to load candle history from disk and we dont need to
pinging exchange multiple times for the same candle.
:params:
dataframe: DataFrame = strategy provided dataframe
"""
feat_params = self.freqai_info["feature_parameters"]
with self.history_lock:
history_data = self.historic_data
for pair in dk.all_pairs:
for tf in feat_params.get("include_timeframes"):
# check if newest candle is already appended
df_dp = strategy.dp.get_pair_dataframe(pair, tf)
if len(df_dp.index) == 0:
continue
if str(history_data[pair][tf].iloc[-1]["date"]) == str(
df_dp.iloc[-1:]["date"].iloc[-1]
):
continue
try:
index = (
df_dp.loc[
df_dp["date"] == history_data[pair][tf].iloc[-1]["date"]
].index[0]
+ 1
)
except IndexError:
logger.warning(
f"Unable to update pair history for {pair}. "
"If this does not resolve itself after 1 additional candle, "
"please report the error to #freqai discord channel"
)
return
history_data[pair][tf] = pd.concat(
[
history_data[pair][tf],
df_dp.iloc[index:],
],
ignore_index=True,
axis=0,
)
def load_all_pair_histories(self, timerange: TimeRange, dk: FreqaiDataKitchen) -> None:
"""
Load pair histories for all whitelist and corr_pairlist pairs.
Only called once upon startup of bot.
:params:
timerange: TimeRange = full timerange required to populate all indicators
for training according to user defined train_period_days
"""
history_data = self.historic_data
for pair in dk.all_pairs:
if pair not in history_data:
history_data[pair] = {}
for tf in self.freqai_info["feature_parameters"].get("include_timeframes"):
history_data[pair][tf] = load_pair_history(
datadir=self.config["datadir"],
timeframe=tf,
pair=pair,
timerange=timerange,
data_format=self.config.get("dataformat_ohlcv", "json"),
candle_type=self.config.get("trading_mode", "spot"),
)
def get_base_and_corr_dataframes(
self, timerange: TimeRange, pair: str, dk: FreqaiDataKitchen
) -> Tuple[Dict[Any, Any], Dict[Any, Any]]:
"""
Searches through our historic_data in memory and returns the dataframes relevant
to the present pair.
:params:
timerange: TimeRange = full timerange required to populate all indicators
for training according to user defined train_period_days
metadata: dict = strategy furnished pair metadata
"""
with self.history_lock:
corr_dataframes: Dict[Any, Any] = {}
base_dataframes: Dict[Any, Any] = {}
historic_data = self.historic_data
pairs = self.freqai_info["feature_parameters"].get(
"include_corr_pairlist", []
)
for tf in self.freqai_info["feature_parameters"].get("include_timeframes"):
base_dataframes[tf] = dk.slice_dataframe(timerange, historic_data[pair][tf])
if pairs:
for p in pairs:
if pair in p:
continue # dont repeat anything from whitelist
if p not in corr_dataframes:
corr_dataframes[p] = {}
corr_dataframes[p][tf] = dk.slice_dataframe(
timerange, historic_data[p][tf]
)
return corr_dataframes, base_dataframes
# to be used if we want to send predictions directly to the follower instead of forcing
# follower to load models and inference
# def save_model_return_values_to_disk(self) -> None:
# with open(self.full_path / str('model_return_values.json'), "w") as fp:
# json.dump(self.model_return_values, fp, default=self.np_encoder)
# def load_model_return_values_from_disk(self, dk: FreqaiDataKitchen) -> FreqaiDataKitchen:
# exists = Path(self.full_path / str('model_return_values.json')).resolve().exists()
# if exists:
# with open(self.full_path / str('model_return_values.json'), "r") as fp:
# self.model_return_values = json.load(fp)
# elif not self.follow_mode:
# logger.info("Could not find existing datadrawer, starting from scratch")
# else:
# logger.warning(f'Follower could not find pair_dictionary at {self.full_path} '
# 'sending null values back to strategy')
# return exists, dk

File diff suppressed because it is too large Load Diff

View File

@ -0,0 +1,684 @@
# import contextlib
import datetime
import logging
import shutil
import threading
import time
from abc import ABC, abstractmethod
from pathlib import Path
from threading import Lock
from typing import Any, Dict, Tuple
import numpy as np
import pandas as pd
from numpy.typing import NDArray
from pandas import DataFrame
from freqtrade.configuration import TimeRange
from freqtrade.enums import RunMode
from freqtrade.exceptions import OperationalException
from freqtrade.exchange import timeframe_to_seconds
from freqtrade.freqai.data_drawer import FreqaiDataDrawer
from freqtrade.freqai.data_kitchen import FreqaiDataKitchen
from freqtrade.strategy.interface import IStrategy
pd.options.mode.chained_assignment = None
logger = logging.getLogger(__name__)
def threaded(fn):
def wrapper(*args, **kwargs):
threading.Thread(target=fn, args=args, kwargs=kwargs).start()
return wrapper
class IFreqaiModel(ABC):
"""
Class containing all tools for training and prediction in the strategy.
Base*PredictionModels inherit from this class.
Record of contribution:
FreqAI was developed by a group of individuals who all contributed specific skillsets to the
project.
Conception and software development:
Robert Caulk @robcaulk
Theoretical brainstorming:
Elin Törnquist @th0rntwig
Code review, software architecture brainstorming:
@xmatthias
Beta testing and bug reporting:
@bloodhunter4rc, Salah Lamkadem @ikonx, @ken11o2, @longyu, @paranoidandy, @smidelis, @smarm
Juha Nykänen @suikula, Wagner Costa @wagnercosta, Johan Vlugt @Jooopieeert
"""
def __init__(self, config: Dict[str, Any]) -> None:
self.config = config
self.assert_config(self.config)
self.freqai_info: Dict[str, Any] = config["freqai"]
self.data_split_parameters: Dict[str, Any] = config.get("freqai", {}).get(
"data_split_parameters", {})
self.model_training_parameters: Dict[str, Any] = config.get("freqai", {}).get(
"model_training_parameters", {})
self.feature_parameters = config.get("freqai", {}).get("feature_parameters")
self.retrain = False
self.first = True
self.set_full_path()
self.follow_mode: bool = self.freqai_info.get("follow_mode", False)
self.dd = FreqaiDataDrawer(Path(self.full_path), self.config, self.follow_mode)
self.identifier: str = self.freqai_info.get("identifier", "no_id_provided")
self.scanning = False
self.keras: bool = self.freqai_info.get("keras", False)
if self.keras and self.freqai_info.get("feature_parameters", {}).get("DI_threshold", 0):
self.freqai_info["feature_parameters"]["DI_threshold"] = 0
logger.warning("DI threshold is not configured for Keras models yet. Deactivating.")
self.CONV_WIDTH = self.freqai_info.get("conv_width", 2)
self.pair_it = 0
self.pair_it_train = 0
self.total_pairs = len(self.config.get("exchange", {}).get("pair_whitelist"))
self.last_trade_database_summary: DataFrame = {}
self.current_trade_database_summary: DataFrame = {}
self.analysis_lock = Lock()
self.inference_time: float = 0
self.train_time: float = 0
self.begin_time: float = 0
self.begin_time_train: float = 0
self.base_tf_seconds = timeframe_to_seconds(self.config['timeframe'])
def assert_config(self, config: Dict[str, Any]) -> None:
if not config.get("freqai", {}):
raise OperationalException("No freqai parameters found in configuration file.")
def start(self, dataframe: DataFrame, metadata: dict, strategy: IStrategy) -> DataFrame:
"""
Entry point to the FreqaiModel from a specific pair, it will train a new model if
necessary before making the prediction.
:param dataframe: Full dataframe coming from strategy - it contains entire
backtesting timerange + additional historical data necessary to train
the model.
:param metadata: pair metadata coming from strategy.
:param strategy: Strategy to train on
"""
self.live = strategy.dp.runmode in (RunMode.DRY_RUN, RunMode.LIVE)
self.dd.set_pair_dict_info(metadata)
if self.live:
self.inference_timer('start')
self.dk = FreqaiDataKitchen(self.config, self.live, metadata["pair"])
dk = self.start_live(dataframe, metadata, strategy, self.dk)
# For backtesting, each pair enters and then gets trained for each window along the
# sliding window defined by "train_period_days" (training window) and "live_retrain_hours"
# (backtest window, i.e. window immediately following the training window).
# FreqAI slides the window and sequentially builds the backtesting results before returning
# the concatenated results for the full backtesting period back to the strategy.
elif not self.follow_mode:
self.dk = FreqaiDataKitchen(self.config, self.live, metadata["pair"])
logger.info(f"Training {len(self.dk.training_timeranges)} timeranges")
with self.analysis_lock:
dataframe = self.dk.use_strategy_to_populate_indicators(
strategy, prediction_dataframe=dataframe, pair=metadata["pair"]
)
dk = self.start_backtesting(dataframe, metadata, self.dk)
dataframe = dk.remove_features_from_df(dk.return_dataframe)
self.clean_up()
if self.live:
self.inference_timer('stop')
return dataframe
def clean_up(self):
"""
Objects that should be handled by GC already between coins, but
are explicitly shown here to help demonstrate the non-persistence of these
objects.
"""
self.model = None
self.dk = None
@threaded
def start_scanning(self, strategy: IStrategy) -> None:
"""
Function designed to constantly scan pairs for retraining on a separate thread (intracandle)
to improve model youth. This function is agnostic to data preparation/collection/storage,
it simply trains on what ever data is available in the self.dd.
:param strategy: IStrategy = The user defined strategy class
"""
while 1:
time.sleep(1)
for pair in self.config.get("exchange", {}).get("pair_whitelist"):
(_, trained_timestamp, _) = self.dd.get_pair_dict_info(pair)
if self.dd.pair_dict[pair]["priority"] != 1:
continue
dk = FreqaiDataKitchen(self.config, self.live, pair)
dk.set_paths(pair, trained_timestamp)
(
retrain,
new_trained_timerange,
data_load_timerange,
) = dk.check_if_new_training_required(trained_timestamp)
dk.set_paths(pair, new_trained_timerange.stopts)
if retrain:
self.train_timer('start')
self.train_model_in_series(
new_trained_timerange, pair, strategy, dk, data_load_timerange
)
self.train_timer('stop')
self.dd.save_historic_predictions_to_disk()
def start_backtesting(
self, dataframe: DataFrame, metadata: dict, dk: FreqaiDataKitchen
) -> FreqaiDataKitchen:
"""
The main broad execution for backtesting. For backtesting, each pair enters and then gets
trained for each window along the sliding window defined by "train_period_days"
(training window) and "backtest_period_days" (backtest window, i.e. window immediately
following the training window). FreqAI slides the window and sequentially builds
the backtesting results before returning the concatenated results for the full
backtesting period back to the strategy.
:param dataframe: DataFrame = strategy passed dataframe
:param metadata: Dict = pair metadata
:param dk: FreqaiDataKitchen = Data management/analysis tool associated to present pair only
:return:
FreqaiDataKitchen = Data management/analysis tool associated to present pair only
"""
self.pair_it += 1
train_it = 0
# Loop enforcing the sliding window training/backtesting paradigm
# tr_train is the training time range e.g. 1 historical month
# tr_backtest is the backtesting time range e.g. the week directly
# following tr_train. Both of these windows slide through the
# entire backtest
for tr_train, tr_backtest in zip(dk.training_timeranges, dk.backtesting_timeranges):
(_, _, _) = self.dd.get_pair_dict_info(metadata["pair"])
train_it += 1
total_trains = len(dk.backtesting_timeranges)
self.training_timerange = tr_train
dataframe_train = dk.slice_dataframe(tr_train, dataframe)
dataframe_backtest = dk.slice_dataframe(tr_backtest, dataframe)
trained_timestamp = tr_train
tr_train_startts_str = datetime.datetime.utcfromtimestamp(tr_train.startts).strftime(
"%Y-%m-%d %H:%M:%S"
)
tr_train_stopts_str = datetime.datetime.utcfromtimestamp(tr_train.stopts).strftime(
"%Y-%m-%d %H:%M:%S"
)
logger.info(
f"Training {metadata['pair']}, {self.pair_it}/{self.total_pairs} pairs"
f" from {tr_train_startts_str} to {tr_train_stopts_str}, {train_it}/{total_trains} "
"trains"
)
dk.data_path = Path(
dk.full_path
/
f"sub-train-{metadata['pair'].split('/')[0]}_{int(trained_timestamp.stopts)}"
)
if not self.model_exists(
metadata["pair"], dk, trained_timestamp=int(trained_timestamp.stopts)
):
dk.find_features(dataframe_train)
self.model = self.train(dataframe_train, metadata["pair"], dk)
self.dd.pair_dict[metadata["pair"]]["trained_timestamp"] = int(
trained_timestamp.stopts)
dk.set_new_model_names(metadata["pair"], trained_timestamp)
self.dd.save_data(self.model, metadata["pair"], dk)
else:
self.model = self.dd.load_data(metadata["pair"], dk)
self.check_if_feature_list_matches_strategy(dataframe_train, dk)
pred_df, do_preds = self.predict(dataframe_backtest, dk)
dk.append_predictions(pred_df, do_preds)
dk.fill_predictions(dataframe)
return dk
def start_live(
self, dataframe: DataFrame, metadata: dict, strategy: IStrategy, dk: FreqaiDataKitchen
) -> FreqaiDataKitchen:
"""
The main broad execution for dry/live. This function will check if a retraining should be
performed, and if so, retrain and reset the model.
:param dataframe: DataFrame = strategy passed dataframe
:param metadata: Dict = pair metadata
:param strategy: IStrategy = currently employed strategy
dk: FreqaiDataKitchen = Data management/analysis tool associated to present pair only
:returns:
dk: FreqaiDataKitchen = Data management/analysis tool associated to present pair only
"""
# update follower
if self.follow_mode:
self.dd.update_follower_metadata()
# get the model metadata associated with the current pair
(_, trained_timestamp, return_null_array) = self.dd.get_pair_dict_info(metadata["pair"])
# if the metadata doesn't exist, the follower returns null arrays to strategy
if self.follow_mode and return_null_array:
logger.info("Returning null array from follower to strategy")
self.dd.return_null_values_to_strategy(dataframe, dk)
return dk
# append the historic data once per round
if self.dd.historic_data:
self.dd.update_historic_data(strategy, dk)
logger.debug(f'Updating historic data on pair {metadata["pair"]}')
if not self.follow_mode:
(_, new_trained_timerange, data_load_timerange) = dk.check_if_new_training_required(
trained_timestamp
)
dk.set_paths(metadata["pair"], new_trained_timerange.stopts)
# download candle history if it is not already in memory
if not self.dd.historic_data:
logger.info(
"Downloading all training data for all pairs in whitelist and "
"corr_pairlist, this may take a while if you do not have the "
"data saved"
)
dk.download_all_data_for_training(data_load_timerange, strategy.dp)
self.dd.load_all_pair_histories(data_load_timerange, dk)
if not self.scanning:
self.scanning = True
self.start_scanning(strategy)
elif self.follow_mode:
dk.set_paths(metadata["pair"], trained_timestamp)
logger.info(
"FreqAI instance set to follow_mode, finding existing pair "
f"using { self.identifier }"
)
# load the model and associated data into the data kitchen
self.model = self.dd.load_data(metadata["pair"], dk)
with self.analysis_lock:
dataframe = self.dk.use_strategy_to_populate_indicators(
strategy, prediction_dataframe=dataframe, pair=metadata["pair"]
)
if not self.model:
logger.warning(
f"No model ready for {metadata['pair']}, returning null values to strategy."
)
self.dd.return_null_values_to_strategy(dataframe, dk)
return dk
# ensure user is feeding the correct indicators to the model
self.check_if_feature_list_matches_strategy(dataframe, dk)
self.build_strategy_return_arrays(dataframe, dk, metadata["pair"], trained_timestamp)
return dk
def build_strategy_return_arrays(
self, dataframe: DataFrame, dk: FreqaiDataKitchen, pair: str, trained_timestamp: int
) -> None:
# hold the historical predictions in memory so we are sending back
# correct array to strategy
if pair not in self.dd.model_return_values:
# first predictions are made on entire historical candle set coming from strategy. This
# allows FreqUI to show full return values.
pred_df, do_preds = self.predict(dataframe, dk)
if pair not in self.dd.historic_predictions:
self.set_initial_historic_predictions(pred_df, dk, pair)
self.dd.set_initial_return_values(pair, pred_df)
dk.return_dataframe = self.dd.attach_return_values_to_return_dataframe(pair, dataframe)
return
elif self.dk.check_if_model_expired(trained_timestamp):
pred_df = DataFrame(np.zeros((2, len(dk.label_list))), columns=dk.label_list)
do_preds = np.ones(2, dtype=np.int_) * 2
dk.DI_values = np.zeros(2)
logger.warning(
f"Model expired for {pair}, returning null values to strategy. Strategy "
"construction should take care to consider this event with "
"prediction == 0 and do_predict == 2"
)
else:
# remaining predictions are made only on the most recent candles for performance and
# historical accuracy reasons.
pred_df, do_preds = self.predict(dataframe.iloc[-self.CONV_WIDTH:], dk, first=False)
if self.freqai_info.get('fit_live_predictions_candles', 0) and self.live:
self.fit_live_predictions(dk, pair)
self.dd.append_model_predictions(pair, pred_df, do_preds, dk, len(dataframe))
dk.return_dataframe = self.dd.attach_return_values_to_return_dataframe(pair, dataframe)
return
def check_if_feature_list_matches_strategy(
self, dataframe: DataFrame, dk: FreqaiDataKitchen
) -> None:
"""
Ensure user is passing the proper feature set if they are reusing an `identifier` pointing
to a folder holding existing models.
:param dataframe: DataFrame = strategy provided dataframe
:param dk: FreqaiDataKitchen = non-persistent data container/analyzer for
current coin/bot loop
"""
dk.find_features(dataframe)
if "training_features_list_raw" in dk.data:
feature_list = dk.data["training_features_list_raw"]
else:
feature_list = dk.training_features_list
if dk.training_features_list != feature_list:
raise OperationalException(
"Trying to access pretrained model with `identifier` "
"but found different features furnished by current strategy."
"Change `identifier` to train from scratch, or ensure the"
"strategy is furnishing the same features as the pretrained"
"model"
)
def data_cleaning_train(self, dk: FreqaiDataKitchen) -> None:
"""
Base data cleaning method for train
Any function inside this method should drop training data points from the filtered_dataframe
based on user decided logic. See FreqaiDataKitchen::use_SVM_to_remove_outliers() for an
example of how outlier data points are dropped from the dataframe used for training.
"""
if self.freqai_info["feature_parameters"].get(
"principal_component_analysis", False
):
dk.principal_component_analysis()
if self.freqai_info["feature_parameters"].get("use_SVM_to_remove_outliers", False):
dk.use_SVM_to_remove_outliers(predict=False)
if self.freqai_info["feature_parameters"].get("DI_threshold", 0):
dk.data["avg_mean_dist"] = dk.compute_distances()
if self.freqai_info["feature_parameters"].get("use_DBSCAN_to_remove_outliers", False):
if dk.pair in self.dd.old_DBSCAN_eps:
eps = self.dd.old_DBSCAN_eps[dk.pair]
else:
eps = None
dk.use_DBSCAN_to_remove_outliers(predict=False, eps=eps)
self.dd.old_DBSCAN_eps[dk.pair] = dk.data['DBSCAN_eps']
def data_cleaning_predict(self, dk: FreqaiDataKitchen, dataframe: DataFrame) -> None:
"""
Base data cleaning method for predict.
These functions each modify dk.do_predict, which is a dataframe with equal length
to the number of candles coming from and returning to the strategy. Inside do_predict,
1 allows prediction and < 0 signals to the strategy that the model is not confident in
the prediction.
See FreqaiDataKitchen::remove_outliers() for an example
of how the do_predict vector is modified. do_predict is ultimately passed back to strategy
for buy signals.
"""
if self.freqai_info["feature_parameters"].get(
"principal_component_analysis", False
):
dk.pca_transform(dataframe)
if self.freqai_info["feature_parameters"].get("use_SVM_to_remove_outliers", False):
dk.use_SVM_to_remove_outliers(predict=True)
if self.freqai_info["feature_parameters"].get("DI_threshold", 0):
dk.check_if_pred_in_training_spaces()
if self.freqai_info["feature_parameters"].get("use_DBSCAN_to_remove_outliers", False):
dk.use_DBSCAN_to_remove_outliers(predict=True)
def model_exists(
self,
pair: str,
dk: FreqaiDataKitchen,
trained_timestamp: int = None,
model_filename: str = "",
scanning: bool = False,
) -> bool:
"""
Given a pair and path, check if a model already exists
:param pair: pair e.g. BTC/USD
:param path: path to model
:return:
:boolean: whether the model file exists or not.
"""
coin, _ = pair.split("/")
if not self.live:
dk.model_filename = model_filename = f"cb_{coin.lower()}_{trained_timestamp}"
path_to_modelfile = Path(dk.data_path / f"{model_filename}_model.joblib")
file_exists = path_to_modelfile.is_file()
if file_exists and not scanning:
logger.info("Found model at %s", dk.data_path / dk.model_filename)
elif not scanning:
logger.info("Could not find model at %s", dk.data_path / dk.model_filename)
return file_exists
def set_full_path(self) -> None:
self.full_path = Path(
self.config["user_data_dir"] / "models" / f"{self.freqai_info['identifier']}"
)
self.full_path.mkdir(parents=True, exist_ok=True)
shutil.copy(
self.config["config_files"][0],
Path(self.full_path, Path(self.config["config_files"][0]).name),
)
def train_model_in_series(
self,
new_trained_timerange: TimeRange,
pair: str,
strategy: IStrategy,
dk: FreqaiDataKitchen,
data_load_timerange: TimeRange,
):
"""
Retrieve data and train model.
:param new_trained_timerange: TimeRange = the timerange to train the model on
:param metadata: dict = strategy provided metadata
:param strategy: IStrategy = user defined strategy object
:param dk: FreqaiDataKitchen = non-persistent data container for current coin/loop
:param data_load_timerange: TimeRange = the amount of data to be loaded
for populate_any_indicators
(larger than new_trained_timerange so that
new_trained_timerange does not contain any NaNs)
"""
corr_dataframes, base_dataframes = self.dd.get_base_and_corr_dataframes(
data_load_timerange, pair, dk
)
with self.analysis_lock:
unfiltered_dataframe = dk.use_strategy_to_populate_indicators(
strategy, corr_dataframes, base_dataframes, pair
)
unfiltered_dataframe = dk.slice_dataframe(new_trained_timerange, unfiltered_dataframe)
# find the features indicated by strategy and store in datakitchen
dk.find_features(unfiltered_dataframe)
model = self.train(unfiltered_dataframe, pair, dk)
self.dd.pair_dict[pair]["trained_timestamp"] = new_trained_timerange.stopts
dk.set_new_model_names(pair, new_trained_timerange)
self.dd.pair_dict[pair]["first"] = False
if self.dd.pair_dict[pair]["priority"] == 1 and self.scanning:
self.dd.pair_to_end_of_training_queue(pair)
self.dd.save_data(model, pair, dk)
if self.freqai_info.get("purge_old_models", False):
self.dd.purge_old_models()
def set_initial_historic_predictions(
self, pred_df: DataFrame, dk: FreqaiDataKitchen, pair: str
) -> None:
"""
This function is called only if the datadrawer failed to load an
existing set of historic predictions. In this case, it builds
the structure and sets fake predictions off the first training
data. After that, FreqAI will append new real predictions to the
set of historic predictions.
These values are used to generate live statistics which can be used
in the strategy for adaptive values. E.g. &*_mean/std are quantities
that can computed based on live predictions from the set of historical
predictions. Those values can be used in the user strategy to better
assess prediction rarity, and thus wait for probabilistically favorable
entries relative to the live historical predictions.
If the user reuses an identifier on a subsequent instance,
this function will not be called. In that case, "real" predictions
will be appended to the loaded set of historic predictions.
:param: df: DataFrame = the dataframe containing the training feature data
:param: model: Any = A model which was `fit` using a common library such as
catboost or lightgbm
:param: dk: FreqaiDataKitchen = object containing methods for data analysis
:param: pair: str = current pair
"""
self.dd.historic_predictions[pair] = pred_df
hist_preds_df = self.dd.historic_predictions[pair]
for label in hist_preds_df.columns:
if hist_preds_df[label].dtype == object:
continue
hist_preds_df[f'{label}_mean'] = 0
hist_preds_df[f'{label}_std'] = 0
hist_preds_df['do_predict'] = 0
if self.freqai_info['feature_parameters'].get('DI_threshold', 0) > 0:
hist_preds_df['DI_values'] = 0
for return_str in dk.data['extra_returns_per_train']:
hist_preds_df[return_str] = 0
# # for keras type models, the conv_window needs to be prepended so
# # viewing is correct in frequi
if self.freqai_info.get('keras', False):
n_lost_points = self.freqai_info.get('conv_width', 2)
zeros_df = DataFrame(np.zeros((n_lost_points, len(hist_preds_df.columns))),
columns=hist_preds_df.columns)
self.dd.historic_predictions[pair] = pd.concat(
[zeros_df, hist_preds_df], axis=0, ignore_index=True)
def fit_live_predictions(self, dk: FreqaiDataKitchen, pair: str) -> None:
"""
Fit the labels with a gaussian distribution
"""
import scipy as spy
# add classes from classifier label types if used
full_labels = dk.label_list + dk.unique_class_list
num_candles = self.freqai_info.get("fit_live_predictions_candles", 100)
dk.data["labels_mean"], dk.data["labels_std"] = {}, {}
for label in full_labels:
if self.dd.historic_predictions[dk.pair][label].dtype == object:
continue
f = spy.stats.norm.fit(self.dd.historic_predictions[dk.pair][label].tail(num_candles))
dk.data["labels_mean"][label], dk.data["labels_std"][label] = f[0], f[1]
return
def inference_timer(self, do='start'):
"""
Timer designed to track the cumulative time spent in FreqAI for one pass through
the whitelist. This will check if the time spent is more than 1/4 the time
of a single candle, and if so, it will warn the user of degraded performance
"""
if do == 'start':
self.pair_it += 1
self.begin_time = time.time()
elif do == 'stop':
end = time.time()
self.inference_time += (end - self.begin_time)
if self.pair_it == self.total_pairs:
logger.info(
f'Total time spent inferencing pairlist {self.inference_time:.2f} seconds')
if self.inference_time > 0.25 * self.base_tf_seconds:
logger.warning('Inference took over 25/% of the candle time. Reduce pairlist to'
' avoid blinding open trades and degrading performance.')
self.pair_it = 0
self.inference_time = 0
return
def train_timer(self, do='start'):
"""
Timer designed to track the cumulative time spent training the full pairlist in
FreqAI.
"""
if do == 'start':
self.pair_it_train += 1
self.begin_time_train = time.time()
elif do == 'stop':
end = time.time()
self.train_time += (end - self.begin_time_train)
if self.pair_it_train == self.total_pairs:
logger.info(
f'Total time spent training pairlist {self.train_time:.2f} seconds')
self.pair_it_train = 0
self.train_time = 0
return
# Following methods which are overridden by user made prediction models.
# See freqai/prediction_models/CatboostPredictionModel.py for an example.
@abstractmethod
def train(self, unfiltered_dataframe: DataFrame, pair: str, dk: FreqaiDataKitchen) -> Any:
"""
Filter the training data and train a model to it. Train makes heavy use of the datahandler
for storing, saving, loading, and analyzing the data.
:param unfiltered_dataframe: Full dataframe for the current training period
:param metadata: pair metadata from strategy.
:return: Trained model which can be used to inference (self.predict)
"""
@abstractmethod
def fit(self, data_dictionary: Dict[str, Any]) -> Any:
"""
Most regressors use the same function names and arguments e.g. user
can drop in LGBMRegressor in place of CatBoostRegressor and all data
management will be properly handled by Freqai.
:param data_dictionary: Dict = the dictionary constructed by DataHandler to hold
all the training and test data/labels.
"""
return
@abstractmethod
def predict(
self, dataframe: DataFrame, dk: FreqaiDataKitchen, first: bool = True
) -> Tuple[DataFrame, NDArray[np.int_]]:
"""
Filter the prediction features data and predict with it.
:param unfiltered_dataframe: Full dataframe for the current backtest period.
:param dk: FreqaiDataKitchen = Data management/analysis tool associated to present pair only
:param first: boolean = whether this is the first prediction or not.
:return:
:predictions: np.array of predictions
:do_predict: np.array of 1s and 0s to indicate places where freqai needed to remove
data (NaNs) or felt uncertain about data (i.e. SVM and/or DI index)
"""

View File

@ -0,0 +1,99 @@
import logging
from typing import Any, Tuple
import numpy as np
import numpy.typing as npt
import pandas as pd
from pandas import DataFrame
from freqtrade.freqai.data_kitchen import FreqaiDataKitchen
from freqtrade.freqai.freqai_interface import IFreqaiModel
logger = logging.getLogger(__name__)
class BaseClassifierModel(IFreqaiModel):
"""
Base class for regression type models (e.g. Catboost, LightGBM, XGboost etc.).
User *must* inherit from this class and set fit() and predict(). See example scripts
such as prediction_models/CatboostPredictionModel.py for guidance.
"""
def train(
self, unfiltered_dataframe: DataFrame, pair: str, dk: FreqaiDataKitchen
) -> Any:
"""
Filter the training data and train a model to it. Train makes heavy use of the datakitchen
for storing, saving, loading, and analyzing the data.
:param unfiltered_dataframe: Full dataframe for the current training period
:param metadata: pair metadata from strategy.
:return:
:model: Trained model which can be used to inference (self.predict)
"""
logger.info("-------------------- Starting training " f"{pair} --------------------")
# filter the features requested by user in the configuration file and elegantly handle NaNs
features_filtered, labels_filtered = dk.filter_features(
unfiltered_dataframe,
dk.training_features_list,
dk.label_list,
training_filter=True,
)
start_date = unfiltered_dataframe["date"].iloc[0].strftime("%Y-%m-%d")
end_date = unfiltered_dataframe["date"].iloc[-1].strftime("%Y-%m-%d")
logger.info(f"-------------------- Training on data from {start_date} to "
f"{end_date}--------------------")
# split data into train/test data.
data_dictionary = dk.make_train_test_datasets(features_filtered, labels_filtered)
if not self.freqai_info.get('fit_live_predictions', 0) or not self.live:
dk.fit_labels()
# normalize all data based on train_dataset only
data_dictionary = dk.normalize_data(data_dictionary)
# optional additional data cleaning/analysis
self.data_cleaning_train(dk)
logger.info(
f'Training model on {len(dk.data_dictionary["train_features"].columns)}' " features"
)
logger.info(f'Training model on {len(data_dictionary["train_features"])} data points')
model = self.fit(data_dictionary)
logger.info(f"--------------------done training {pair}--------------------")
return model
def predict(
self, unfiltered_dataframe: DataFrame, dk: FreqaiDataKitchen, first: bool = False
) -> Tuple[DataFrame, npt.NDArray[np.int_]]:
"""
Filter the prediction features data and predict with it.
:param: unfiltered_dataframe: Full dataframe for the current backtest period.
:return:
:pred_df: dataframe containing the predictions
:do_predict: np.array of 1s and 0s to indicate places where freqai needed to remove
data (NaNs) or felt uncertain about data (PCA and DI index)
"""
dk.find_features(unfiltered_dataframe)
filtered_dataframe, _ = dk.filter_features(
unfiltered_dataframe, dk.training_features_list, training_filter=False
)
filtered_dataframe = dk.normalize_data_from_metadata(filtered_dataframe)
dk.data_dictionary["prediction_features"] = filtered_dataframe
self.data_cleaning_predict(dk, filtered_dataframe)
predictions = self.model.predict(dk.data_dictionary["prediction_features"])
pred_df = DataFrame(predictions, columns=dk.label_list)
predictions_prob = self.model.predict_proba(dk.data_dictionary["prediction_features"])
pred_df_prob = DataFrame(predictions_prob, columns=self.model.classes_)
pred_df = pd.concat([pred_df, pred_df_prob], axis=1)
return (pred_df, dk.do_predict)

View File

@ -0,0 +1,96 @@
import logging
from typing import Any, Tuple
import numpy as np
import numpy.typing as npt
from pandas import DataFrame
from freqtrade.freqai.data_kitchen import FreqaiDataKitchen
from freqtrade.freqai.freqai_interface import IFreqaiModel
logger = logging.getLogger(__name__)
class BaseRegressionModel(IFreqaiModel):
"""
Base class for regression type models (e.g. Catboost, LightGBM, XGboost etc.).
User *must* inherit from this class and set fit() and predict(). See example scripts
such as prediction_models/CatboostPredictionModel.py for guidance.
"""
def train(
self, unfiltered_dataframe: DataFrame, pair: str, dk: FreqaiDataKitchen
) -> Any:
"""
Filter the training data and train a model to it. Train makes heavy use of the datakitchen
for storing, saving, loading, and analyzing the data.
:param unfiltered_dataframe: Full dataframe for the current training period
:param metadata: pair metadata from strategy.
:return:
:model: Trained model which can be used to inference (self.predict)
"""
logger.info("-------------------- Starting training " f"{pair} --------------------")
# filter the features requested by user in the configuration file and elegantly handle NaNs
features_filtered, labels_filtered = dk.filter_features(
unfiltered_dataframe,
dk.training_features_list,
dk.label_list,
training_filter=True,
)
start_date = unfiltered_dataframe["date"].iloc[0].strftime("%Y-%m-%d")
end_date = unfiltered_dataframe["date"].iloc[-1].strftime("%Y-%m-%d")
logger.info(f"-------------------- Training on data from {start_date} to "
f"{end_date}--------------------")
# split data into train/test data.
data_dictionary = dk.make_train_test_datasets(features_filtered, labels_filtered)
if not self.freqai_info.get('fit_live_predictions', 0) or not self.live:
dk.fit_labels()
# normalize all data based on train_dataset only
data_dictionary = dk.normalize_data(data_dictionary)
# optional additional data cleaning/analysis
self.data_cleaning_train(dk)
logger.info(
f'Training model on {len(dk.data_dictionary["train_features"].columns)}' " features"
)
logger.info(f'Training model on {len(data_dictionary["train_features"])} data points')
model = self.fit(data_dictionary)
logger.info(f"--------------------done training {pair}--------------------")
return model
def predict(
self, unfiltered_dataframe: DataFrame, dk: FreqaiDataKitchen, first: bool = False
) -> Tuple[DataFrame, npt.NDArray[np.int_]]:
"""
Filter the prediction features data and predict with it.
:param: unfiltered_dataframe: Full dataframe for the current backtest period.
:return:
:pred_df: dataframe containing the predictions
:do_predict: np.array of 1s and 0s to indicate places where freqai needed to remove
data (NaNs) or felt uncertain about data (PCA and DI index)
"""
dk.find_features(unfiltered_dataframe)
filtered_dataframe, _ = dk.filter_features(
unfiltered_dataframe, dk.training_features_list, training_filter=False
)
filtered_dataframe = dk.normalize_data_from_metadata(filtered_dataframe)
dk.data_dictionary["prediction_features"] = filtered_dataframe
# optional additional data cleaning/analysis
self.data_cleaning_predict(dk, filtered_dataframe)
predictions = self.model.predict(dk.data_dictionary["prediction_features"])
pred_df = DataFrame(predictions, columns=dk.label_list)
pred_df = dk.denormalize_labels_from_metadata(pred_df)
return (pred_df, dk.do_predict)

View File

@ -0,0 +1,64 @@
import logging
from typing import Any
from pandas import DataFrame
from freqtrade.freqai.data_kitchen import FreqaiDataKitchen
from freqtrade.freqai.freqai_interface import IFreqaiModel
logger = logging.getLogger(__name__)
class BaseTensorFlowModel(IFreqaiModel):
"""
Base class for TensorFlow type models.
User *must* inherit from this class and set fit() and predict().
"""
def train(
self, unfiltered_dataframe: DataFrame, pair: str, dk: FreqaiDataKitchen
) -> Any:
"""
Filter the training data and train a model to it. Train makes heavy use of the datakitchen
for storing, saving, loading, and analyzing the data.
:param unfiltered_dataframe: Full dataframe for the current training period
:param metadata: pair metadata from strategy.
:return:
:model: Trained model which can be used to inference (self.predict)
"""
logger.info("-------------------- Starting training " f"{pair} --------------------")
# filter the features requested by user in the configuration file and elegantly handle NaNs
features_filtered, labels_filtered = dk.filter_features(
unfiltered_dataframe,
dk.training_features_list,
dk.label_list,
training_filter=True,
)
start_date = unfiltered_dataframe["date"].iloc[0].strftime("%Y-%m-%d")
end_date = unfiltered_dataframe["date"].iloc[-1].strftime("%Y-%m-%d")
logger.info(f"-------------------- Training on data from {start_date} to "
f"{end_date}--------------------")
# split data into train/test data.
data_dictionary = dk.make_train_test_datasets(features_filtered, labels_filtered)
if not self.freqai_info.get('fit_live_predictions', 0) or not self.live:
dk.fit_labels()
# normalize all data based on train_dataset only
data_dictionary = dk.normalize_data(data_dictionary)
# optional additional data cleaning/analysis
self.data_cleaning_train(dk)
logger.info(
f'Training model on {len(dk.data_dictionary["train_features"].columns)}' " features"
)
logger.info(f'Training model on {len(data_dictionary["train_features"])} data points')
model = self.fit(data_dictionary)
logger.info(f"--------------------done training {pair}--------------------")
return model

View File

@ -0,0 +1,41 @@
import logging
from typing import Any, Dict
from catboost import CatBoostClassifier, Pool
from freqtrade.freqai.prediction_models.BaseClassifierModel import BaseClassifierModel
logger = logging.getLogger(__name__)
class CatboostClassifier(BaseClassifierModel):
"""
User created prediction model. The class needs to override three necessary
functions, predict(), train(), fit(). The class inherits ModelHandler which
has its own DataHandler where data is held, saved, loaded, and managed.
"""
def fit(self, data_dictionary: Dict) -> Any:
"""
User sets up the training and test data to fit their desired model here
:params:
:data_dictionary: the dictionary constructed by DataHandler to hold
all the training and test data/labels.
"""
train_data = Pool(
data=data_dictionary["train_features"],
label=data_dictionary["train_labels"],
weight=data_dictionary["train_weights"],
)
cbr = CatBoostClassifier(
allow_writing_files=False,
loss_function='MultiClass',
**self.model_training_parameters,
)
cbr.fit(train_data)
return cbr

View File

@ -0,0 +1,53 @@
import gc
import logging
from typing import Any, Dict
from catboost import CatBoostRegressor, Pool
from freqtrade.freqai.prediction_models.BaseRegressionModel import BaseRegressionModel
logger = logging.getLogger(__name__)
class CatboostRegressor(BaseRegressionModel):
"""
User created prediction model. The class needs to override three necessary
functions, predict(), train(), fit(). The class inherits ModelHandler which
has its own DataHandler where data is held, saved, loaded, and managed.
"""
def fit(self, data_dictionary: Dict) -> Any:
"""
User sets up the training and test data to fit their desired model here
:param data_dictionary: the dictionary constructed by DataHandler to hold
all the training and test data/labels.
"""
train_data = Pool(
data=data_dictionary["train_features"],
label=data_dictionary["train_labels"],
weight=data_dictionary["train_weights"],
)
if self.freqai_info.get('data_split_parameters', {}).get('test_size', 0.1) == 0:
test_data = None
else:
test_data = Pool(
data=data_dictionary["test_features"],
label=data_dictionary["test_labels"],
weight=data_dictionary["test_weights"],
)
model = CatBoostRegressor(
allow_writing_files=False,
**self.model_training_parameters,
)
model.fit(X=train_data, eval_set=test_data)
# some evidence that catboost pools have memory leaks:
# https://github.com/catboost/catboost/issues/1835
del train_data, test_data
gc.collect()
return model

View File

@ -0,0 +1,44 @@
import logging
from typing import Any, Dict
from catboost import CatBoostRegressor # , Pool
from sklearn.multioutput import MultiOutputRegressor
from freqtrade.freqai.prediction_models.BaseRegressionModel import BaseRegressionModel
logger = logging.getLogger(__name__)
class CatboostRegressorMultiTarget(BaseRegressionModel):
"""
User created prediction model. The class needs to override three necessary
functions, predict(), train(), fit(). The class inherits ModelHandler which
has its own DataHandler where data is held, saved, loaded, and managed.
"""
def fit(self, data_dictionary: Dict) -> Any:
"""
User sets up the training and test data to fit their desired model here
:param data_dictionary: the dictionary constructed by DataHandler to hold
all the training and test data/labels.
"""
cbr = CatBoostRegressor(
allow_writing_files=False,
**self.model_training_parameters,
)
X = data_dictionary["train_features"]
y = data_dictionary["train_labels"]
eval_set = (data_dictionary["test_features"], data_dictionary["test_labels"])
sample_weight = data_dictionary["train_weights"]
model = MultiOutputRegressor(estimator=cbr)
model.fit(X=X, y=y, sample_weight=sample_weight) # , eval_set=eval_set)
if self.freqai_info.get('data_split_parameters', {}).get('test_size', 0.1) != 0:
train_score = model.score(X, y)
test_score = model.score(*eval_set)
logger.info(f"Train score {train_score}, Test score {test_score}")
return model

View File

@ -0,0 +1,43 @@
import logging
from typing import Any, Dict
from lightgbm import LGBMClassifier
from freqtrade.freqai.prediction_models.BaseClassifierModel import BaseClassifierModel
logger = logging.getLogger(__name__)
class LightGBMClassifier(BaseClassifierModel):
"""
User created prediction model. The class needs to override three necessary
functions, predict(), train(), fit(). The class inherits ModelHandler which
has its own DataHandler where data is held, saved, loaded, and managed.
"""
def fit(self, data_dictionary: Dict) -> Any:
"""
User sets up the training and test data to fit their desired model here
:params:
:data_dictionary: the dictionary constructed by DataHandler to hold
all the training and test data/labels.
"""
if self.freqai_info.get('data_split_parameters', {}).get('test_size', 0.1) == 0:
eval_set = None
test_weights = None
else:
eval_set = (data_dictionary["test_features"].to_numpy(),
data_dictionary["test_labels"].to_numpy()[:, 0])
test_weights = data_dictionary["test_weights"]
X = data_dictionary["train_features"].to_numpy()
y = data_dictionary["train_labels"].to_numpy()[:, 0]
train_weights = data_dictionary["train_weights"]
model = LGBMClassifier(**self.model_training_parameters)
model.fit(X=X, y=y, eval_set=eval_set, sample_weight=train_weights,
eval_sample_weight=[test_weights])
return model

View File

@ -0,0 +1,43 @@
import logging
from typing import Any, Dict
from lightgbm import LGBMRegressor
from freqtrade.freqai.prediction_models.BaseRegressionModel import BaseRegressionModel
logger = logging.getLogger(__name__)
class LightGBMRegressor(BaseRegressionModel):
"""
User created prediction model. The class needs to override three necessary
functions, predict(), train(), fit(). The class inherits ModelHandler which
has its own DataHandler where data is held, saved, loaded, and managed.
"""
def fit(self, data_dictionary: Dict) -> Any:
"""
Most regressors use the same function names and arguments e.g. user
can drop in LGBMRegressor in place of CatBoostRegressor and all data
management will be properly handled by Freqai.
:param data_dictionary: the dictionary constructed by DataHandler to hold
all the training and test data/labels.
"""
if self.freqai_info.get('data_split_parameters', {}).get('test_size', 0.1) == 0:
eval_set = None
eval_weights = None
else:
eval_set = (data_dictionary["test_features"], data_dictionary["test_labels"])
eval_weights = data_dictionary["test_weights"]
X = data_dictionary["train_features"]
y = data_dictionary["train_labels"]
train_weights = data_dictionary["train_weights"]
model = LGBMRegressor(**self.model_training_parameters)
model.fit(X=X, y=y, eval_set=eval_set, sample_weight=train_weights,
eval_sample_weight=[eval_weights])
return model

View File

@ -0,0 +1,39 @@
import logging
from typing import Any, Dict
from lightgbm import LGBMRegressor
from sklearn.multioutput import MultiOutputRegressor
from freqtrade.freqai.prediction_models.BaseRegressionModel import BaseRegressionModel
logger = logging.getLogger(__name__)
class LightGBMRegressorMultiTarget(BaseRegressionModel):
"""
User created prediction model. The class needs to override three necessary
functions, predict(), train(), fit(). The class inherits ModelHandler which
has its own DataHandler where data is held, saved, loaded, and managed.
"""
def fit(self, data_dictionary: Dict) -> Any:
"""
User sets up the training and test data to fit their desired model here
:param data_dictionary: the dictionary constructed by DataHandler to hold
all the training and test data/labels.
"""
lgb = LGBMRegressor(**self.model_training_parameters)
X = data_dictionary["train_features"]
y = data_dictionary["train_labels"]
eval_set = (data_dictionary["test_features"], data_dictionary["test_labels"])
sample_weight = data_dictionary["train_weights"]
model = MultiOutputRegressor(estimator=lgb)
model.fit(X=X, y=y, sample_weight=sample_weight) # , eval_set=eval_set)
train_score = model.score(X, y)
test_score = model.score(*eval_set)
logger.info(f"Train score {train_score}, Test score {test_score}")
return model

View File

@ -21,17 +21,17 @@ from freqtrade.enums import (ExitCheckTuple, ExitType, RPCMessageType, RunMode,
State, TradingMode)
from freqtrade.exceptions import (DependencyException, ExchangeError, InsufficientFundsError,
InvalidOrderException, PricingError)
from freqtrade.exchange import timeframe_to_minutes, timeframe_to_seconds
from freqtrade.exchange.exchange import timeframe_to_next_date
from freqtrade.exchange import timeframe_to_minutes, timeframe_to_next_date, timeframe_to_seconds
from freqtrade.misc import safe_value_fallback, safe_value_fallback2
from freqtrade.mixins import LoggingMixin
from freqtrade.persistence import Order, PairLocks, Trade, cleanup_db, init_db
from freqtrade.persistence import Order, PairLocks, Trade, init_db
from freqtrade.plugins.pairlistmanager import PairListManager
from freqtrade.plugins.protectionmanager import ProtectionManager
from freqtrade.resolvers import ExchangeResolver, StrategyResolver
from freqtrade.rpc import RPCManager
from freqtrade.strategy.interface import IStrategy
from freqtrade.strategy.strategy_wrapper import strategy_safe_wrapper
from freqtrade.util import FtPrecise
from freqtrade.wallets import Wallets
@ -149,7 +149,7 @@ class FreqtradeBot(LoggingMixin):
self.check_for_open_trades()
self.rpc.cleanup()
cleanup_db()
Trade.commit()
self.exchange.close()
def startup(self) -> None:
@ -158,6 +158,8 @@ class FreqtradeBot(LoggingMixin):
performs startup tasks
"""
self.rpc.startup_messages(self.config, self.pairlists, self.protections)
# Update older trades with precision and precision mode
self.startup_backpopulate_precision()
if not self.edge:
# Adjust stoploss if it was changed
Trade.stoploss_reinitialization(self.strategy.stoploss)
@ -214,6 +216,7 @@ class FreqtradeBot(LoggingMixin):
if self.trading_mode == TradingMode.FUTURES:
self._schedule.run_pending()
Trade.commit()
self.rpc.process_msg_queue(self.dataprovider._msg_queue)
self.last_process = datetime.now(timezone.utc)
def process_stopped(self) -> None:
@ -236,7 +239,7 @@ class FreqtradeBot(LoggingMixin):
'status':
f"{len(open_trades)} open trades active.\n\n"
f"Handle these trades manually on {self.exchange.name}, "
f"or '/start' the bot again and use '/stopbuy' "
f"or '/start' the bot again and use '/stopentry' "
f"to handle open trades gracefully. \n"
f"{'Note: Trades are simulated (dry run).' if self.config['dry_run'] else ''}",
}
@ -267,7 +270,7 @@ class FreqtradeBot(LoggingMixin):
Return the number of free open trades slots or 0 if
max number of open trades reached
"""
open_trades = len(Trade.get_open_trades())
open_trades = Trade.get_open_trade_count()
return max(0, self.config['max_open_trades'] - open_trades)
def update_funding_fees(self):
@ -284,6 +287,18 @@ class FreqtradeBot(LoggingMixin):
else:
return 0.0
def startup_backpopulate_precision(self):
trades = Trade.get_trades([Trade.contract_size.is_(None)])
for trade in trades:
if trade.exchange != self.exchange.id:
continue
trade.precision_mode = self.exchange.precisionMode
trade.amount_precision = self.exchange.get_precision_amount(trade.pair)
trade.price_precision = self.exchange.get_precision_price(trade.pair)
trade.contract_size = self.exchange.get_contract_size(trade.pair)
Trade.commit()
def startup_update_open_orders(self):
"""
Updates open orders based on order list kept in the database.
@ -403,7 +418,7 @@ class FreqtradeBot(LoggingMixin):
whitelist = copy.deepcopy(self.active_pair_whitelist)
if not whitelist:
logger.info("Active pair whitelist is empty.")
self.log_once("Active pair whitelist is empty.", logger.info)
return trades_created
# Remove pairs for currently opened trades from the whitelist
for trade in Trade.get_open_trades():
@ -412,8 +427,8 @@ class FreqtradeBot(LoggingMixin):
logger.debug('Ignoring %s in pair whitelist', trade.pair)
if not whitelist:
logger.info("No currency pair in active pair whitelist, "
"but checking to exit open trades.")
self.log_once("No currency pair in active pair whitelist, "
"but checking to exit open trades.", logger.info)
return trades_created
if PairLocks.is_global_lock(side='*'):
# This only checks for total locks (both sides).
@ -524,39 +539,61 @@ class FreqtradeBot(LoggingMixin):
If the strategy triggers the adjustment, a new order gets issued.
Once that completes, the existing trade is modified to match new data.
"""
if self.strategy.max_entry_position_adjustment > -1:
count_of_buys = trade.nr_of_successful_entries
if count_of_buys > self.strategy.max_entry_position_adjustment:
logger.debug(f"Max adjustment entries for {trade.pair} has been reached.")
return
else:
logger.debug("Max adjustment entries is set to unlimited.")
current_rate = self.exchange.get_rate(
trade.pair, side='entry', is_short=trade.is_short, refresh=True)
current_profit = trade.calc_profit_ratio(current_rate)
current_entry_rate, current_exit_rate = self.exchange.get_rates(
trade.pair, True, trade.is_short)
min_stake_amount = self.exchange.get_min_pair_stake_amount(trade.pair,
current_rate,
current_entry_profit = trade.calc_profit_ratio(current_entry_rate)
current_exit_profit = trade.calc_profit_ratio(current_exit_rate)
min_entry_stake = self.exchange.get_min_pair_stake_amount(trade.pair,
current_entry_rate,
self.strategy.stoploss)
max_stake_amount = self.exchange.get_max_pair_stake_amount(trade.pair, current_rate)
min_exit_stake = self.exchange.get_min_pair_stake_amount(trade.pair,
current_exit_rate,
self.strategy.stoploss)
max_entry_stake = self.exchange.get_max_pair_stake_amount(trade.pair, current_entry_rate)
stake_available = self.wallets.get_available_stake_amount()
logger.debug(f"Calling adjust_trade_position for pair {trade.pair}")
stake_amount = strategy_safe_wrapper(self.strategy.adjust_trade_position,
default_retval=None)(
trade=trade, current_time=datetime.now(timezone.utc), current_rate=current_rate,
current_profit=current_profit, min_stake=min_stake_amount,
max_stake=min(max_stake_amount, stake_available))
trade=trade,
current_time=datetime.now(timezone.utc), current_rate=current_entry_rate,
current_profit=current_entry_profit, min_stake=min_entry_stake,
max_stake=min(max_entry_stake, stake_available),
current_entry_rate=current_entry_rate, current_exit_rate=current_exit_rate,
current_entry_profit=current_entry_profit, current_exit_profit=current_exit_profit
)
if stake_amount is not None and stake_amount > 0.0:
# We should increase our position
self.execute_entry(trade.pair, stake_amount, price=current_rate,
if self.strategy.max_entry_position_adjustment > -1:
count_of_entries = trade.nr_of_successful_entries
if count_of_entries > self.strategy.max_entry_position_adjustment:
logger.debug(f"Max adjustment entries for {trade.pair} has been reached.")
return
else:
logger.debug("Max adjustment entries is set to unlimited.")
self.execute_entry(trade.pair, stake_amount, price=current_entry_rate,
trade=trade, is_short=trade.is_short)
if stake_amount is not None and stake_amount < 0.0:
# We should decrease our position
# TODO: Selling part of the trade not implemented yet.
logger.error(f"Unable to decrease trade position / sell partially"
f" for pair {trade.pair}, feature not implemented.")
amount = abs(float(FtPrecise(stake_amount) / FtPrecise(current_exit_rate)))
if amount > trade.amount:
# This is currently ineffective as remaining would become < min tradable
# Fixing this would require checking for 0.0 there -
# if we decide that this callback is allowed to "fully exit"
logger.info(
f"Adjusting amount to trade.amount as it is higher. {amount} > {trade.amount}")
amount = trade.amount
remaining = (trade.amount - amount) * current_exit_rate
if remaining < min_exit_stake:
logger.info(f'Remaining amount of {remaining} would be too small.')
return
self.execute_trade_exit(trade, current_exit_rate, exit_check=ExitCheckTuple(
exit_type=ExitType.PARTIAL_EXIT), sub_trade_amt=amount)
def _check_depth_of_market(self, pair: str, conf: Dict, side: SignalDirection) -> bool:
"""
@ -600,7 +637,8 @@ class FreqtradeBot(LoggingMixin):
ordertype: Optional[str] = None,
enter_tag: Optional[str] = None,
trade: Optional[Trade] = None,
order_adjust: bool = False
order_adjust: bool = False,
leverage_: Optional[float] = None,
) -> bool:
"""
Executes a limit buy for the given pair
@ -616,7 +654,7 @@ class FreqtradeBot(LoggingMixin):
pos_adjust = trade is not None
enter_limit_requested, stake_amount, leverage = self.get_valid_enter_price_and_stake(
pair, price, stake_amount, trade_side, enter_tag, trade, order_adjust)
pair, price, stake_amount, trade_side, enter_tag, trade, order_adjust, leverage_)
if not stake_amount:
return False
@ -713,7 +751,11 @@ class FreqtradeBot(LoggingMixin):
leverage=leverage,
is_short=is_short,
trading_mode=self.trading_mode,
funding_fees=funding_fees
funding_fees=funding_fees,
amount_precision=self.exchange.get_precision_amount(pair),
price_precision=self.exchange.get_precision_price(pair),
precision_mode=self.exchange.precisionMode,
contract_size=self.exchange.get_contract_size(pair),
)
else:
# This is additional buy, we reset fee_open_currency so timeout checking can work
@ -730,7 +772,7 @@ class FreqtradeBot(LoggingMixin):
# Updating wallets
self.wallets.update()
self._notify_enter(trade, order, order_type)
self._notify_enter(trade, order_obj, order_type, sub_trade=pos_adjust)
if pos_adjust:
if order_status == 'closed':
@ -739,8 +781,8 @@ class FreqtradeBot(LoggingMixin):
else:
logger.info(f"DCA order {order_status}, will wait for resolution: {trade}")
# Update fees if order is closed
if order_status == 'closed':
# Update fees if order is non-opened
if order_status in constants.NON_OPEN_EXCHANGE_STATES:
self.update_trade_state(trade, order_id, order)
return True
@ -763,6 +805,7 @@ class FreqtradeBot(LoggingMixin):
entry_tag: Optional[str],
trade: Optional[Trade],
order_adjust: bool,
leverage_: Optional[float],
) -> Tuple[float, float, float]:
if price:
@ -785,8 +828,11 @@ class FreqtradeBot(LoggingMixin):
if not enter_limit_requested:
raise PricingError('Could not determine entry price.')
if trade is None:
if self.trading_mode != TradingMode.SPOT and trade is None:
max_leverage = self.exchange.get_max_leverage(pair, stake_amount)
if leverage_:
leverage = leverage_
else:
leverage = strategy_safe_wrapper(self.strategy.leverage, default_retval=1.0)(
pair=pair,
current_time=datetime.now(timezone.utc),
@ -794,7 +840,7 @@ class FreqtradeBot(LoggingMixin):
proposed_leverage=1.0,
max_leverage=max_leverage,
side=trade_side, entry_tag=entry_tag,
) if self.trading_mode != TradingMode.SPOT else 1.0
)
# Cap leverage between 1.0 and max_leverage.
leverage = min(max(leverage, 1.0), max_leverage)
else:
@ -829,13 +875,14 @@ class FreqtradeBot(LoggingMixin):
return enter_limit_requested, stake_amount, leverage
def _notify_enter(self, trade: Trade, order: Dict, order_type: Optional[str] = None,
fill: bool = False) -> None:
def _notify_enter(self, trade: Trade, order: Order, order_type: Optional[str] = None,
fill: bool = False, sub_trade: bool = False) -> None:
"""
Sends rpc notification when a entry order occurred.
"""
msg_type = RPCMessageType.ENTRY_FILL if fill else RPCMessageType.ENTRY
open_rate = safe_value_fallback(order, 'average', 'price')
open_rate = order.safe_price
if open_rate is None:
open_rate = trade.open_rate
@ -859,15 +906,17 @@ class FreqtradeBot(LoggingMixin):
'stake_amount': trade.stake_amount,
'stake_currency': self.config['stake_currency'],
'fiat_currency': self.config.get('fiat_display_currency', None),
'amount': safe_value_fallback(order, 'filled', 'amount') or trade.amount,
'amount': order.safe_amount_after_fee,
'open_date': trade.open_date or datetime.utcnow(),
'current_rate': current_rate,
'sub_trade': sub_trade,
}
# Send the message
self.rpc.send_msg(msg)
def _notify_enter_cancel(self, trade: Trade, order_type: str, reason: str) -> None:
def _notify_enter_cancel(self, trade: Trade, order_type: str, reason: str,
sub_trade: bool = False) -> None:
"""
Sends rpc notification when a entry order cancel occurred.
"""
@ -892,6 +941,7 @@ class FreqtradeBot(LoggingMixin):
'open_date': trade.open_date,
'current_rate': current_rate,
'reason': reason,
'sub_trade': sub_trade,
}
# Send the message
@ -1015,7 +1065,7 @@ class FreqtradeBot(LoggingMixin):
trade.stoploss_order_id = None
logger.error(f'Unable to place a stoploss order on exchange. {e}')
logger.warning('Exiting the trade forcefully')
self.execute_trade_exit(trade, trade.stop_loss, exit_check=ExitCheckTuple(
self.execute_trade_exit(trade, stop_price, exit_check=ExitCheckTuple(
exit_type=ExitType.EMERGENCY_EXIT))
except ExchangeError:
@ -1085,7 +1135,7 @@ class FreqtradeBot(LoggingMixin):
if (trade.is_open
and stoploss_order
and stoploss_order['status'] in ('canceled', 'cancelled')):
if self.create_stoploss_order(trade=trade, stop_price=trade.stop_loss):
if self.create_stoploss_order(trade=trade, stop_price=trade.stoploss_or_liquidation):
return False
else:
trade.stoploss_order_id = None
@ -1114,7 +1164,7 @@ class FreqtradeBot(LoggingMixin):
:param order: Current on exchange stoploss order
:return: None
"""
stoploss_norm = self.exchange.price_to_precision(trade.pair, trade.stop_loss)
stoploss_norm = self.exchange.price_to_precision(trade.pair, trade.stoploss_or_liquidation)
if self.exchange.stoploss_adjust(stoploss_norm, order, side=trade.exit_side):
# we check if the update is necessary
@ -1132,7 +1182,7 @@ class FreqtradeBot(LoggingMixin):
f"for pair {trade.pair}")
# Create new stoploss order
if not self.create_stoploss_order(trade=trade, stop_price=trade.stop_loss):
if not self.create_stoploss_order(trade=trade, stop_price=stoploss_norm):
logger.warning(f"Could not create trailing stoploss order "
f"for pair {trade.pair}.")
@ -1365,16 +1415,22 @@ class FreqtradeBot(LoggingMixin):
trade.open_order_id = None
trade.exit_reason = None
cancelled = True
self.wallets.update()
else:
# TODO: figure out how to handle partially complete sell orders
reason = constants.CANCEL_REASON['PARTIALLY_FILLED_KEEP_OPEN']
cancelled = False
self.wallets.update()
order_obj = trade.select_order_by_order_id(order['id'])
if not order_obj:
raise DependencyException(
f"Order_obj not found for {order['id']}. This should not have happened.")
sub_trade = order_obj.amount != trade.amount
self._notify_exit_cancel(
trade,
order_type=self.strategy.order_types['exit'],
reason=reason
reason=reason, order=order_obj, sub_trade=sub_trade
)
return cancelled
@ -1415,6 +1471,7 @@ class FreqtradeBot(LoggingMixin):
*,
exit_tag: Optional[str] = None,
ordertype: Optional[str] = None,
sub_trade_amt: float = None,
) -> bool:
"""
Executes a trade exit for the given trade and limit
@ -1431,15 +1488,10 @@ class FreqtradeBot(LoggingMixin):
)
exit_type = 'exit'
exit_reason = exit_tag or exit_check.exit_reason
if exit_check.exit_type in (ExitType.STOP_LOSS, ExitType.TRAILING_STOP_LOSS):
if exit_check.exit_type in (
ExitType.STOP_LOSS, ExitType.TRAILING_STOP_LOSS, ExitType.LIQUIDATION):
exit_type = 'stoploss'
# if stoploss is on exchange and we are on dry_run mode,
# we consider the sell price stop price
if (self.config['dry_run'] and exit_type == 'stoploss'
and self.strategy.order_types['stoploss_on_exchange']):
limit = trade.stop_loss
# set custom_exit_price if available
proposed_limit_rate = limit
current_profit = trade.calc_profit_ratio(limit)
@ -1460,14 +1512,17 @@ class FreqtradeBot(LoggingMixin):
# Emergency sells (default to market!)
order_type = self.strategy.order_types.get("emergency_exit", "market")
amount = self._safe_exit_amount(trade.pair, trade.amount)
amount = self._safe_exit_amount(trade.pair, sub_trade_amt or trade.amount)
time_in_force = self.strategy.order_time_in_force['exit']
if not strategy_safe_wrapper(self.strategy.confirm_trade_exit, default_retval=True)(
if (exit_check.exit_type != ExitType.LIQUIDATION
and not sub_trade_amt
and not strategy_safe_wrapper(
self.strategy.confirm_trade_exit, default_retval=True)(
pair=trade.pair, trade=trade, order_type=order_type, amount=amount, rate=limit,
time_in_force=time_in_force, exit_reason=exit_reason,
sell_reason=exit_reason, # sellreason -> compatibility
current_time=datetime.now(timezone.utc)):
current_time=datetime.now(timezone.utc))):
logger.info(f"User denied exit for {trade.pair}.")
return False
@ -1497,11 +1552,12 @@ class FreqtradeBot(LoggingMixin):
trade.close_rate_requested = limit
trade.exit_reason = exit_reason
if not sub_trade_amt:
# Lock pair for one candle to prevent immediate re-trading
self.strategy.lock_pair(trade.pair, datetime.now(timezone.utc),
reason='Auto lock')
self._notify_exit(trade, order_type)
self._notify_exit(trade, order_type, sub_trade=bool(sub_trade_amt), order=order_obj)
# In case of market sell orders the order can be closed immediately
if order.get('status', 'unknown') in ('closed', 'expired'):
self.update_trade_state(trade, trade.open_order_id, order)
@ -1509,16 +1565,27 @@ class FreqtradeBot(LoggingMixin):
return True
def _notify_exit(self, trade: Trade, order_type: str, fill: bool = False) -> None:
def _notify_exit(self, trade: Trade, order_type: str, fill: bool = False,
sub_trade: bool = False, order: Order = None) -> None:
"""
Sends rpc notification when a sell occurred.
"""
profit_rate = trade.close_rate if trade.close_rate else trade.close_rate_requested
profit_trade = trade.calc_profit(rate=profit_rate)
# Use cached rates here - it was updated seconds ago.
current_rate = self.exchange.get_rate(
trade.pair, side='exit', is_short=trade.is_short, refresh=False) if not fill else None
# second condition is for mypy only; order will always be passed during sub trade
if sub_trade and order is not None:
amount = order.safe_filled if fill else order.amount
profit_rate = order.safe_price
profit = trade.calc_profit(rate=profit_rate, amount=amount, open_rate=trade.open_rate)
profit_ratio = trade.calc_profit_ratio(profit_rate, amount, trade.open_rate)
else:
profit_rate = trade.close_rate if trade.close_rate else trade.close_rate_requested
profit = trade.calc_profit(rate=profit_rate) + (0.0 if fill else trade.realized_profit)
profit_ratio = trade.calc_profit_ratio(profit_rate)
amount = trade.amount
gain = "profit" if profit_ratio > 0 else "loss"
msg = {
@ -1532,11 +1599,11 @@ class FreqtradeBot(LoggingMixin):
'gain': gain,
'limit': profit_rate,
'order_type': order_type,
'amount': trade.amount,
'amount': amount,
'open_rate': trade.open_rate,
'close_rate': trade.close_rate,
'close_rate': profit_rate,
'current_rate': current_rate,
'profit_amount': profit_trade,
'profit_amount': profit,
'profit_ratio': profit_ratio,
'buy_tag': trade.enter_tag,
'enter_tag': trade.enter_tag,
@ -1544,19 +1611,18 @@ class FreqtradeBot(LoggingMixin):
'exit_reason': trade.exit_reason,
'open_date': trade.open_date,
'close_date': trade.close_date or datetime.utcnow(),
'stake_amount': trade.stake_amount,
'stake_currency': self.config['stake_currency'],
'fiat_currency': self.config.get('fiat_display_currency'),
'sub_trade': sub_trade,
'cumulative_profit': trade.realized_profit,
}
if 'fiat_display_currency' in self.config:
msg.update({
'fiat_currency': self.config['fiat_display_currency'],
})
# Send the message
self.rpc.send_msg(msg)
def _notify_exit_cancel(self, trade: Trade, order_type: str, reason: str) -> None:
def _notify_exit_cancel(self, trade: Trade, order_type: str, reason: str,
order: Order, sub_trade: bool = False) -> None:
"""
Sends rpc notification when a sell cancel occurred.
"""
@ -1582,7 +1648,7 @@ class FreqtradeBot(LoggingMixin):
'gain': gain,
'limit': profit_rate or 0,
'order_type': order_type,
'amount': trade.amount,
'amount': order.safe_amount_after_fee,
'open_rate': trade.open_rate,
'current_rate': current_rate,
'profit_amount': profit_trade,
@ -1596,6 +1662,8 @@ class FreqtradeBot(LoggingMixin):
'stake_currency': self.config['stake_currency'],
'fiat_currency': self.config.get('fiat_display_currency', None),
'reason': reason,
'sub_trade': sub_trade,
'stake_amount': trade.stake_amount,
}
if 'fiat_display_currency' in self.config:
@ -1650,40 +1718,51 @@ class FreqtradeBot(LoggingMixin):
self.handle_order_fee(trade, order_obj, order)
trade.update_trade(order_obj)
# TODO: is the below necessary? it's already done in update_trade for filled buys
trade.recalc_trade_from_orders()
Trade.commit()
if order['status'] in constants.NON_OPEN_EXCHANGE_STATES:
if order.get('status') in constants.NON_OPEN_EXCHANGE_STATES:
# If a entry order was closed, force update on stoploss on exchange
if order.get('side') == trade.entry_side:
trade = self.cancel_stoploss_on_exchange(trade)
# TODO: Margin will need to use interest_rate as well.
# interest_rate = self.exchange.get_interest_rate()
trade.set_isolated_liq(self.exchange.get_liquidation_price(
leverage=trade.leverage,
pair=trade.pair,
amount=trade.amount,
open_rate=trade.open_rate,
is_short=trade.is_short
))
if not self.edge:
# TODO: should shorting/leverage be supported by Edge,
# then this will need to be fixed.
trade.adjust_stop_loss(trade.open_rate, self.strategy.stoploss, initial=True)
if order.get('side') == trade.entry_side or trade.amount > 0:
# Must also run for partial exits
# TODO: Margin will need to use interest_rate as well.
# interest_rate = self.exchange.get_interest_rate()
trade.set_liquidation_price(self.exchange.get_liquidation_price(
leverage=trade.leverage,
pair=trade.pair,
amount=trade.amount,
stake_amount=trade.stake_amount,
open_rate=trade.open_rate,
is_short=trade.is_short
))
# Updating wallets when order is closed
self.wallets.update()
Trade.commit()
if not trade.is_open:
self.order_close_notify(trade, order_obj, stoploss_order, send_msg)
return False
def order_close_notify(
self, trade: Trade, order: Order, stoploss_order: bool, send_msg: bool):
"""send "fill" notifications"""
sub_trade = not isclose(order.safe_amount_after_fee,
trade.amount, abs_tol=constants.MATH_CLOSE_PREC)
if order.ft_order_side == trade.exit_side:
# Exit notification
if send_msg and not stoploss_order and not trade.open_order_id:
self._notify_exit(trade, '', True)
self._notify_exit(trade, '', fill=True, sub_trade=sub_trade, order=order)
if not trade.is_open:
self.handle_protections(trade.pair, trade.trade_direction)
elif send_msg and not trade.open_order_id and not stoploss_order:
# Enter fill
self._notify_enter(trade, order, fill=True)
return False
self._notify_enter(trade, order, fill=True, sub_trade=sub_trade)
def handle_protections(self, pair: str, side: LongShort) -> None:
prot_trig = self.protections.stop_per_pair(pair, side=side)
@ -1806,6 +1885,9 @@ class FreqtradeBot(LoggingMixin):
if fee_rate is not None and fee_rate < 0.02:
# Only update if fee-rate is < 2%
trade.update_fee(fee_cost, fee_currency, fee_rate, order.get('side', ''))
else:
logger.warning(
f"Not updating {order.get('side', '')}-fee - rate: {fee_rate}, {fee_currency}.")
if not isclose(amount, order_amount, abs_tol=constants.MATH_CLOSE_PREC):
# * Leverage could be a cause for this warning

View File

@ -1,20 +1,20 @@
from decimal import Decimal
from math import ceil
from freqtrade.exceptions import OperationalException
from freqtrade.util import FtPrecise
one = Decimal(1.0)
four = Decimal(4.0)
twenty_four = Decimal(24.0)
one = FtPrecise(1.0)
four = FtPrecise(4.0)
twenty_four = FtPrecise(24.0)
def interest(
exchange_name: str,
borrowed: Decimal,
rate: Decimal,
hours: Decimal
) -> Decimal:
borrowed: FtPrecise,
rate: FtPrecise,
hours: FtPrecise
) -> FtPrecise:
"""
Equation to calculate interest on margin trades
@ -31,13 +31,13 @@ def interest(
"""
exchange_name = exchange_name.lower()
if exchange_name == "binance":
return borrowed * rate * ceil(hours) / twenty_four
return borrowed * rate * FtPrecise(ceil(hours)) / twenty_four
elif exchange_name == "kraken":
# Rounded based on https://kraken-fees-calculator.github.io/
return borrowed * rate * (one + ceil(hours / four))
return borrowed * rate * (one + FtPrecise(ceil(hours / four)))
elif exchange_name == "ftx":
# As Explained under #Interest rates section in
# https://help.ftx.com/hc/en-us/articles/360053007671-Spot-Margin-Trading-Explainer
return borrowed * rate * ceil(hours) / twenty_four
return borrowed * rate * FtPrecise(ceil(hours)) / twenty_four
else:
raise OperationalException(f"Leverage not available on {exchange_name} with freqtrade")

151
freqtrade/optimize/backtesting.py Executable file → Normal file
View File

@ -23,7 +23,8 @@ from freqtrade.data.dataprovider import DataProvider
from freqtrade.enums import (BacktestState, CandleType, ExitCheckTuple, ExitType, RunMode,
TradingMode)
from freqtrade.exceptions import DependencyException, OperationalException
from freqtrade.exchange import timeframe_to_minutes, timeframe_to_seconds
from freqtrade.exchange import (amount_to_contract_precision, price_to_precision,
timeframe_to_minutes, timeframe_to_seconds)
from freqtrade.mixins import LoggingMixin
from freqtrade.optimize.backtest_caching import get_strategy_run_id
from freqtrade.optimize.bt_progress import BTProgress
@ -89,6 +90,9 @@ class Backtesting:
self.dataprovider = DataProvider(self.config, self.exchange)
if self.config.get('strategy_list'):
if self.config.get('freqai', {}).get('enabled', False):
raise OperationalException(
"You can't use strategy_list and freqai at the same time.")
for strat in list(self.config['strategy_list']):
stratconf = deepcopy(self.config)
stratconf['strategy'] = strat
@ -128,6 +132,7 @@ class Backtesting:
self.fee = config['fee']
else:
self.fee = self.exchange.get_fee(symbol=self.pairlists.whitelist[0])
self.precision_mode = self.exchange.precisionMode
self.timerange = TimeRange.parse_timerange(
None if self.config.get('timerange') is None else str(self.config.get('timerange')))
@ -207,6 +212,15 @@ class Backtesting:
"""
self.progress.init_step(BacktestState.DATALOAD, 1)
if self.config.get('freqai', {}).get('enabled', False):
startup_candles = int(self.config.get('freqai', {}).get('startup_candles', 0))
if not startup_candles:
raise OperationalException('FreqAI backtesting module requires user set '
'startup_candles in config.')
self.required_startup += int(self.config.get('freqai', {}).get('startup_candles', 0))
logger.info(f'Increasing startup_candle_count for freqai to {self.required_startup}')
self.config['startup_candle_count'] = self.required_startup
data = history.load_data(
datadir=self.config['datadir'],
pairs=self.pairlists.whitelist,
@ -253,7 +267,7 @@ class Backtesting:
funding_rates_dict = history.load_data(
datadir=self.config['datadir'],
pairs=self.pairlists.whitelist,
timeframe=self.exchange._ft_has['mark_ohlcv_timeframe'],
timeframe=self.exchange.get_option('mark_ohlcv_timeframe'),
timerange=self.timerange,
startup_candles=0,
fail_without_data=True,
@ -265,12 +279,12 @@ class Backtesting:
mark_rates_dict = history.load_data(
datadir=self.config['datadir'],
pairs=self.pairlists.whitelist,
timeframe=self.exchange._ft_has['mark_ohlcv_timeframe'],
timeframe=self.exchange.get_option('mark_ohlcv_timeframe'),
timerange=self.timerange,
startup_candles=0,
fail_without_data=True,
data_format=self.config.get('dataformat_ohlcv', 'json'),
candle_type=CandleType.from_string(self.exchange._ft_has["mark_ohlcv_price"])
candle_type=CandleType.from_string(self.exchange.get_option("mark_ohlcv_price"))
)
# Combine data to avoid combining the data per trade.
unavailable_pairs = []
@ -381,7 +395,8 @@ class Backtesting:
Get close rate for backtesting result
"""
# Special handling if high or low hit STOP_LOSS or ROI
if exit.exit_type in (ExitType.STOP_LOSS, ExitType.TRAILING_STOP_LOSS):
if exit.exit_type in (
ExitType.STOP_LOSS, ExitType.TRAILING_STOP_LOSS, ExitType.LIQUIDATION):
return self._get_close_rate_for_stoploss(row, trade, exit, trade_dur)
elif exit.exit_type == (ExitType.ROI):
return self._get_close_rate_for_roi(row, trade, exit, trade_dur)
@ -396,11 +411,16 @@ class Backtesting:
is_short = trade.is_short or False
leverage = trade.leverage or 1.0
side_1 = -1 if is_short else 1
if exit.exit_type == ExitType.LIQUIDATION and trade.liquidation_price:
stoploss_value = trade.liquidation_price
else:
stoploss_value = trade.stop_loss
if is_short:
if trade.stop_loss < row[LOW_IDX]:
if stoploss_value < row[LOW_IDX]:
return row[OPEN_IDX]
else:
if trade.stop_loss > row[HIGH_IDX]:
if stoploss_value > row[HIGH_IDX]:
return row[OPEN_IDX]
# Special case: trailing triggers within same candle as trade opened. Assume most
@ -433,7 +453,7 @@ class Backtesting:
return max(row[LOW_IDX], stop_rate)
# Set close_rate to stoploss
return trade.stop_loss
return stoploss_value
def _get_close_rate_for_roi(self, row: Tuple, trade: LocalTrade, exit: ExitCheckTuple,
trade_dur: int) -> float:
@ -497,26 +517,53 @@ class Backtesting:
def _get_adjust_trade_entry_for_candle(self, trade: LocalTrade, row: Tuple
) -> LocalTrade:
current_profit = trade.calc_profit_ratio(row[OPEN_IDX])
min_stake = self.exchange.get_min_pair_stake_amount(trade.pair, row[OPEN_IDX], -0.1)
max_stake = self.exchange.get_max_pair_stake_amount(trade.pair, row[OPEN_IDX])
current_rate = row[OPEN_IDX]
current_date = row[DATE_IDX].to_pydatetime()
current_profit = trade.calc_profit_ratio(current_rate)
min_stake = self.exchange.get_min_pair_stake_amount(trade.pair, current_rate, -0.1)
max_stake = self.exchange.get_max_pair_stake_amount(trade.pair, current_rate)
stake_available = self.wallets.get_available_stake_amount()
stake_amount = strategy_safe_wrapper(self.strategy.adjust_trade_position,
default_retval=None)(
trade=trade, # type: ignore[arg-type]
current_time=row[DATE_IDX].to_pydatetime(), current_rate=row[OPEN_IDX],
current_time=current_date, current_rate=current_rate,
current_profit=current_profit, min_stake=min_stake,
max_stake=min(max_stake, stake_available))
max_stake=min(max_stake, stake_available),
current_entry_rate=current_rate, current_exit_rate=current_rate,
current_entry_profit=current_profit, current_exit_profit=current_profit)
# Check if we should increase our position
if stake_amount is not None and stake_amount > 0.0:
check_adjust_entry = True
if self.strategy.max_entry_position_adjustment > -1:
entry_count = trade.nr_of_successful_entries
check_adjust_entry = (entry_count <= self.strategy.max_entry_position_adjustment)
if check_adjust_entry:
pos_trade = self._enter_trade(
trade.pair, row, 'short' if trade.is_short else 'long', stake_amount, trade)
if pos_trade is not None:
self.wallets.update()
return pos_trade
if stake_amount is not None and stake_amount < 0.0:
amount = abs(stake_amount) / current_rate
if amount > trade.amount:
# This is currently ineffective as remaining would become < min tradable
amount = trade.amount
remaining = (trade.amount - amount) * current_rate
if remaining < min_stake:
# Remaining stake is too low to be sold.
return trade
exit_ = ExitCheckTuple(ExitType.PARTIAL_EXIT)
pos_trade = self._get_exit_for_signal(trade, row, exit_, amount)
if pos_trade is not None:
order = pos_trade.orders[-1]
if self._get_order_filled(order.price, row):
order.close_bt_order(current_date, trade)
trade.recalc_trade_from_orders()
self.wallets.update()
return pos_trade
return trade
def _get_order_filled(self, rate: float, row: Tuple) -> bool:
@ -528,11 +575,6 @@ class Backtesting:
# Check if we need to adjust our current positions
if self.strategy.position_adjustment_enable:
check_adjust_entry = True
if self.strategy.max_entry_position_adjustment > -1:
entry_count = trade.nr_of_successful_entries
check_adjust_entry = (entry_count <= self.strategy.max_entry_position_adjustment)
if check_adjust_entry:
trade = self._get_adjust_trade_entry_for_candle(trade, row)
enter = row[SHORT_IDX] if trade.is_short else row[LONG_IDX]
@ -548,14 +590,15 @@ class Backtesting:
return t
return None
def _get_exit_for_signal(self, trade: LocalTrade, row: Tuple,
exit_: ExitCheckTuple) -> Optional[LocalTrade]:
def _get_exit_for_signal(
self, trade: LocalTrade, row: Tuple, exit_: ExitCheckTuple,
amount: Optional[float] = None) -> Optional[LocalTrade]:
exit_candle_time: datetime = row[DATE_IDX].to_pydatetime()
if exit_.exit_flag:
trade.close_date = exit_candle_time
exit_reason = exit_.exit_reason
amount_ = amount if amount is not None else trade.amount
trade_dur = int((trade.close_date_utc - trade.open_date_utc).total_seconds() // 60)
try:
close_rate = self._get_close_rate(row, trade, exit_, trade_dur)
@ -564,7 +607,8 @@ class Backtesting:
# call the custom exit price,with default value as previous close_rate
current_profit = trade.calc_profit_ratio(close_rate)
order_type = self.strategy.order_types['exit']
if exit_.exit_type in (ExitType.EXIT_SIGNAL, ExitType.CUSTOM_EXIT):
if exit_.exit_type in (ExitType.EXIT_SIGNAL, ExitType.CUSTOM_EXIT,
ExitType.PARTIAL_EXIT):
# Checks and adds an exit tag, after checking that the length of the
# row has the length for an exit tag column
if (
@ -592,21 +636,34 @@ class Backtesting:
# Confirm trade exit:
time_in_force = self.strategy.order_time_in_force['exit']
if not strategy_safe_wrapper(self.strategy.confirm_trade_exit, default_retval=True)(
if (exit_.exit_type not in (ExitType.LIQUIDATION, ExitType.PARTIAL_EXIT)
and not strategy_safe_wrapper(
self.strategy.confirm_trade_exit, default_retval=True)(
pair=trade.pair,
trade=trade, # type: ignore[arg-type]
order_type='limit',
amount=trade.amount,
order_type=order_type,
amount=amount_,
rate=close_rate,
time_in_force=time_in_force,
sell_reason=exit_reason, # deprecated
exit_reason=exit_reason,
current_time=exit_candle_time):
current_time=exit_candle_time)):
return None
trade.exit_reason = exit_reason
return self._exit_trade(trade, row, close_rate, amount_)
return None
def _exit_trade(self, trade: LocalTrade, sell_row: Tuple,
close_rate: float, amount: float = None) -> Optional[LocalTrade]:
self.order_id_counter += 1
exit_candle_time = sell_row[DATE_IDX].to_pydatetime()
order_type = self.strategy.order_types['exit']
# amount = amount or trade.amount
amount = amount_to_contract_precision(amount or trade.amount, trade.amount_precision,
self.precision_mode, trade.contract_size)
rate = price_to_precision(close_rate, trade.price_precision, self.precision_mode)
order = Order(
id=self.order_id_counter,
ft_trade_id=trade.id,
@ -620,18 +677,16 @@ class Backtesting:
side=trade.exit_side,
order_type=order_type,
status="open",
price=close_rate,
average=close_rate,
amount=trade.amount,
price=rate,
average=rate,
amount=amount,
filled=0,
remaining=trade.amount,
cost=trade.amount * close_rate,
remaining=amount,
cost=amount * rate,
)
trade.orders.append(order)
return trade
return None
def _get_exit_trade_entry(self, trade: LocalTrade, row: Tuple) -> Optional[LocalTrade]:
exit_candle_time: datetime = row[DATE_IDX].to_pydatetime()
@ -773,7 +828,17 @@ class Backtesting:
if stake_amount and (not min_stake_amount or stake_amount > min_stake_amount):
self.order_id_counter += 1
base_currency = self.exchange.get_pair_base_currency(pair)
amount = round((stake_amount / propose_rate) * leverage, 8)
precision_price = self.exchange.get_precision_price(pair)
propose_rate = price_to_precision(propose_rate, precision_price, self.precision_mode)
amount_p = (stake_amount / propose_rate) * leverage
contract_size = self.exchange.get_contract_size(pair)
precision_amount = self.exchange.get_precision_amount(pair)
amount = amount_to_contract_precision(amount_p, precision_amount, self.precision_mode,
contract_size)
# Backcalculate actual stake amount.
stake_amount = amount * propose_rate / leverage
is_short = (direction == 'short')
# Necessary for Margin trading. Disabled until support is enabled.
# interest_rate = self.exchange.get_interest_rate()
@ -802,15 +867,20 @@ class Backtesting:
trading_mode=self.trading_mode,
leverage=leverage,
# interest_rate=interest_rate,
amount_precision=precision_amount,
price_precision=precision_price,
precision_mode=self.precision_mode,
contract_size=contract_size,
orders=[],
)
trade.adjust_stop_loss(trade.open_rate, self.strategy.stoploss, initial=True)
trade.set_isolated_liq(self.exchange.get_liquidation_price(
trade.set_liquidation_price(self.exchange.get_liquidation_price(
pair=pair,
open_rate=propose_rate,
amount=amount,
stake_amount=trade.stake_amount,
leverage=leverage,
is_short=is_short,
))
@ -858,6 +928,8 @@ class Backtesting:
# Ignore trade if entry-order did not fill yet
continue
exit_row = data[pair][-1]
self._exit_trade(trade, exit_row, exit_row[OPEN_IDX], trade.amount)
trade.orders[-1].close_bt_order(exit_row[DATE_IDX].to_pydatetime(), trade)
trade.close_date = exit_row[DATE_IDX].to_pydatetime()
trade.exit_reason = ExitType.FORCE_EXIT.value
@ -999,7 +1071,7 @@ class Backtesting:
return None
return row
def backtest(self, processed: Dict,
def backtest(self, processed: Dict, # noqa: max-complexity: 13
start_date: datetime, end_date: datetime,
max_open_trades: int = 0, position_stacking: bool = False,
enable_protections: bool = False) -> Dict[str, Any]:
@ -1101,6 +1173,11 @@ class Backtesting:
if order and self._get_order_filled(order.price, row):
order.close_bt_order(current_time, trade)
trade.open_order_id = None
sub_trade = order.safe_amount_after_fee != trade.amount
if sub_trade:
order.close_bt_order(current_time, trade)
trade.recalc_trade_from_orders()
else:
trade.close_date = current_time
trade.close(order.price, show_msg=False)

View File

@ -24,13 +24,15 @@ from pandas import DataFrame
from freqtrade.constants import DATETIME_PRINT_FORMAT, FTHYPT_FILEVERSION, LAST_BT_RESULT_FN
from freqtrade.data.converter import trim_dataframes
from freqtrade.data.history import get_timerange
from freqtrade.enums import HyperoptState
from freqtrade.exceptions import OperationalException
from freqtrade.misc import deep_merge_dicts, file_dump_json, plural
from freqtrade.optimize.backtesting import Backtesting
# Import IHyperOpt and IHyperOptLoss to allow unpickling classes from these modules
from freqtrade.optimize.hyperopt_auto import HyperOptAuto
from freqtrade.optimize.hyperopt_loss_interface import IHyperOptLoss
from freqtrade.optimize.hyperopt_tools import HyperoptTools, hyperopt_serializer
from freqtrade.optimize.hyperopt_tools import (HyperoptStateContainer, HyperoptTools,
hyperopt_serializer)
from freqtrade.optimize.optimize_reports import generate_strategy_stats
from freqtrade.resolvers.hyperopt_resolver import HyperOptLossResolver
@ -74,10 +76,14 @@ class Hyperopt:
self.dimensions: List[Dimension] = []
self.config = config
self.min_date: datetime
self.max_date: datetime
self.backtesting = Backtesting(self.config)
self.pairlist = self.backtesting.pairlists.whitelist
self.custom_hyperopt: HyperOptAuto
self.analyze_per_epoch = self.config.get('analyze_per_epoch', False)
HyperoptStateContainer.set_state(HyperoptState.STARTUP)
if not self.config.get('hyperopt'):
self.custom_hyperopt = HyperOptAuto(self.config)
@ -290,6 +296,7 @@ class Hyperopt:
Called once per epoch to optimize whatever is configured.
Keep this function as optimized as possible!
"""
HyperoptStateContainer.set_state(HyperoptState.OPTIMIZE)
backtest_start_time = datetime.now(timezone.utc)
params_dict = self._get_params_dict(self.dimensions, raw_params)
@ -321,6 +328,10 @@ class Hyperopt:
with self.data_pickle_file.open('rb') as f:
processed = load(f, mmap_mode='r')
if self.analyze_per_epoch:
# Data is not yet analyzed, rerun populate_indicators.
processed = self.advise_and_trim(processed)
bt_results = self.backtesting.backtest(
processed=processed,
start_date=self.min_date,
@ -406,22 +417,33 @@ class Hyperopt:
def _set_random_state(self, random_state: Optional[int]) -> int:
return random_state or random.randint(1, 2**16 - 1)
def prepare_hyperopt_data(self) -> None:
data, timerange = self.backtesting.load_bt_data()
self.backtesting.load_bt_data_detail()
logger.info("Dataload complete. Calculating indicators")
def advise_and_trim(self, data: Dict[str, DataFrame]) -> Dict[str, DataFrame]:
preprocessed = self.backtesting.strategy.advise_all_indicators(data)
# Trim startup period from analyzed dataframe to get correct dates for output.
processed = trim_dataframes(preprocessed, timerange, self.backtesting.required_startup)
processed = trim_dataframes(preprocessed, self.timerange, self.backtesting.required_startup)
self.min_date, self.max_date = get_timerange(processed)
return processed
logger.info(f'Hyperopting with data from {self.min_date.strftime(DATETIME_PRINT_FORMAT)} '
def prepare_hyperopt_data(self) -> None:
HyperoptStateContainer.set_state(HyperoptState.DATALOAD)
data, self.timerange = self.backtesting.load_bt_data()
self.backtesting.load_bt_data_detail()
logger.info("Dataload complete. Calculating indicators")
if not self.analyze_per_epoch:
HyperoptStateContainer.set_state(HyperoptState.INDICATORS)
preprocessed = self.advise_and_trim(data)
logger.info(f'Hyperopting with data from '
f'{self.min_date.strftime(DATETIME_PRINT_FORMAT)} '
f'up to {self.max_date.strftime(DATETIME_PRINT_FORMAT)} '
f'({(self.max_date - self.min_date).days} days)..')
# Store non-trimmed data - will be trimmed after signal generation.
dump(preprocessed, self.data_pickle_file)
else:
dump(data, self.data_pickle_file)
def get_asked_points(self, n_points: int) -> Tuple[List[List[Any]], List[bool]]:
"""
@ -483,6 +505,7 @@ class Hyperopt:
self.backtesting.exchange._api_async = None
self.backtesting.exchange.loop = None # type: ignore
self.backtesting.exchange._loop_lock = None # type: ignore
self.backtesting.exchange._cache_lock = None # type: ignore
# self.backtesting.exchange = None # type: ignore
self.backtesting.pairlists = None # type: ignore

View File

@ -13,6 +13,7 @@ from colorama import Fore, Style
from pandas import isna, json_normalize
from freqtrade.constants import FTHYPT_FILEVERSION, USERPATH_STRATEGIES
from freqtrade.enums import HyperoptState
from freqtrade.exceptions import OperationalException
from freqtrade.misc import deep_merge_dicts, round_coin_value, round_dict, safe_value_fallback2
from freqtrade.optimize.hyperopt_epoch_filters import hyperopt_filter_epochs
@ -32,6 +33,15 @@ def hyperopt_serializer(x):
return str(x)
class HyperoptStateContainer():
""" Singleton class to track state of hyperopt"""
state: HyperoptState = HyperoptState.OPTIMIZE
@classmethod
def set_state(cls, value: HyperoptState):
cls.state = value
class HyperoptTools():
@staticmethod

View File

@ -1,5 +1,5 @@
# flake8: noqa: F401
from freqtrade.persistence.models import cleanup_db, init_db
from freqtrade.persistence.models import init_db
from freqtrade.persistence.pairlock_middleware import PairLocks
from freqtrade.persistence.trade_model import LocalTrade, Order, Trade

View File

@ -95,6 +95,7 @@ def migrate_trades_and_orders_table(
exit_reason = get_column_def(cols, 'sell_reason', get_column_def(cols, 'exit_reason', 'null'))
strategy = get_column_def(cols, 'strategy', 'null')
enter_tag = get_column_def(cols, 'buy_tag', get_column_def(cols, 'enter_tag', 'null'))
realized_profit = get_column_def(cols, 'realized_profit', '0.0')
trading_mode = get_column_def(cols, 'trading_mode', 'null')
@ -129,6 +130,11 @@ def migrate_trades_and_orders_table(
get_column_def(cols, 'sell_order_status', 'null'))
amount_requested = get_column_def(cols, 'amount_requested', 'amount')
amount_precision = get_column_def(cols, 'amount_precision', 'null')
price_precision = get_column_def(cols, 'price_precision', 'null')
precision_mode = get_column_def(cols, 'precision_mode', 'null')
contract_size = get_column_def(cols, 'contract_size', 'null')
# Schema migration necessary
with engine.begin() as connection:
connection.execute(text(f"alter table trades rename to {trade_back_name}"))
@ -155,7 +161,8 @@ def migrate_trades_and_orders_table(
max_rate, min_rate, exit_reason, exit_order_status, strategy, enter_tag,
timeframe, open_trade_value, close_profit_abs,
trading_mode, leverage, liquidation_price, is_short,
interest_rate, funding_fees
interest_rate, funding_fees, realized_profit,
amount_precision, price_precision, precision_mode, contract_size
)
select id, lower(exchange), pair, {base_currency} base_currency,
{stake_currency} stake_currency,
@ -181,7 +188,9 @@ def migrate_trades_and_orders_table(
{open_trade_value} open_trade_value, {close_profit_abs} close_profit_abs,
{trading_mode} trading_mode, {leverage} leverage, {liquidation_price} liquidation_price,
{is_short} is_short, {interest_rate} interest_rate,
{funding_fees} funding_fees
{funding_fees} funding_fees, {realized_profit} realized_profit,
{amount_precision} amount_precision, {price_precision} price_precision,
{precision_mode} precision_mode, {contract_size} contract_size
from {trade_back_name}
"""))
@ -297,8 +306,11 @@ def check_migrate(engine, decl_base, previous_tables) -> None:
# Check if migration necessary
# Migrates both trades and orders table!
if not has_column(cols_orders, 'stop_price'):
# if not has_column(cols_trades, 'base_currency'):
# if ('orders' not in previous_tables
# or not has_column(cols_orders, 'stop_price')):
migrating = False
if not has_column(cols_trades, 'contract_size'):
migrating = True
logger.info(f"Running database migration for trades - "
f"backup: {table_back_name}, {order_table_bak_name}")
migrate_trades_and_orders_table(
@ -306,6 +318,7 @@ def check_migrate(engine, decl_base, previous_tables) -> None:
order_table_bak_name, cols_orders)
if not has_column(cols_pairlocks, 'side'):
migrating = True
logger.info(f"Running database migration for pairlocks - "
f"backup: {pairlock_table_bak_name}")
@ -320,3 +333,6 @@ def check_migrate(engine, decl_base, previous_tables) -> None:
set_sqlite_to_wal(engine)
fix_old_dry_orders(engine)
if migrating:
logger.info("Database migration finished.")

View File

@ -53,7 +53,7 @@ def init_db(db_url: str) -> None:
# https://docs.sqlalchemy.org/en/13/orm/contextual.html#thread-local-scope
# Scoped sessions proxy requests to the appropriate thread-local session.
# We should use the scoped_session object - not a seperately initialized version
Trade._session = scoped_session(sessionmaker(bind=engine, autoflush=True))
Trade._session = scoped_session(sessionmaker(bind=engine, autoflush=False))
Trade.query = Trade._session.query_property()
Order.query = Trade._session.query_property()
PairLock.query = Trade._session.query_property()
@ -61,11 +61,3 @@ def init_db(db_url: str) -> None:
previous_tables = inspect(engine).get_table_names()
_DECL_BASE.metadata.create_all(engine)
check_migrate(engine, decl_base=_DECL_BASE, previous_tables=previous_tables)
def cleanup_db() -> None:
"""
Flushes all pending operations to disk.
:return: None
"""
Trade.commit()

View File

@ -3,18 +3,21 @@ This module contains the class to persist trades into SQLite
"""
import logging
from datetime import datetime, timedelta, timezone
from decimal import Decimal
from math import isclose
from typing import Any, Dict, List, Optional
from sqlalchemy import (Boolean, Column, DateTime, Enum, Float, ForeignKey, Integer, String,
UniqueConstraint, desc, func)
from sqlalchemy.orm import Query, lazyload, relationship
from freqtrade.constants import DATETIME_PRINT_FORMAT, NON_OPEN_EXCHANGE_STATES, BuySell, LongShort
from freqtrade.constants import (DATETIME_PRINT_FORMAT, MATH_CLOSE_PREC, NON_OPEN_EXCHANGE_STATES,
BuySell, LongShort)
from freqtrade.enums import ExitType, TradingMode
from freqtrade.exceptions import DependencyException, OperationalException
from freqtrade.exchange import amount_to_contract_precision, price_to_precision
from freqtrade.leverage import interest
from freqtrade.persistence.base import _DECL_BASE
from freqtrade.util import FtPrecise
logger = logging.getLogger(__name__)
@ -176,10 +179,9 @@ class Order(_DECL_BASE):
self.remaining = 0
self.status = 'closed'
self.ft_is_open = False
if (self.ft_order_side == trade.entry_side
and len(trade.select_filled_orders(trade.entry_side)) == 1):
if (self.ft_order_side == trade.entry_side):
trade.open_rate = self.price
trade.recalc_open_trade_value()
trade.recalc_trade_from_orders()
trade.adjust_stop_loss(trade.open_rate, trade.stop_loss_pct, refresh=True)
@staticmethod
@ -195,7 +197,7 @@ class Order(_DECL_BASE):
if filtered_orders:
oobj = filtered_orders[0]
oobj.update_from_ccxt_object(order)
Order.query.session.commit()
Trade.commit()
else:
logger.warning(f"Did not find order for {order}.")
@ -237,6 +239,7 @@ class LocalTrade():
trades: List['LocalTrade'] = []
trades_open: List['LocalTrade'] = []
total_profit: float = 0
realized_profit: float = 0
id: int = 0
@ -290,6 +293,10 @@ class LocalTrade():
timeframe: Optional[int] = None
trading_mode: TradingMode = TradingMode.SPOT
amount_precision: Optional[float] = None
price_precision: Optional[float] = None
precision_mode: Optional[int] = None
contract_size: Optional[float] = None
# Leverage trading properties
liquidation_price: Optional[float] = None
@ -302,6 +309,16 @@ class LocalTrade():
# Futures properties
funding_fees: Optional[float] = None
@property
def stoploss_or_liquidation(self) -> float:
if self.liquidation_price:
if self.is_short:
return min(self.stop_loss, self.liquidation_price)
else:
return max(self.stop_loss, self.liquidation_price)
return self.stop_loss
@property
def buy_tag(self) -> Optional[str]:
"""
@ -437,6 +454,7 @@ class LocalTrade():
if self.close_date else None),
'close_timestamp': int(self.close_date.replace(
tzinfo=timezone.utc).timestamp() * 1000) if self.close_date else None,
'realized_profit': self.realized_profit or 0.0,
'close_rate': self.close_rate,
'close_rate_requested': self.close_rate_requested,
'close_profit': self.close_profit, # Deprecated
@ -497,7 +515,7 @@ class LocalTrade():
self.max_rate = max(current_price, self.max_rate or self.open_rate)
self.min_rate = min(current_price_low, self.min_rate or self.open_rate)
def set_isolated_liq(self, liquidation_price: Optional[float]):
def set_liquidation_price(self, liquidation_price: Optional[float]):
"""
Method you should use to set self.liquidation price.
Assures stop_loss is not passed the liquidation price
@ -506,22 +524,14 @@ class LocalTrade():
return
self.liquidation_price = liquidation_price
def _set_stop_loss(self, stop_loss: float, percent: float):
def __set_stop_loss(self, stop_loss: float, percent: float):
"""
Method you should use to set self.stop_loss.
Assures stop_loss is not passed the liquidation price
Method used internally to set self.stop_loss.
"""
if self.liquidation_price is not None:
if self.is_short:
sl = min(stop_loss, self.liquidation_price)
else:
sl = max(stop_loss, self.liquidation_price)
else:
sl = stop_loss
stop_loss_norm = price_to_precision(stop_loss, self.price_precision, self.precision_mode)
if not self.stop_loss:
self.initial_stop_loss = sl
self.stop_loss = sl
self.initial_stop_loss = stop_loss_norm
self.stop_loss = stop_loss_norm
self.stop_loss_pct = -1 * abs(percent)
self.stoploss_last_update = datetime.utcnow()
@ -543,19 +553,14 @@ class LocalTrade():
leverage = self.leverage or 1.0
if self.is_short:
new_loss = float(current_price * (1 + abs(stoploss / leverage)))
# If trading with leverage, don't set the stoploss below the liquidation price
if self.liquidation_price:
new_loss = min(self.liquidation_price, new_loss)
else:
new_loss = float(current_price * (1 - abs(stoploss / leverage)))
# If trading with leverage, don't set the stoploss below the liquidation price
if self.liquidation_price:
new_loss = max(self.liquidation_price, new_loss)
# no stop loss assigned yet
if self.initial_stop_loss_pct is None or refresh:
self._set_stop_loss(new_loss, stoploss)
self.initial_stop_loss = new_loss
self.__set_stop_loss(new_loss, stoploss)
self.initial_stop_loss = price_to_precision(
new_loss, self.price_precision, self.precision_mode)
self.initial_stop_loss_pct = -1 * abs(stoploss)
# evaluate if the stop loss needs to be updated
@ -569,7 +574,7 @@ class LocalTrade():
# ? decreasing the minimum stoploss
if (higher_stop and not self.is_short) or (lower_stop and self.is_short):
logger.debug(f"{self.pair} - Adjusting stoploss...")
self._set_stop_loss(new_loss, stoploss)
self.__set_stop_loss(new_loss, stoploss)
else:
logger.debug(f"{self.pair} - Keeping current stoploss...")
@ -601,14 +606,30 @@ class LocalTrade():
if self.is_open:
payment = "SELL" if self.is_short else "BUY"
logger.info(f'{order.order_type.upper()}_{payment} has been fulfilled for {self}.')
# condition to avoid reset value when updating fees
if self.open_order_id == order.order_id:
self.open_order_id = None
else:
logger.warning(
f'Got different open_order_id {self.open_order_id} != {order.order_id}')
self.recalc_trade_from_orders()
elif order.ft_order_side == self.exit_side:
if self.is_open:
payment = "BUY" if self.is_short else "SELL"
# * On margin shorts, you buy a little bit more than the amount (amount + interest)
logger.info(f'{order.order_type.upper()}_{payment} has been fulfilled for {self}.')
# condition to avoid reset value when updating fees
if self.open_order_id == order.order_id:
self.open_order_id = None
else:
logger.warning(
f'Got different open_order_id {self.open_order_id} != {order.order_id}')
amount_tr = amount_to_contract_precision(self.amount, self.amount_precision,
self.precision_mode, self.contract_size)
if isclose(order.safe_amount_after_fee, amount_tr, abs_tol=MATH_CLOSE_PREC):
self.close(order.safe_price)
else:
self.recalc_trade_from_orders()
elif order.ft_order_side == 'stoploss':
self.stoploss_order_id = None
self.close_rate_requested = self.stop_loss
@ -627,11 +648,11 @@ class LocalTrade():
"""
self.close_rate = rate
self.close_date = self.close_date or datetime.utcnow()
self.close_profit = self.calc_profit_ratio(rate)
self.close_profit_abs = self.calc_profit(rate)
self.close_profit_abs = self.calc_profit(rate) + self.realized_profit
self.is_open = False
self.exit_order_status = 'closed'
self.open_order_id = None
self.recalc_trade_from_orders(is_closing=True)
if show_msg:
logger.info(
'Marking %s as closed as the trade is fulfilled and found no open orders for it.',
@ -677,13 +698,13 @@ class LocalTrade():
"""
return len([o for o in self.orders if o.ft_order_side == self.exit_side])
def _calc_open_trade_value(self) -> float:
def _calc_open_trade_value(self, amount: float, open_rate: float) -> float:
"""
Calculate the open_rate including open_fee.
:return: Price in of the open trade incl. Fees
"""
open_trade = Decimal(self.amount) * Decimal(self.open_rate)
fees = open_trade * Decimal(self.fee_open)
open_trade = FtPrecise(amount) * FtPrecise(open_rate)
fees = open_trade * FtPrecise(self.fee_open)
if self.is_short:
return float(open_trade - fees)
else:
@ -694,39 +715,39 @@ class LocalTrade():
Recalculate open_trade_value.
Must be called whenever open_rate, fee_open is changed.
"""
self.open_trade_value = self._calc_open_trade_value()
self.open_trade_value = self._calc_open_trade_value(self.amount, self.open_rate)
def calculate_interest(self) -> Decimal:
def calculate_interest(self) -> FtPrecise:
"""
Calculate interest for this trade. Only applicable for Margin trading.
"""
zero = Decimal(0.0)
zero = FtPrecise(0.0)
# If nothing was borrowed
if self.trading_mode != TradingMode.MARGIN or self.has_no_leverage:
return zero
open_date = self.open_date.replace(tzinfo=None)
now = (self.close_date or datetime.now(timezone.utc)).replace(tzinfo=None)
sec_per_hour = Decimal(3600)
total_seconds = Decimal((now - open_date).total_seconds())
sec_per_hour = FtPrecise(3600)
total_seconds = FtPrecise((now - open_date).total_seconds())
hours = total_seconds / sec_per_hour or zero
rate = Decimal(self.interest_rate)
borrowed = Decimal(self.borrowed)
rate = FtPrecise(self.interest_rate)
borrowed = FtPrecise(self.borrowed)
return interest(exchange_name=self.exchange, borrowed=borrowed, rate=rate, hours=hours)
def _calc_base_close(self, amount: Decimal, rate: float, fee: float) -> Decimal:
def _calc_base_close(self, amount: FtPrecise, rate: float, fee: float) -> FtPrecise:
close_trade = amount * Decimal(rate)
fees = close_trade * Decimal(fee)
close_trade = amount * FtPrecise(rate)
fees = close_trade * FtPrecise(fee)
if self.is_short:
return close_trade + fees
else:
return close_trade - fees
def calc_close_trade_value(self, rate: float) -> float:
def calc_close_trade_value(self, rate: float, amount: float = None) -> float:
"""
Calculate the Trade's close value including fees
:param rate: rate to compare with.
@ -735,96 +756,145 @@ class LocalTrade():
if rate is None and not self.close_rate:
return 0.0
amount = Decimal(self.amount)
amount1 = FtPrecise(amount or self.amount)
trading_mode = self.trading_mode or TradingMode.SPOT
if trading_mode == TradingMode.SPOT:
return float(self._calc_base_close(amount, rate, self.fee_close))
return float(self._calc_base_close(amount1, rate, self.fee_close))
elif (trading_mode == TradingMode.MARGIN):
total_interest = self.calculate_interest()
if self.is_short:
amount = amount + total_interest
return float(self._calc_base_close(amount, rate, self.fee_close))
amount1 = amount1 + total_interest
return float(self._calc_base_close(amount1, rate, self.fee_close))
else:
# Currency already owned for longs, no need to purchase
return float(self._calc_base_close(amount, rate, self.fee_close) - total_interest)
return float(self._calc_base_close(amount1, rate, self.fee_close) - total_interest)
elif (trading_mode == TradingMode.FUTURES):
funding_fees = self.funding_fees or 0.0
# Positive funding_fees -> Trade has gained from fees.
# Negative funding_fees -> Trade had to pay the fees.
if self.is_short:
return float(self._calc_base_close(amount, rate, self.fee_close)) - funding_fees
return float(self._calc_base_close(amount1, rate, self.fee_close)) - funding_fees
else:
return float(self._calc_base_close(amount, rate, self.fee_close)) + funding_fees
return float(self._calc_base_close(amount1, rate, self.fee_close)) + funding_fees
else:
raise OperationalException(
f"{self.trading_mode.value} trading is not yet available using freqtrade")
def calc_profit(self, rate: float) -> float:
def calc_profit(self, rate: float, amount: float = None, open_rate: float = None) -> float:
"""
Calculate the absolute profit in stake currency between Close and Open trade
:param rate: close rate to compare with.
:param amount: Amount to use for the calculation. Falls back to trade.amount if not set.
:param open_rate: open_rate to use. Defaults to self.open_rate if not provided.
:return: profit in stake currency as float
"""
close_trade_value = self.calc_close_trade_value(rate)
close_trade_value = self.calc_close_trade_value(rate, amount)
if amount is None or open_rate is None:
open_trade_value = self.open_trade_value
else:
open_trade_value = self._calc_open_trade_value(amount, open_rate)
if self.is_short:
profit = self.open_trade_value - close_trade_value
profit = open_trade_value - close_trade_value
else:
profit = close_trade_value - self.open_trade_value
profit = close_trade_value - open_trade_value
return float(f"{profit:.8f}")
def calc_profit_ratio(self, rate: float) -> float:
def calc_profit_ratio(
self, rate: float, amount: float = None, open_rate: float = None) -> float:
"""
Calculates the profit as ratio (including fee).
:param rate: rate to compare with.
:param amount: Amount to use for the calculation. Falls back to trade.amount if not set.
:param open_rate: open_rate to use. Defaults to self.open_rate if not provided.
:return: profit ratio as float
"""
close_trade_value = self.calc_close_trade_value(rate)
close_trade_value = self.calc_close_trade_value(rate, amount)
if amount is None or open_rate is None:
open_trade_value = self.open_trade_value
else:
open_trade_value = self._calc_open_trade_value(amount, open_rate)
short_close_zero = (self.is_short and close_trade_value == 0.0)
long_close_zero = (not self.is_short and self.open_trade_value == 0.0)
long_close_zero = (not self.is_short and open_trade_value == 0.0)
leverage = self.leverage or 1.0
if (short_close_zero or long_close_zero):
return 0.0
else:
if self.is_short:
profit_ratio = (1 - (close_trade_value / self.open_trade_value)) * leverage
profit_ratio = (1 - (close_trade_value / open_trade_value)) * leverage
else:
profit_ratio = ((close_trade_value / self.open_trade_value) - 1) * leverage
profit_ratio = ((close_trade_value / open_trade_value) - 1) * leverage
return float(f"{profit_ratio:.8f}")
def recalc_trade_from_orders(self):
total_amount = 0.0
total_stake = 0.0
def recalc_trade_from_orders(self, *, is_closing: bool = False):
ZERO = FtPrecise(0.0)
current_amount = FtPrecise(0.0)
current_stake = FtPrecise(0.0)
total_stake = 0.0 # Total stake after all buy orders (does not subtract!)
avg_price = FtPrecise(0.0)
close_profit = 0.0
close_profit_abs = 0.0
profit = None
for o in self.orders:
if (o.ft_is_open or
(o.ft_order_side != self.entry_side) or
(o.status not in NON_OPEN_EXCHANGE_STATES)):
if o.ft_is_open or not o.filled:
continue
tmp_amount = o.safe_amount_after_fee
tmp_price = o.average or o.price
if tmp_amount > 0.0 and tmp_price is not None:
total_amount += tmp_amount
total_stake += tmp_price * tmp_amount
tmp_amount = FtPrecise(o.safe_amount_after_fee)
tmp_price = FtPrecise(o.safe_price)
if total_amount > 0:
is_exit = o.ft_order_side != self.entry_side
side = FtPrecise(-1 if is_exit else 1)
if tmp_amount > ZERO and tmp_price is not None:
current_amount += tmp_amount * side
price = avg_price if is_exit else tmp_price
current_stake += price * tmp_amount * side
if current_amount > ZERO:
avg_price = current_stake / current_amount
if is_exit:
# Process partial exits
exit_rate = o.safe_price
exit_amount = o.safe_amount_after_fee
profit = self.calc_profit(rate=exit_rate, amount=exit_amount,
open_rate=float(avg_price))
close_profit_abs += profit
close_profit = self.calc_profit_ratio(
exit_rate, amount=exit_amount, open_rate=avg_price)
else:
total_stake = total_stake + self._calc_open_trade_value(tmp_amount, price)
if close_profit:
self.close_profit = close_profit
self.realized_profit = close_profit_abs
self.close_profit_abs = profit
current_amount_tr = amount_to_contract_precision(
float(current_amount), self.amount_precision, self.precision_mode, self.contract_size)
if current_amount_tr > 0.0:
# Trade is still open
# Leverage not updated, as we don't allow changing leverage through DCA at the moment.
self.open_rate = total_stake / total_amount
self.stake_amount = total_stake / (self.leverage or 1.0)
self.amount = total_amount
self.fee_open_cost = self.fee_open * total_stake
self.open_rate = float(current_stake / current_amount)
self.amount = current_amount_tr
self.stake_amount = float(current_stake) / (self.leverage or 1.0)
self.fee_open_cost = self.fee_open * float(current_stake)
self.recalc_open_trade_value()
if self.stop_loss_pct is not None and self.open_rate is not None:
self.adjust_stop_loss(self.open_rate, self.stop_loss_pct)
elif is_closing and total_stake > 0:
# Close profit abs / maximum owned
# Fees are considered as they are part of close_profit_abs
self.close_profit = (close_profit_abs / total_stake) * self.leverage
self.close_profit_abs = close_profit_abs
def select_order_by_order_id(self, order_id: str) -> Optional[Order]:
"""
@ -846,7 +916,7 @@ class LocalTrade():
"""
orders = self.orders
if order_side:
orders = [o for o in self.orders if o.ft_order_side == order_side]
orders = [o for o in orders if o.ft_order_side == order_side]
if is_open is not None:
orders = [o for o in orders if o.ft_is_open == is_open]
if len(orders) > 0:
@ -861,9 +931,9 @@ class LocalTrade():
:return: array of Order objects
"""
return [o for o in self.orders if ((o.ft_order_side == order_side) or (order_side is None))
and o.ft_is_open is False and
(o.filled or 0) > 0 and
o.status in NON_OPEN_EXCHANGE_STATES]
and o.ft_is_open is False
and o.filled
and o.status in NON_OPEN_EXCHANGE_STATES]
def select_filled_or_open_orders(self) -> List['Order']:
"""
@ -975,6 +1045,16 @@ class LocalTrade():
"""
return Trade.get_trades_proxy(is_open=True)
@staticmethod
def get_open_trade_count() -> int:
"""
get open trade count
"""
if Trade.use_db:
return Trade.query.filter(Trade.is_open.is_(True)).count()
else:
return len(LocalTrade.trades_open)
@staticmethod
def stoploss_reinitialization(desired_stoploss):
"""
@ -1028,6 +1108,7 @@ class Trade(_DECL_BASE, LocalTrade):
open_trade_value = Column(Float)
close_rate: Optional[float] = Column(Float)
close_rate_requested = Column(Float)
realized_profit = Column(Float, default=0.0)
close_profit = Column(Float)
close_profit_abs = Column(Float)
stake_amount = Column(Float, nullable=False)
@ -1059,6 +1140,10 @@ class Trade(_DECL_BASE, LocalTrade):
timeframe = Column(Integer, nullable=True)
trading_mode = Column(Enum(TradingMode), nullable=True)
amount_precision = Column(Float, nullable=True)
price_precision = Column(Float, nullable=True)
precision_mode = Column(Integer, nullable=True)
contract_size = Column(Float, nullable=True)
# Leverage trading properties
leverage = Column(Float, nullable=True, default=1.0)
@ -1073,6 +1158,7 @@ class Trade(_DECL_BASE, LocalTrade):
def __init__(self, **kwargs):
super().__init__(**kwargs)
self.realized_profit = 0
self.recalc_open_trade_value()
def delete(self) -> None:
@ -1087,6 +1173,10 @@ class Trade(_DECL_BASE, LocalTrade):
def commit():
Trade.query.session.commit()
@staticmethod
def rollback():
Trade.query.session.rollback()
@staticmethod
def get_trades_proxy(*, pair: str = None, is_open: bool = None,
open_date: datetime = None, close_date: datetime = None,

View File

@ -8,11 +8,11 @@ from typing import Any, Dict, List, Optional
import arrow
from pandas import DataFrame
from freqtrade.configuration import PeriodicCache
from freqtrade.constants import ListPairsWithTimeframes
from freqtrade.exceptions import OperationalException
from freqtrade.misc import plural
from freqtrade.plugins.pairlist.IPairList import IPairList
from freqtrade.util import PeriodicCache
logger = logging.getLogger(__name__)

View File

@ -51,6 +51,11 @@ class PrecisionFilter(IPairList):
:param ticker: ticker dict as returned from ccxt.fetch_tickers()
:return: True if the pair can stay, false if it should be removed
"""
if ticker.get('last', None) is None:
self.log_once(f"Removed {ticker['symbol']} from whitelist, because "
"ticker['last'] is empty (Usually no trade in the last 24h).",
logger.info)
return False
stop_price = ticker['last'] * self._stoploss
# Adjust stop-prices to precision

View File

@ -4,14 +4,14 @@ Volume PairList provider
Provides dynamic pair list based on trade volumes
"""
import logging
from datetime import datetime, timedelta, timezone
from typing import Any, Dict, List
import arrow
from cachetools import TTLCache
from freqtrade.constants import ListPairsWithTimeframes
from freqtrade.exceptions import OperationalException
from freqtrade.exchange import timeframe_to_minutes
from freqtrade.exchange import timeframe_to_minutes, timeframe_to_prev_date
from freqtrade.misc import format_ms_time
from freqtrade.plugins.pairlist.IPairList import IPairList
@ -73,7 +73,7 @@ class VolumePairList(IPairList):
if (not self._use_range and not (
self._exchange.exchange_has('fetchTickers')
and self._exchange._ft_has["tickers_have_quoteVolume"])):
and self._exchange.get_option("tickers_have_quoteVolume"))):
raise OperationalException(
"Exchange does not support dynamic whitelist in this configuration. "
"Please edit your config and either remove Volumepairlist, "
@ -158,16 +158,16 @@ class VolumePairList(IPairList):
filtered_tickers: List[Dict[str, Any]] = [{'symbol': k} for k in pairlist]
# get lookback period in ms, for exchange ohlcv fetch
since_ms = int(arrow.utcnow()
.floor('minute')
.shift(minutes=-(self._lookback_period * self._tf_in_min)
- self._tf_in_min)
.int_timestamp) * 1000
since_ms = int(timeframe_to_prev_date(
self._lookback_timeframe,
datetime.now(timezone.utc) + timedelta(
minutes=-(self._lookback_period * self._tf_in_min) - self._tf_in_min)
).timestamp()) * 1000
to_ms = int(arrow.utcnow()
.floor('minute')
.shift(minutes=-self._tf_in_min)
.int_timestamp) * 1000
to_ms = int(timeframe_to_prev_date(
self._lookback_timeframe,
datetime.now(timezone.utc) - timedelta(minutes=self._tf_in_min)
).timestamp()) * 1000
# todo: utc date output for starting date
self.log_once(f"Using volume range of {self._lookback_period} candles, timeframe: "
@ -193,7 +193,7 @@ class VolumePairList(IPairList):
) in candles else None
# in case of candle data calculate typical price and quoteVolume for candle
if pair_candles is not None and not pair_candles.empty:
if self._exchange._ft_has["ohlcv_volume_currency"] == "base":
if self._exchange.get_option("ohlcv_volume_currency") == "base":
pair_candles['typical_price'] = (pair_candles['high'] + pair_candles['low']
+ pair_candles['close']) / 3

View File

@ -1,5 +1,5 @@
import re
from typing import List
from typing import Any, Dict, List
def expand_pairlist(wildcardpl: List[str], available_pairs: List[str],
@ -40,3 +40,13 @@ def expand_pairlist(wildcardpl: List[str], available_pairs: List[str],
except re.error as err:
raise ValueError(f"Wildcard error in {pair_wc}, {err}")
return result
def dynamic_expand_pairlist(config: Dict[str, Any], markets: List[str]) -> List[str]:
expanded_pairs = expand_pairlist(config['pairs'], markets)
if config.get('freqai', {}).get('enabled', False):
corr_pairlist = config['freqai']['feature_parameters']['include_corr_pairlist']
expanded_pairs += [pair for pair in corr_pairlist
if pair not in config['pairs']]
return expanded_pairs

View File

@ -49,7 +49,7 @@ class StoplossGuard(IProtection):
trades1 = Trade.get_trades_proxy(pair=pair, is_open=False, close_date=look_back_until)
trades = [trade for trade in trades1 if (str(trade.exit_reason) in (
ExitType.TRAILING_STOP_LOSS.value, ExitType.STOP_LOSS.value,
ExitType.STOPLOSS_ON_EXCHANGE.value)
ExitType.STOPLOSS_ON_EXCHANGE.value, ExitType.LIQUIDATION.value)
and trade.close_profit and trade.close_profit < self._profit_limit)]
if self._only_per_side:

View File

@ -0,0 +1,57 @@
# pragma pylint: disable=attribute-defined-outside-init
"""
This module load a custom model for freqai
"""
import logging
from pathlib import Path
from typing import Dict
from freqtrade.constants import USERPATH_FREQAIMODELS
from freqtrade.exceptions import OperationalException
from freqtrade.freqai.freqai_interface import IFreqaiModel
from freqtrade.resolvers import IResolver
logger = logging.getLogger(__name__)
class FreqaiModelResolver(IResolver):
"""
This class contains all the logic to load custom hyperopt loss class
"""
object_type = IFreqaiModel
object_type_str = "FreqaiModel"
user_subdir = USERPATH_FREQAIMODELS
initial_search_path = (
Path(__file__).parent.parent.joinpath("freqai/prediction_models").resolve()
)
@staticmethod
def load_freqaimodel(config: Dict) -> IFreqaiModel:
"""
Load the custom class from config parameter
:param config: configuration dictionary
"""
disallowed_models = ["BaseRegressionModel", "BaseTensorFlowModel"]
freqaimodel_name = config.get("freqaimodel")
if not freqaimodel_name:
raise OperationalException(
"No freqaimodel set. Please use `--freqaimodel` to "
"specify the FreqaiModel class to use.\n"
)
if freqaimodel_name in disallowed_models:
raise OperationalException(
f"{freqaimodel_name} is a baseclass and cannot be used directly. Please choose "
"an existing child class or inherit from this baseclass.\n"
)
freqaimodel = FreqaiModelResolver.load_object(
freqaimodel_name,
config,
kwargs={"config": config},
extra_dir=config.get("freqaimodel_path"),
)
return freqaimodel

View File

@ -193,7 +193,10 @@ class IResolver:
:return: List of dicts containing 'name', 'class' and 'location' entries
"""
logger.debug(f"Searching for {cls.object_type.__name__} '{directory}'")
objects = []
objects: List[Dict[str, Any]] = []
if not directory.is_dir():
logger.info(f"'{directory}' is not a directory, skipping.")
return objects
for entry in directory.iterdir():
if (
recursive and entry.is_dir()

View File

@ -194,11 +194,11 @@ class OrderSchema(BaseModel):
pair: str
order_id: str
status: str
remaining: float
remaining: Optional[float]
amount: float
safe_price: float
cost: float
filled: float
filled: Optional[float]
ft_order_side: str
order_type: str
is_open: bool
@ -325,11 +325,13 @@ class ForceEnterPayload(BaseModel):
ordertype: Optional[OrderTypeValues]
stakeamount: Optional[float]
entry_tag: Optional[str]
leverage: Optional[float]
class ForceExitPayload(BaseModel):
tradeid: str
ordertype: Optional[OrderTypeValues]
amount: Optional[float]
class BlacklistPayload(BaseModel):

View File

@ -37,7 +37,8 @@ logger = logging.getLogger(__name__)
# 2.14: Add entry/exit orders to trade response
# 2.15: Add backtest history endpoints
# 2.16: Additional daily metrics
API_VERSION = 2.16
# 2.17: Forceentry - leverage, partial force_exit
API_VERSION = 2.17
# Public API, requires no auth.
router_public = APIRouter()
@ -142,12 +143,11 @@ def show_config(rpc: Optional[RPC] = Depends(get_rpc_optional), config=Depends(g
@router.post('/forcebuy', response_model=ForceEnterResponse, tags=['trading'])
def force_entry(payload: ForceEnterPayload, rpc: RPC = Depends(get_rpc)):
ordertype = payload.ordertype.value if payload.ordertype else None
stake_amount = payload.stakeamount if payload.stakeamount else None
entry_tag = payload.entry_tag if payload.entry_tag else 'force_entry'
trade = rpc._rpc_force_entry(payload.pair, payload.price, order_side=payload.side,
order_type=ordertype, stake_amount=stake_amount,
enter_tag=entry_tag)
order_type=ordertype, stake_amount=payload.stakeamount,
enter_tag=payload.entry_tag or 'force_entry',
leverage=payload.leverage)
if trade:
return ForceEnterResponse.parse_obj(trade.to_json())
@ -161,7 +161,7 @@ def force_entry(payload: ForceEnterPayload, rpc: RPC = Depends(get_rpc)):
@router.post('/forcesell', response_model=ResultMsg, tags=['trading'])
def forceexit(payload: ForceExitPayload, rpc: RPC = Depends(get_rpc)):
ordertype = payload.ordertype.value if payload.ordertype else None
return rpc._rpc_force_exit(payload.tradeid, ordertype)
return rpc._rpc_force_exit(payload.tradeid, ordertype, amount=payload.amount)
@router.get('/blacklist', response_model=BlacklistResponse, tags=['info', 'pairlist'])
@ -216,9 +216,10 @@ def stop(rpc: RPC = Depends(get_rpc)):
return rpc._rpc_stop()
@router.post('/stopentry', response_model=StatusMsg, tags=['botcontrol'])
@router.post('/stopbuy', response_model=StatusMsg, tags=['botcontrol'])
def stop_buy(rpc: RPC = Depends(get_rpc)):
return rpc._rpc_stopbuy()
return rpc._rpc_stopentry()
@router.post('/reload_config', response_model=StatusMsg, tags=['botcontrol'])

View File

@ -18,9 +18,9 @@ def get_rpc_optional() -> Optional[RPC]:
def get_rpc() -> Optional[Iterator[RPC]]:
_rpc = get_rpc_optional()
if _rpc:
Trade.query.session.rollback()
Trade.rollback()
yield _rpc
Trade.query.session.rollback()
Trade.rollback()
else:
raise RPCException('Bot is not in the correct state')

View File

@ -1,4 +1,5 @@
from pathlib import Path
from typing import Optional
from fastapi import APIRouter
from fastapi.exceptions import HTTPException
@ -50,8 +51,12 @@ async def index_html(rest_of_path: str):
filename = uibase / rest_of_path
# It's security relevant to check "relative_to".
# Without this, Directory-traversal is possible.
media_type: Optional[str] = None
if filename.suffix == '.js':
# Force text/javascript for .js files - Circumvent faulty system configuration
media_type = 'application/javascript'
if filename.is_file() and is_relative_to(filename, uibase):
return FileResponse(str(filename))
return FileResponse(str(filename), media_type=media_type)
index_file = uibase / 'index.html'
if not index_file.is_file():

View File

@ -12,6 +12,7 @@ from pycoingecko import CoinGeckoAPI
from requests.exceptions import RequestException
from freqtrade.constants import SUPPORTED_FIAT
from freqtrade.mixins.logging_mixin import LoggingMixin
logger = logging.getLogger(__name__)
@ -27,7 +28,7 @@ coingecko_mapping = {
}
class CryptoToFiatConverter:
class CryptoToFiatConverter(LoggingMixin):
"""
Main class to initiate Crypto to FIAT.
This object contains a list of pair Crypto, FIAT
@ -54,6 +55,7 @@ class CryptoToFiatConverter:
# Timeout: 6h
self._pair_price: TTLCache = TTLCache(maxsize=500, ttl=6 * 60 * 60)
LoggingMixin.__init__(self, logger, 3600)
self._load_cryptomap()
def _load_cryptomap(self) -> None:
@ -177,7 +179,9 @@ class CryptoToFiatConverter:
if not _gekko_id:
# return 0 for unsupported stake currencies (fiat-convert should not break the bot)
logger.warning("unsupported crypto-symbol %s - returning 0.0", crypto_symbol)
self.log_once(
f"unsupported crypto-symbol {crypto_symbol.upper()} - returning 0.0",
logger.warning)
return 0.0
try:

View File

@ -179,8 +179,10 @@ class RPC:
else:
current_rate = trade.close_rate
if len(trade.select_filled_orders(trade.entry_side)) > 0:
current_profit = trade.calc_profit_ratio(current_rate)
current_profit_abs = trade.calc_profit(current_rate)
current_profit = trade.calc_profit_ratio(
current_rate) if not isnan(current_rate) else NAN
current_profit_abs = trade.calc_profit(
current_rate) if not isnan(current_rate) else NAN
current_profit_fiat: Optional[float] = None
# Calculate fiat profit
if self._fiat_converter:
@ -201,7 +203,7 @@ class RPC:
trade_dict = trade.to_json()
trade_dict.update(dict(
close_profit=trade.close_profit if trade.close_profit is not None else None,
close_profit=trade.close_profit if not trade.is_open else None,
current_rate=current_rate,
current_profit=current_profit, # Deprecated
current_profit_pct=round(current_profit * 100, 2), # Deprecated
@ -239,7 +241,10 @@ class RPC:
trade.pair, side='exit', is_short=trade.is_short, refresh=False)
except (PricingError, ExchangeError):
current_rate = NAN
if len(trade.select_filled_orders(trade.entry_side)) > 0:
trade_profit = NAN
profit_str = f'{NAN:.2%}'
else:
if trade.nr_of_successful_entries > 0:
trade_profit = trade.calc_profit(current_rate)
profit_str = f'{trade.calc_profit_ratio(current_rate):.2%}'
else:
@ -424,21 +429,20 @@ class RPC:
for trade in trades:
current_rate: float = 0.0
if not trade.open_rate:
continue
if trade.close_date:
durations.append((trade.close_date - trade.open_date).total_seconds())
if not trade.is_open:
profit_ratio = trade.close_profit
profit_closed_coin.append(trade.close_profit_abs)
profit_abs = trade.close_profit_abs
profit_closed_coin.append(profit_abs)
profit_closed_ratio.append(profit_ratio)
if trade.close_profit >= 0:
winning_trades += 1
winning_profit += trade.close_profit_abs
winning_profit += profit_abs
else:
losing_trades += 1
losing_profit += trade.close_profit_abs
losing_profit += profit_abs
else:
# Get current rate
try:
@ -446,11 +450,15 @@ class RPC:
trade.pair, side='exit', is_short=trade.is_short, refresh=False)
except (PricingError, ExchangeError):
current_rate = NAN
if isnan(current_rate):
profit_ratio = NAN
profit_abs = NAN
else:
profit_ratio = trade.calc_profit_ratio(rate=current_rate)
profit_abs = trade.calc_profit(
rate=trade.close_rate or current_rate) + trade.realized_profit
profit_all_coin.append(
trade.calc_profit(rate=trade.close_rate or current_rate)
)
profit_all_coin.append(profit_abs)
profit_all_ratio.append(profit_ratio)
best_pair = Trade.get_best_pair(start_date)
@ -649,7 +657,7 @@ class RPC:
self._freqtrade.state = State.RELOAD_CONFIG
return {'status': 'Reloading config ...'}
def _rpc_stopbuy(self) -> Dict[str, str]:
def _rpc_stopentry(self) -> Dict[str, str]:
"""
Handler to stop buying, but handle open trades gracefully.
"""
@ -657,14 +665,10 @@ class RPC:
# Set 'max_open_trades' to 0
self._freqtrade.config['max_open_trades'] = 0
return {'status': 'No more buy will occur from now. Run /reload_config to reset.'}
return {'status': 'No more entries will occur from now. Run /reload_config to reset.'}
def _rpc_force_exit(self, trade_id: str, ordertype: Optional[str] = None) -> Dict[str, str]:
"""
Handler for forceexit <id>.
Sells the given trade at current price
"""
def _exec_force_exit(trade: Trade) -> None:
def __exec_force_exit(self, trade: Trade, ordertype: Optional[str],
amount: Optional[float] = None) -> None:
# Check if there is there is an open order
fully_canceled = False
if trade.open_order_id:
@ -685,10 +689,26 @@ class RPC:
exit_check = ExitCheckTuple(exit_type=ExitType.FORCE_EXIT)
order_type = ordertype or self._freqtrade.strategy.order_types.get(
"force_exit", self._freqtrade.strategy.order_types["exit"])
sub_amount: Optional[float] = None
if amount and amount < trade.amount:
# Partial exit ...
min_exit_stake = self._freqtrade.exchange.get_min_pair_stake_amount(
trade.pair, current_rate, trade.stop_loss_pct)
remaining = (trade.amount - amount) * current_rate
if remaining < min_exit_stake:
raise RPCException(f'Remaining amount of {remaining} would be too small.')
sub_amount = amount
self._freqtrade.execute_trade_exit(
trade, current_rate, exit_check, ordertype=order_type)
# ---- EOF def _exec_forcesell ----
trade, current_rate, exit_check, ordertype=order_type,
sub_trade_amt=sub_amount)
def _rpc_force_exit(self, trade_id: str, ordertype: Optional[str] = None, *,
amount: Optional[float] = None) -> Dict[str, str]:
"""
Handler for forceexit <id>.
Sells the given trade at current price
"""
if self._freqtrade.state != State.RUNNING:
raise RPCException('trader is not running')
@ -697,7 +717,7 @@ class RPC:
if trade_id == 'all':
# Execute sell for all open orders
for trade in Trade.get_open_trades():
_exec_force_exit(trade)
self.__exec_force_exit(trade, ordertype)
Trade.commit()
self._freqtrade.wallets.update()
return {'result': 'Created sell orders for all open trades.'}
@ -710,7 +730,7 @@ class RPC:
logger.warning('force_exit: Invalid argument received')
raise RPCException('invalid argument')
_exec_force_exit(trade)
self.__exec_force_exit(trade, ordertype, amount)
Trade.commit()
self._freqtrade.wallets.update()
return {'result': f'Created sell order for trade {trade_id}.'}
@ -719,7 +739,8 @@ class RPC:
order_type: Optional[str] = None,
order_side: SignalDirection = SignalDirection.LONG,
stake_amount: Optional[float] = None,
enter_tag: Optional[str] = 'force_entry') -> Optional[Trade]:
enter_tag: Optional[str] = 'force_entry',
leverage: Optional[float] = None) -> Optional[Trade]:
"""
Handler for forcebuy <asset> <price>
Buys a pair trade at the given or current price
@ -761,6 +782,7 @@ class RPC:
ordertype=order_type, trade=trade,
is_short=is_short,
enter_tag=enter_tag,
leverage_=leverage,
):
Trade.commit()
trade = Trade.get_trades([Trade.is_open.is_(True), Trade.pair == pair]).first()
@ -875,7 +897,7 @@ class RPC:
lock.active = False
lock.lock_end_time = datetime.now(timezone.utc)
PairLock.query.session.commit()
Trade.commit()
return self._rpc_locks()

View File

@ -2,6 +2,7 @@
This module contains class to manage RPC communications (Telegram, API, ...)
"""
import logging
from collections import deque
from typing import Any, Dict, List
from freqtrade.enums import RPCMessageType
@ -77,6 +78,17 @@ class RPCManager:
except NotImplementedError:
logger.error(f"Message type '{msg['type']}' not implemented by handler {mod.name}.")
def process_msg_queue(self, queue: deque) -> None:
"""
Process all messages in the queue.
"""
while queue:
msg = queue.popleft()
self.send_msg({
'type': RPCMessageType.STRATEGY_MSG,
'msg': msg,
})
def startup_messages(self, config: Dict[str, Any], pairlist, protections) -> None:
if config['dry_run']:
self.send_msg({

View File

@ -16,8 +16,8 @@ from typing import Any, Callable, Dict, List, Optional, Union
import arrow
from tabulate import tabulate
from telegram import (CallbackQuery, InlineKeyboardButton, InlineKeyboardMarkup, KeyboardButton,
ParseMode, ReplyKeyboardMarkup, Update)
from telegram import (MAX_MESSAGE_LENGTH, CallbackQuery, InlineKeyboardButton, InlineKeyboardMarkup,
KeyboardButton, ParseMode, ReplyKeyboardMarkup, Update)
from telegram.error import BadRequest, NetworkError, TelegramError
from telegram.ext import CallbackContext, CallbackQueryHandler, CommandHandler, Updater
from telegram.utils.helpers import escape_markdown
@ -35,8 +35,6 @@ logger = logging.getLogger(__name__)
logger.debug('Included module rpc.telegram ...')
MAX_TELEGRAM_MESSAGE_LENGTH = 4096
@dataclass
class TimeunitMappings:
@ -72,7 +70,7 @@ def authorized_only(command_handler: Callable[..., None]) -> Callable[..., Any]:
)
return wrapper
# Rollback session to avoid getting data stored in a transaction.
Trade.query.session.rollback()
Trade.rollback()
logger.debug(
'Executing handler: %s for chat_id: %s',
command_handler.__name__,
@ -116,17 +114,20 @@ class Telegram(RPCHandler):
# TODO: DRY! - its not good to list all valid cmds here. But otherwise
# this needs refactoring of the whole telegram module (same
# problem in _help()).
valid_keys: List[str] = [r'/start$', r'/stop$', r'/status$', r'/status table$',
valid_keys: List[str] = [
r'/start$', r'/stop$', r'/status$', r'/status table$',
r'/trades$', r'/performance$', r'/buys', r'/entries',
r'/sells', r'/exits', r'/mix_tags',
r'/daily$', r'/daily \d+$', r'/profit$', r'/profit \d+',
r'/stats$', r'/count$', r'/locks$', r'/balance$',
r'/stopbuy$', r'/reload_config$', r'/show_config$',
r'/logs$', r'/whitelist$', r'/blacklist$', r'/bl_delete$',
r'/stopbuy$', r'/stopentry$', r'/reload_config$', r'/show_config$',
r'/logs$', r'/whitelist$', r'/whitelist(\ssorted|\sbaseonly)+$',
r'/blacklist$', r'/bl_delete$',
r'/weekly$', r'/weekly \d+$', r'/monthly$', r'/monthly \d+$',
r'/forcebuy$', r'/forcelong$', r'/forceshort$',
r'/forcesell$', r'/forceexit$',
r'/edge$', r'/health$', r'/help$', r'/version$']
r'/edge$', r'/health$', r'/help$', r'/version$'
]
# Create keys for generation
valid_keys_print = [k.replace('$', '') for k in valid_keys]
@ -183,7 +184,7 @@ class Telegram(RPCHandler):
CommandHandler(['unlock', 'delete_locks'], self._delete_locks),
CommandHandler(['reload_config', 'reload_conf'], self._reload_config),
CommandHandler(['show_config', 'show_conf'], self._show_config),
CommandHandler('stopbuy', self._stopbuy),
CommandHandler(['stopbuy', 'stopentry'], self._stopentry),
CommandHandler('whitelist', self._whitelist),
CommandHandler('blacklist', self._blacklist),
CommandHandler(['blacklist_delete', 'bl_delete'], self._blacklist_delete),
@ -315,20 +316,36 @@ class Telegram(RPCHandler):
msg['profit_fiat'] = self._rpc._fiat_converter.convert_amount(
msg['profit_amount'], msg['stake_currency'], msg['fiat_currency'])
msg['profit_extra'] = (
f" ({msg['gain']}: {msg['profit_amount']:.8f} {msg['stake_currency']}"
f" / {msg['profit_fiat']:.3f} {msg['fiat_currency']})")
f" / {msg['profit_fiat']:.3f} {msg['fiat_currency']}")
else:
msg['profit_extra'] = ''
msg['profit_extra'] = (
f" ({msg['gain']}: {msg['profit_amount']:.8f} {msg['stake_currency']}"
f"{msg['profit_extra']})")
is_fill = msg['type'] == RPCMessageType.EXIT_FILL
is_sub_trade = msg.get('sub_trade')
is_sub_profit = msg['profit_amount'] != msg.get('cumulative_profit')
profit_prefix = ('Sub ' if is_sub_profit
else 'Cumulative ') if is_sub_trade else ''
cp_extra = ''
if is_sub_profit and is_sub_trade:
if self._rpc._fiat_converter:
cp_fiat = self._rpc._fiat_converter.convert_amount(
msg['cumulative_profit'], msg['stake_currency'], msg['fiat_currency'])
cp_extra = f" / {cp_fiat:.3f} {msg['fiat_currency']}"
else:
cp_extra = ''
cp_extra = f"*Cumulative Profit:* (`{msg['cumulative_profit']:.8f} " \
f"{msg['stake_currency']}{cp_extra}`)\n"
message = (
f"{msg['emoji']} *{self._exchange_from_msg(msg)}:* "
f"{'Exited' if is_fill else 'Exiting'} {msg['pair']} (#{msg['trade_id']})\n"
f"{self._add_analyzed_candle(msg['pair'])}"
f"*{'Profit' if is_fill else 'Unrealized Profit'}:* "
f"*{f'{profit_prefix}Profit' if is_fill else f'Unrealized {profit_prefix}Profit'}:* "
f"`{msg['profit_ratio']:.2%}{msg['profit_extra']}`\n"
f"{cp_extra}"
f"*Enter Tag:* `{msg['enter_tag']}`\n"
f"*Exit Reason:* `{msg['exit_reason']}`\n"
f"*Duration:* `{msg['duration']} ({msg['duration_min']:.1f} min)`\n"
f"*Direction:* `{msg['direction']}`\n"
f"{msg['leverage_text']}"
f"*Amount:* `{msg['amount']:.8f}`\n"
@ -336,11 +353,25 @@ class Telegram(RPCHandler):
)
if msg['type'] == RPCMessageType.EXIT:
message += (f"*Current Rate:* `{msg['current_rate']:.8f}`\n"
f"*Close Rate:* `{msg['limit']:.8f}`")
f"*Exit Rate:* `{msg['limit']:.8f}`")
elif msg['type'] == RPCMessageType.EXIT_FILL:
message += f"*Close Rate:* `{msg['close_rate']:.8f}`"
message += f"*Exit Rate:* `{msg['close_rate']:.8f}`"
if msg.get('sub_trade'):
if self._rpc._fiat_converter:
msg['stake_amount_fiat'] = self._rpc._fiat_converter.convert_amount(
msg['stake_amount'], msg['stake_currency'], msg['fiat_currency'])
else:
msg['stake_amount_fiat'] = 0
rem = round_coin_value(msg['stake_amount'], msg['stake_currency'])
message += f"\n*Remaining:* `({rem}"
if msg.get('fiat_currency', None):
message += f", {round_coin_value(msg['stake_amount_fiat'], msg['fiat_currency'])}"
message += ")`"
else:
message += f"\n*Duration:* `{msg['duration']} ({msg['duration_min']:.1f} min)`"
return message
def compose_message(self, msg: Dict[str, Any], msg_type: RPCMessageType) -> str:
@ -353,7 +384,8 @@ class Telegram(RPCHandler):
elif msg_type in (RPCMessageType.ENTRY_CANCEL, RPCMessageType.EXIT_CANCEL):
msg['message_side'] = 'enter' if msg_type in [RPCMessageType.ENTRY_CANCEL] else 'exit'
message = (f"\N{WARNING SIGN} *{self._exchange_from_msg(msg)}:* "
f"Cancelling {msg['message_side']} Order for {msg['pair']} "
f"Cancelling {'partial ' if msg.get('sub_trade') else ''}"
f"{msg['message_side']} Order for {msg['pair']} "
f"(#{msg['trade_id']}). Reason: {msg['reason']}.")
elif msg_type == RPCMessageType.PROTECTION_TRIGGER:
@ -376,7 +408,8 @@ class Telegram(RPCHandler):
elif msg_type == RPCMessageType.STARTUP:
message = f"{msg['status']}"
elif msg_type == RPCMessageType.STRATEGY_MSG:
message = f"{msg['msg']}"
else:
raise NotImplementedError(f"Unknown message type: {msg_type}")
return message
@ -423,54 +456,63 @@ class Telegram(RPCHandler):
else:
return "\N{CROSS MARK}"
def _prepare_entry_details(self, filled_orders: List, quote_currency: str, is_open: bool):
def _prepare_order_details(self, filled_orders: List, quote_currency: str, is_open: bool):
"""
Prepare details of trade with entry adjustment enabled
"""
lines: List[str] = []
lines_detail: List[str] = []
if len(filled_orders) > 0:
first_avg = filled_orders[0]["safe_price"]
for x, order in enumerate(filled_orders):
if not order['ft_is_entry'] or order['is_open'] is True:
lines: List[str] = []
if order['is_open'] is True:
continue
wording = 'Entry' if order['ft_is_entry'] else 'Exit'
cur_entry_datetime = arrow.get(order["order_filled_date"])
cur_entry_amount = order["amount"]
cur_entry_amount = order["filled"] or order["amount"]
cur_entry_average = order["safe_price"]
lines.append(" ")
if x == 0:
lines.append(f"*Entry #{x+1}:*")
lines.append(f"*{wording} #{x+1}:*")
lines.append(
f"*Entry Amount:* {cur_entry_amount} ({order['cost']:.8f} {quote_currency})")
lines.append(f"*Average Entry Price:* {cur_entry_average}")
f"*Amount:* {cur_entry_amount} ({order['cost']:.8f} {quote_currency})")
lines.append(f"*Average Price:* {cur_entry_average}")
else:
sumA = 0
sumB = 0
for y in range(x):
sumA += (filled_orders[y]["amount"] * filled_orders[y]["safe_price"])
sumB += filled_orders[y]["amount"]
amount = filled_orders[y]["filled"] or filled_orders[y]["amount"]
sumA += amount * filled_orders[y]["safe_price"]
sumB += amount
prev_avg_price = sumA / sumB
# TODO: This calculation ignores fees.
price_to_1st_entry = ((cur_entry_average - first_avg) / first_avg)
minus_on_entry = 0
if prev_avg_price:
minus_on_entry = (cur_entry_average - prev_avg_price) / prev_avg_price
dur_entry = cur_entry_datetime - arrow.get(
filled_orders[x - 1]["order_filled_date"])
days = dur_entry.days
hours, remainder = divmod(dur_entry.seconds, 3600)
minutes, seconds = divmod(remainder, 60)
lines.append(f"*Entry #{x+1}:* at {minus_on_entry:.2%} avg profit")
lines.append(f"*{wording} #{x+1}:* at {minus_on_entry:.2%} avg profit")
if is_open:
lines.append("({})".format(cur_entry_datetime
.humanize(granularity=["day", "hour", "minute"])))
lines.append(
f"*Entry Amount:* {cur_entry_amount} ({order['cost']:.8f} {quote_currency})")
lines.append(f"*Average Entry Price:* {cur_entry_average} "
f"*Amount:* {cur_entry_amount} ({order['cost']:.8f} {quote_currency})")
lines.append(f"*Average {wording} Price:* {cur_entry_average} "
f"({price_to_1st_entry:.2%} from 1st entry rate)")
lines.append(f"*Order filled at:* {order['order_filled_date']}")
lines.append(f"({days}d {hours}h {minutes}m {seconds}s from previous entry)")
return lines
lines.append(f"*Order filled:* {order['order_filled_date']}")
# TODO: is this really useful?
# dur_entry = cur_entry_datetime - arrow.get(
# filled_orders[x - 1]["order_filled_date"])
# days = dur_entry.days
# hours, remainder = divmod(dur_entry.seconds, 3600)
# minutes, seconds = divmod(remainder, 60)
# lines.append(
# f"({days}d {hours}h {minutes}m {seconds}s from previous {wording.lower()})")
lines_detail.append("\n".join(lines))
return lines_detail
@authorized_only
def _status(self, update: Update, context: CallbackContext) -> None:
@ -485,7 +527,14 @@ class Telegram(RPCHandler):
if context.args and 'table' in context.args:
self._status_table(update, context)
return
else:
self._status_msg(update, context)
def _status_msg(self, update: Update, context: CallbackContext) -> None:
"""
handler for `/status` and `/status <id>`.
"""
try:
# Check if there's at least one numerical ID provided.
@ -497,7 +546,6 @@ class Telegram(RPCHandler):
results = self._rpc._rpc_trade_status(trade_ids=trade_ids)
position_adjust = self._config.get('position_adjustment_enable', False)
max_entries = self._config.get('max_entry_position_adjustment', -1)
messages = []
for r in results:
r['open_date_hum'] = arrow.get(r['open_date']).humanize()
r['num_entries'] = len([o for o in r['orders'] if o['ft_is_entry']])
@ -528,6 +576,8 @@ class Telegram(RPCHandler):
])
if r['is_open']:
if r.get('realized_profit'):
lines.append("*Realized Profit:* `{realized_profit:.8f}`")
if (r['stop_loss_abs'] != r['initial_stop_loss_abs']
and r['initial_stop_loss_ratio'] is not None):
# Adding initial stoploss only if it is different from stoploss
@ -540,24 +590,34 @@ class Telegram(RPCHandler):
lines.append("*Stoploss distance:* `{stoploss_current_dist:.8f}` "
"`({stoploss_current_dist_ratio:.2%})`")
if r['open_order']:
if r['exit_order_status']:
lines.append("*Open Order:* `{open_order}` - `{exit_order_status}`")
else:
lines.append("*Open Order:* `{open_order}`")
lines.append(
"*Open Order:* `{open_order}`"
+ "- `{exit_order_status}`" if r['exit_order_status'] else "")
lines_detail = self._prepare_entry_details(
lines_detail = self._prepare_order_details(
r['orders'], r['quote_currency'], r['is_open'])
lines.extend(lines_detail if lines_detail else "")
# Filter empty lines using list-comprehension
messages.append("\n".join([line for line in lines if line]).format(**r))
for msg in messages:
self._send_msg(msg)
self.__send_status_msg(lines, r)
except RPCException as e:
self._send_msg(str(e))
def __send_status_msg(self, lines: List[str], r: Dict[str, Any]) -> None:
"""
Send status message.
"""
msg = ''
for line in lines:
if line:
if (len(msg) + len(line) + 1) < MAX_MESSAGE_LENGTH:
msg += line + '\n'
else:
self._send_msg(msg.format(**r))
msg = "*Trade ID:* `{trade_id}` - continued\n" + line + '\n'
self._send_msg(msg.format(**r))
@authorized_only
def _status_table(self, update: Update, context: CallbackContext) -> None:
"""
@ -860,7 +920,7 @@ class Telegram(RPCHandler):
total_dust_currencies += 1
# Handle overflowing message length
if len(output + curr_output) >= MAX_TELEGRAM_MESSAGE_LENGTH:
if len(output + curr_output) >= MAX_MESSAGE_LENGTH:
self._send_msg(output)
output = curr_output
else:
@ -926,7 +986,7 @@ class Telegram(RPCHandler):
self._send_msg(f"Status: `{msg['status']}`")
@authorized_only
def _stopbuy(self, update: Update, context: CallbackContext) -> None:
def _stopentry(self, update: Update, context: CallbackContext) -> None:
"""
Handler for /stop_buy.
Sets max_open_trades to 0 and gracefully sells all open trades
@ -934,7 +994,7 @@ class Telegram(RPCHandler):
:param update: message update
:return: None
"""
msg = self._rpc._rpc_stopbuy()
msg = self._rpc._rpc_stopentry()
self._send_msg(f"Status: `{msg['status']}`")
@authorized_only
@ -1123,7 +1183,7 @@ class Telegram(RPCHandler):
f"({trade['profit_ratio']:.2%}) "
f"({trade['count']})</code>\n")
if len(output + stat_line) >= MAX_TELEGRAM_MESSAGE_LENGTH:
if len(output + stat_line) >= MAX_MESSAGE_LENGTH:
self._send_msg(output, parse_mode=ParseMode.HTML)
output = stat_line
else:
@ -1158,7 +1218,7 @@ class Telegram(RPCHandler):
f"({trade['profit_ratio']:.2%}) "
f"({trade['count']})</code>\n")
if len(output + stat_line) >= MAX_TELEGRAM_MESSAGE_LENGTH:
if len(output + stat_line) >= MAX_MESSAGE_LENGTH:
self._send_msg(output, parse_mode=ParseMode.HTML)
output = stat_line
else:
@ -1193,7 +1253,7 @@ class Telegram(RPCHandler):
f"({trade['profit_ratio']:.2%}) "
f"({trade['count']})</code>\n")
if len(output + stat_line) >= MAX_TELEGRAM_MESSAGE_LENGTH:
if len(output + stat_line) >= MAX_MESSAGE_LENGTH:
self._send_msg(output, parse_mode=ParseMode.HTML)
output = stat_line
else:
@ -1228,7 +1288,7 @@ class Telegram(RPCHandler):
f"({trade['profit']:.2%}) "
f"({trade['count']})</code>\n")
if len(output + stat_line) >= MAX_TELEGRAM_MESSAGE_LENGTH:
if len(output + stat_line) >= MAX_MESSAGE_LENGTH:
self._send_msg(output, parse_mode=ParseMode.HTML)
output = stat_line
else:
@ -1311,6 +1371,12 @@ class Telegram(RPCHandler):
try:
whitelist = self._rpc._rpc_whitelist()
if context.args:
if "sorted" in context.args:
whitelist['whitelist'] = sorted(whitelist['whitelist'])
if "baseonly" in context.args:
whitelist['whitelist'] = [pair.split("/")[0] for pair in whitelist['whitelist']]
message = f"Using whitelist `{whitelist['method']}` with {whitelist['length']} pairs\n"
message += f"`{', '.join(whitelist['whitelist'])}`"
@ -1367,7 +1433,7 @@ class Telegram(RPCHandler):
escape_markdown(logrec[2], version=2),
escape_markdown(logrec[3], version=2),
escape_markdown(logrec[4], version=2))
if len(msgs + msg) + 10 >= MAX_TELEGRAM_MESSAGE_LENGTH:
if len(msgs + msg) + 10 >= MAX_MESSAGE_LENGTH:
# Send message immediately if it would become too long
self._send_msg(msgs, parse_mode=ParseMode.MARKDOWN_V2)
msgs = msg + '\n'
@ -1424,13 +1490,14 @@ class Telegram(RPCHandler):
"------------\n"
"*/start:* `Starts the trader`\n"
"*/stop:* Stops the trader\n"
"*/stopbuy:* `Stops buying, but handles open trades gracefully` \n"
"*/stopentry:* `Stops entering, but handles open trades gracefully` \n"
"*/forceexit <trade_id>|all:* `Instantly exits the given trade or all trades, "
"regardless of profit`\n"
"*/fx <trade_id>|all:* `Alias to /forceexit`\n"
f"{force_enter_text if self._config.get('force_entry_enable', False) else ''}"
"*/delete <trade_id>:* `Instantly delete the given trade in the database`\n"
"*/whitelist:* `Show current whitelist` \n"
"*/whitelist [sorted] [baseonly]:* `Show current whitelist. Optionally in "
"order and/or only displaying the base currency of each pairing.`\n"
"*/blacklist [pair]:* `Show current blacklist, or adds one or more pairs "
"to the blacklist.` \n"
"*/blacklist_delete [pairs]| /bl_delete [pairs]:* "

View File

@ -145,11 +145,29 @@ class IStrategy(ABC, HyperStrategyMixin):
informative_data.candle_type = config['candle_type_def']
self._ft_informative.append((informative_data, cls_method))
def load_freqAI_model(self) -> None:
if self.config.get('freqai', {}).get('enabled', False):
# Import here to avoid importing this if freqAI is disabled
from freqtrade.resolvers.freqaimodel_resolver import FreqaiModelResolver
self.freqai = FreqaiModelResolver.load_freqaimodel(self.config)
self.freqai_info = self.config["freqai"]
else:
# Gracious failures if freqAI is disabled but "start" is called.
class DummyClass():
def start(self, *args, **kwargs):
raise OperationalException(
'freqAI is not enabled. '
'Please enable it in your config to use this strategy.')
self.freqai = DummyClass() # type: ignore
def ft_bot_start(self, **kwargs) -> None:
"""
Strategy init - runs after dataprovider has been added.
Must call bot_start()
"""
self.load_freqAI_model()
strategy_safe_wrapper(self.bot_start)()
self.ft_load_hyper_params(self.config.get('runmode') == RunMode.HYPEROPT)
@ -463,10 +481,13 @@ class IStrategy(ABC, HyperStrategyMixin):
def adjust_trade_position(self, trade: Trade, current_time: datetime,
current_rate: float, current_profit: float,
min_stake: Optional[float], max_stake: float,
current_entry_rate: float, current_exit_rate: float,
current_entry_profit: float, current_exit_profit: float,
**kwargs) -> Optional[float]:
"""
Custom trade adjustment logic, returning the stake amount that a trade should be increased.
This means extra buy orders with additional fees.
Custom trade adjustment logic, returning the stake amount that a trade should be
increased or decreased.
This means extra buy or sell orders with additional fees.
Only called when `position_adjustment_enable` is set to True.
For full documentation please go to https://www.freqtrade.io/en/latest/strategy-advanced/
@ -477,10 +498,16 @@ class IStrategy(ABC, HyperStrategyMixin):
:param current_time: datetime object, containing the current datetime
:param current_rate: Current buy rate.
:param current_profit: Current profit (as ratio), calculated based on current_rate.
:param min_stake: Minimal stake size allowed by exchange.
:param max_stake: Balance available for trading.
:param min_stake: Minimal stake size allowed by exchange (for both entries and exits)
:param max_stake: Maximum stake allowed (either through balance, or by exchange limits).
:param current_entry_rate: Current rate using entry pricing.
:param current_exit_rate: Current rate using exit pricing.
:param current_entry_profit: Current profit using entry pricing.
:param current_exit_profit: Current profit using exit pricing.
:param **kwargs: Ensure to keep this here so updates to this won't break your strategy.
:return float: Stake amount to adjust your trade
:return float: Stake amount to adjust your trade,
Positive values to increase position, Negative values to decrease position.
Return None for no action.
"""
return None
@ -548,6 +575,22 @@ class IStrategy(ABC, HyperStrategyMixin):
"""
return None
def populate_any_indicators(self, pair: str, df: DataFrame, tf: str,
informative: DataFrame = None,
set_generalized_indicators: bool = False) -> DataFrame:
"""
Function designed to automatically generate, name and merge features
from user indicated timeframes in the configuration file. User can add
additional features here, but must follow the naming convention.
This method is *only* used in FreqaiDataKitchen class and therefore
it is only called if FreqAI is active.
:param pair: pair to be used as informative
:param df: strategy dataframe which will receive merges from informatives
:param tf: timeframe of the dataframe which will modify the feature names
:param informative: the dataframe associated with the informative pair
"""
return df
###
# END - Intended to be overridden by strategy
###
@ -574,9 +617,6 @@ class IStrategy(ABC, HyperStrategyMixin):
)
informative_pairs.append(pair_tf)
else:
if not self.dp:
raise OperationalException('@informative decorator with unspecified asset '
'requires DataProvider instance.')
for pair in self.dp.current_whitelist():
informative_pairs.append((pair, inf_data.timeframe, candle_type))
return list(set(informative_pairs))
@ -670,7 +710,6 @@ class IStrategy(ABC, HyperStrategyMixin):
# Defs that only make change on new candle data.
dataframe = self.analyze_ticker(dataframe, metadata)
self._last_candle_seen_per_pair[pair] = dataframe.iloc[-1]['date']
if self.dp:
self.dp._set_cached_df(
pair, self.timeframe, dataframe,
candle_type=self.config.get('candle_type_def', CandleType.SPOT))
@ -694,8 +733,6 @@ class IStrategy(ABC, HyperStrategyMixin):
The analyzed dataframe is then accessible via `dp.get_analyzed_dataframe()`.
:param pair: Pair to analyze.
"""
if not self.dp:
raise OperationalException("DataProvider not found.")
dataframe = self.dp.ohlcv(
pair, self.timeframe, candle_type=self.config.get('candle_type_def', CandleType.SPOT)
)
@ -963,7 +1000,7 @@ class IStrategy(ABC, HyperStrategyMixin):
# ROI
# Trailing stoploss
if stoplossflag.exit_type == ExitType.STOP_LOSS:
if stoplossflag.exit_type in (ExitType.STOP_LOSS, ExitType.LIQUIDATION):
logger.debug(f"{trade.pair} - Stoploss hit. exit_type={stoplossflag.exit_type}")
exits.append(stoplossflag)
@ -1035,6 +1072,17 @@ class IStrategy(ABC, HyperStrategyMixin):
sl_higher_long = (trade.stop_loss >= (low or current_rate) and not trade.is_short)
sl_lower_short = (trade.stop_loss <= (high or current_rate) and trade.is_short)
liq_higher_long = (trade.liquidation_price
and trade.liquidation_price >= (low or current_rate)
and not trade.is_short)
liq_lower_short = (trade.liquidation_price
and trade.liquidation_price <= (high or current_rate)
and trade.is_short)
if (liq_higher_long or liq_lower_short):
logger.debug(f"{trade.pair} - Liquidation price hit. exit_type=ExitType.LIQUIDATION")
return ExitCheckTuple(exit_type=ExitType.LIQUIDATION)
# evaluate if the stoploss was hit if stoploss is not on exchange
# in Dry-Run, this handles stoploss logic as well, as the logic will not be different to
# regular stoploss handling.
@ -1052,13 +1100,6 @@ class IStrategy(ABC, HyperStrategyMixin):
f"stoploss is {trade.stop_loss:.6f}, "
f"initial stoploss was at {trade.initial_stop_loss:.6f}, "
f"trade opened at {trade.open_rate:.6f}")
new_stoploss = (
trade.stop_loss + trade.initial_stop_loss
if trade.is_short else
trade.stop_loss - trade.initial_stop_loss
)
logger.debug(f"{trade.pair} - Trailing stop saved "
f"{new_stoploss:.6f}")
return ExitCheckTuple(exit_type=exit_type)

View File

@ -7,6 +7,9 @@ from abc import ABC, abstractmethod
from contextlib import suppress
from typing import Any, Optional, Sequence, Union
from freqtrade.enums.hyperoptstate import HyperoptState
from freqtrade.optimize.hyperopt_tools import HyperoptStateContainer
with suppress(ImportError):
from skopt.space import Integer, Real, Categorical
@ -57,6 +60,13 @@ class BaseParameter(ABC):
Get-space - will be used by Hyperopt to get the hyperopt Space
"""
def can_optimize(self):
return (
self.in_space
and self.optimize
and HyperoptStateContainer.state != HyperoptState.OPTIMIZE
)
class NumericParameter(BaseParameter):
""" Internal parameter used for Numeric purposes """
@ -133,7 +143,7 @@ class IntParameter(NumericParameter):
Returns a List with 1 item (`value`) in "non-hyperopt" mode, to avoid
calculating 100ds of indicators.
"""
if self.in_space and self.optimize:
if self.can_optimize():
# Scikit-optimize ranges are "inclusive", while python's "range" is exclusive
return range(self.low, self.high + 1)
else:
@ -212,7 +222,7 @@ class DecimalParameter(NumericParameter):
Returns a List with 1 item (`value`) in "non-hyperopt" mode, to avoid
calculating 100ds of indicators.
"""
if self.in_space and self.optimize:
if self.can_optimize():
low = int(self.low * pow(10, self._decimals))
high = int(self.high * pow(10, self._decimals)) + 1
return [round(n * pow(0.1, self._decimals), self._decimals) for n in range(low, high)]
@ -261,7 +271,7 @@ class CategoricalParameter(BaseParameter):
Returns a List with 1 item (`value`) in "non-hyperopt" mode, to avoid
calculating 100ds of indicators.
"""
if self.in_space and self.optimize:
if self.can_optimize():
return self.opt_range
else:
return [self.value]

View File

@ -0,0 +1,328 @@
import logging
from functools import reduce
import pandas as pd
import talib.abstract as ta
from pandas import DataFrame
from technical import qtpylib
from freqtrade.exchange import timeframe_to_prev_date
from freqtrade.persistence import Trade
from freqtrade.strategy import DecimalParameter, IntParameter, IStrategy, merge_informative_pair
logger = logging.getLogger(__name__)
class FreqaiExampleStrategy(IStrategy):
"""
Example strategy showing how the user connects their own
IFreqaiModel to the strategy. Namely, the user uses:
self.freqai.start(dataframe, metadata)
to make predictions on their data. populate_any_indicators() automatically
generates the variety of features indicated by the user in the
canonical freqtrade configuration file under config['freqai'].
"""
minimal_roi = {"0": 0.1, "240": -1}
plot_config = {
"main_plot": {},
"subplots": {
"prediction": {"prediction": {"color": "blue"}},
"target_roi": {
"target_roi": {"color": "brown"},
},
"do_predict": {
"do_predict": {"color": "brown"},
},
},
}
process_only_new_candles = True
stoploss = -0.05
use_exit_signal = True
startup_candle_count: int = 300
can_short = False
linear_roi_offset = DecimalParameter(
0.00, 0.02, default=0.005, space="sell", optimize=False, load=True
)
max_roi_time_long = IntParameter(0, 800, default=400, space="sell", optimize=False, load=True)
def informative_pairs(self):
whitelist_pairs = self.dp.current_whitelist()
corr_pairs = self.config["freqai"]["feature_parameters"]["include_corr_pairlist"]
informative_pairs = []
for tf in self.config["freqai"]["feature_parameters"]["include_timeframes"]:
for pair in whitelist_pairs:
informative_pairs.append((pair, tf))
for pair in corr_pairs:
if pair in whitelist_pairs:
continue # avoid duplication
informative_pairs.append((pair, tf))
return informative_pairs
def populate_any_indicators(
self, pair, df, tf, informative=None, set_generalized_indicators=False
):
"""
Function designed to automatically generate, name and merge features
from user indicated timeframes in the configuration file. User controls the indicators
passed to the training/prediction by prepending indicators with `'%-' + coin `
(see convention below). I.e. user should not prepend any supporting metrics
(e.g. bb_lowerband below) with % unless they explicitly want to pass that metric to the
model.
:param pair: pair to be used as informative
:param df: strategy dataframe which will receive merges from informatives
:param tf: timeframe of the dataframe which will modify the feature names
:param informative: the dataframe associated with the informative pair
"""
coin = pair.split('/')[0]
if informative is None:
informative = self.dp.get_pair_dataframe(pair, tf)
# first loop is automatically duplicating indicators for time periods
for t in self.freqai_info["feature_parameters"]["indicator_periods_candles"]:
t = int(t)
informative[f"%-{coin}rsi-period_{t}"] = ta.RSI(informative, timeperiod=t)
informative[f"%-{coin}mfi-period_{t}"] = ta.MFI(informative, timeperiod=t)
informative[f"%-{coin}adx-period_{t}"] = ta.ADX(informative, window=t)
informative[f"%-{coin}sma-period_{t}"] = ta.SMA(informative, timeperiod=t)
informative[f"%-{coin}ema-period_{t}"] = ta.EMA(informative, timeperiod=t)
informative[f"%-{coin}mfi-period_{t}"] = ta.MFI(informative, timeperiod=t)
bollinger = qtpylib.bollinger_bands(
qtpylib.typical_price(informative), window=t, stds=2.2
)
informative[f"{coin}bb_lowerband-period_{t}"] = bollinger["lower"]
informative[f"{coin}bb_middleband-period_{t}"] = bollinger["mid"]
informative[f"{coin}bb_upperband-period_{t}"] = bollinger["upper"]
informative[f"%-{coin}bb_width-period_{t}"] = (
informative[f"{coin}bb_upperband-period_{t}"]
- informative[f"{coin}bb_lowerband-period_{t}"]
) / informative[f"{coin}bb_middleband-period_{t}"]
informative[f"%-{coin}close-bb_lower-period_{t}"] = (
informative["close"] / informative[f"{coin}bb_lowerband-period_{t}"]
)
informative[f"%-{coin}roc-period_{t}"] = ta.ROC(informative, timeperiod=t)
informative[f"%-{coin}relative_volume-period_{t}"] = (
informative["volume"] / informative["volume"].rolling(t).mean()
)
informative[f"%-{coin}pct-change"] = informative["close"].pct_change()
informative[f"%-{coin}raw_volume"] = informative["volume"]
informative[f"%-{coin}raw_price"] = informative["close"]
indicators = [col for col in informative if col.startswith("%")]
# This loop duplicates and shifts all indicators to add a sense of recency to data
for n in range(self.freqai_info["feature_parameters"]["include_shifted_candles"] + 1):
if n == 0:
continue
informative_shift = informative[indicators].shift(n)
informative_shift = informative_shift.add_suffix("_shift-" + str(n))
informative = pd.concat((informative, informative_shift), axis=1)
df = merge_informative_pair(df, informative, self.config["timeframe"], tf, ffill=True)
skip_columns = [
(s + "_" + tf) for s in ["date", "open", "high", "low", "close", "volume"]
]
df = df.drop(columns=skip_columns)
# Add generalized indicators here (because in live, it will call this
# function to populate indicators during training). Notice how we ensure not to
# add them multiple times
if set_generalized_indicators:
df["%-day_of_week"] = (df["date"].dt.dayofweek + 1) / 7
df["%-hour_of_day"] = (df["date"].dt.hour + 1) / 25
# user adds targets here by prepending them with &- (see convention below)
df["&-s_close"] = (
df["close"]
.shift(-self.freqai_info["feature_parameters"]["label_period_candles"])
.rolling(self.freqai_info["feature_parameters"]["label_period_candles"])
.mean()
/ df["close"]
- 1
)
# Classifiers are typically set up with strings as targets:
# df['&s-up_or_down'] = np.where( df["close"].shift(-100) >
# df["close"], 'up', 'down')
# If user wishes to use multiple targets, they can add more by
# appending more columns with '&'. User should keep in mind that multi targets
# requires a multioutput prediction model such as
# templates/CatboostPredictionMultiModel.py,
# df["&-s_range"] = (
# df["close"]
# .shift(-self.freqai_info["feature_parameters"]["label_period_candles"])
# .rolling(self.freqai_info["feature_parameters"]["label_period_candles"])
# .max()
# -
# df["close"]
# .shift(-self.freqai_info["feature_parameters"]["label_period_candles"])
# .rolling(self.freqai_info["feature_parameters"]["label_period_candles"])
# .min()
# )
return df
def populate_indicators(self, dataframe: DataFrame, metadata: dict) -> DataFrame:
# All indicators must be populated by populate_any_indicators() for live functionality
# to work correctly.
# the model will return all labels created by user in `populate_any_indicators`
# (& appended targets), an indication of whether or not the prediction should be accepted,
# the target mean/std values for each of the labels created by user in
# `populate_any_indicators()` for each training period.
dataframe = self.freqai.start(dataframe, metadata, self)
dataframe["target_roi"] = dataframe["&-s_close_mean"] + dataframe["&-s_close_std"] * 1.25
dataframe["sell_roi"] = dataframe["&-s_close_mean"] - dataframe["&-s_close_std"] * 1.25
return dataframe
def populate_entry_trend(self, df: DataFrame, metadata: dict) -> DataFrame:
enter_long_conditions = [df["do_predict"] == 1, df["&-s_close"] > df["target_roi"]]
if enter_long_conditions:
df.loc[
reduce(lambda x, y: x & y, enter_long_conditions), ["enter_long", "enter_tag"]
] = (1, "long")
enter_short_conditions = [df["do_predict"] == 1, df["&-s_close"] < df["sell_roi"]]
if enter_short_conditions:
df.loc[
reduce(lambda x, y: x & y, enter_short_conditions), ["enter_short", "enter_tag"]
] = (1, "short")
return df
def populate_exit_trend(self, df: DataFrame, metadata: dict) -> DataFrame:
exit_long_conditions = [df["do_predict"] == 1, df["&-s_close"] < df["sell_roi"] * 0.25]
if exit_long_conditions:
df.loc[reduce(lambda x, y: x & y, exit_long_conditions), "exit_long"] = 1
exit_short_conditions = [df["do_predict"] == 1, df["&-s_close"] > df["target_roi"] * 0.25]
if exit_short_conditions:
df.loc[reduce(lambda x, y: x & y, exit_short_conditions), "exit_short"] = 1
return df
def get_ticker_indicator(self):
return int(self.config["timeframe"][:-1])
def custom_exit(
self, pair: str, trade: Trade, current_time, current_rate, current_profit, **kwargs
):
dataframe, _ = self.dp.get_analyzed_dataframe(pair=pair, timeframe=self.timeframe)
trade_date = timeframe_to_prev_date(self.config["timeframe"], trade.open_date_utc)
trade_candle = dataframe.loc[(dataframe["date"] == trade_date)]
if trade_candle.empty:
return None
trade_candle = trade_candle.squeeze()
follow_mode = self.config.get("freqai", {}).get("follow_mode", False)
if not follow_mode:
pair_dict = self.freqai.dd.pair_dict
else:
pair_dict = self.freqai.dd.follower_dict
entry_tag = trade.enter_tag
if (
"prediction" + entry_tag not in pair_dict[pair]
or pair_dict[pair]['extras']["prediction" + entry_tag] == 0
):
pair_dict[pair]['extras']["prediction" + entry_tag] = abs(trade_candle["&-s_close"])
if not follow_mode:
self.freqai.dd.save_drawer_to_disk()
else:
self.freqai.dd.save_follower_dict_to_disk()
roi_price = pair_dict[pair]['extras']["prediction" + entry_tag]
roi_time = self.max_roi_time_long.value
roi_decay = roi_price * (
1 - ((current_time - trade.open_date_utc).seconds) / (roi_time * 60)
)
if roi_decay < 0:
roi_decay = self.linear_roi_offset.value
else:
roi_decay += self.linear_roi_offset.value
if current_profit > roi_decay:
return "roi_custom_win"
if current_profit < -roi_decay:
return "roi_custom_loss"
def confirm_trade_exit(
self,
pair: str,
trade: Trade,
order_type: str,
amount: float,
rate: float,
time_in_force: str,
exit_reason: str,
current_time,
**kwargs,
) -> bool:
entry_tag = trade.enter_tag
follow_mode = self.config.get("freqai", {}).get("follow_mode", False)
if not follow_mode:
pair_dict = self.freqai.dd.pair_dict
else:
pair_dict = self.freqai.dd.follower_dict
pair_dict[pair]['extras']["prediction" + entry_tag] = 0
if not follow_mode:
self.freqai.dd.save_drawer_to_disk()
else:
self.freqai.dd.save_follower_dict_to_disk()
return True
def confirm_trade_entry(
self,
pair: str,
order_type: str,
amount: float,
rate: float,
time_in_force: str,
current_time,
entry_tag,
side: str,
**kwargs,
) -> bool:
df, _ = self.dp.get_analyzed_dataframe(pair, self.timeframe)
last_candle = df.iloc[-1].squeeze()
if side == "long":
if rate > (last_candle["close"] * (1 + 0.0025)):
return False
else:
if rate < (last_candle["close"] * (1 - 0.0025)):
return False
return True

Some files were not shown because too many files have changed in this diff Show More