mirror of
https://github.com/freqtrade/freqtrade.git
synced 2024-11-14 20:23:57 +00:00
163 lines
6.0 KiB
Python
163 lines
6.0 KiB
Python
import logging
|
|
from typing import Optional
|
|
|
|
import numpy as np
|
|
import pandas as pd
|
|
|
|
from freqtrade.configuration import TimeRange
|
|
from freqtrade.constants import DEFAULT_DATAFRAME_COLUMNS, DEFAULT_TRADES_COLUMNS, TradeList
|
|
from freqtrade.enums import CandleType
|
|
|
|
from .idatahandler import IDataHandler
|
|
|
|
|
|
logger = logging.getLogger(__name__)
|
|
|
|
|
|
class HDF5DataHandler(IDataHandler):
|
|
|
|
_columns = DEFAULT_DATAFRAME_COLUMNS
|
|
|
|
def ohlcv_store(
|
|
self, pair: str, timeframe: str, data: pd.DataFrame, candle_type: CandleType) -> None:
|
|
"""
|
|
Store data in hdf5 file.
|
|
:param pair: Pair - used to generate filename
|
|
:param timeframe: Timeframe - used to generate filename
|
|
:param data: Dataframe containing OHLCV data
|
|
:param candle_type: Any of the enum CandleType (must match trading mode!)
|
|
:return: None
|
|
"""
|
|
key = self._pair_ohlcv_key(pair, timeframe)
|
|
_data = data.copy()
|
|
|
|
filename = self._pair_data_filename(self._datadir, pair, timeframe, candle_type)
|
|
self.create_dir_if_needed(filename)
|
|
|
|
_data.loc[:, self._columns].to_hdf(
|
|
filename, key, mode='a', complevel=9, complib='blosc',
|
|
format='table', data_columns=['date']
|
|
)
|
|
|
|
def _ohlcv_load(self, pair: str, timeframe: str,
|
|
timerange: Optional[TimeRange], candle_type: CandleType
|
|
) -> pd.DataFrame:
|
|
"""
|
|
Internal method used to load data for one pair from disk.
|
|
Implements the loading and conversion to a Pandas dataframe.
|
|
Timerange trimming and dataframe validation happens outside of this method.
|
|
:param pair: Pair to load data
|
|
:param timeframe: Timeframe (e.g. "5m")
|
|
:param timerange: Limit data to be loaded to this timerange.
|
|
Optionally implemented by subclasses to avoid loading
|
|
all data where possible.
|
|
:param candle_type: Any of the enum CandleType (must match trading mode!)
|
|
:return: DataFrame with ohlcv data, or empty DataFrame
|
|
"""
|
|
key = self._pair_ohlcv_key(pair, timeframe)
|
|
filename = self._pair_data_filename(
|
|
self._datadir,
|
|
pair,
|
|
timeframe,
|
|
candle_type=candle_type
|
|
)
|
|
|
|
if not filename.exists():
|
|
# Fallback mode for 1M files
|
|
filename = self._pair_data_filename(
|
|
self._datadir, pair, timeframe, candle_type=candle_type, no_timeframe_modify=True)
|
|
if not filename.exists():
|
|
return pd.DataFrame(columns=self._columns)
|
|
where = []
|
|
if timerange:
|
|
if timerange.starttype == 'date':
|
|
where.append(f"date >= Timestamp({timerange.startts * 1e9})")
|
|
if timerange.stoptype == 'date':
|
|
where.append(f"date <= Timestamp({timerange.stopts * 1e9})")
|
|
|
|
pairdata = pd.read_hdf(filename, key=key, mode="r", where=where)
|
|
|
|
if list(pairdata.columns) != self._columns:
|
|
raise ValueError("Wrong dataframe format")
|
|
pairdata = pairdata.astype(dtype={'open': 'float', 'high': 'float',
|
|
'low': 'float', 'close': 'float', 'volume': 'float'})
|
|
pairdata = pairdata.reset_index(drop=True)
|
|
return pairdata
|
|
|
|
def ohlcv_append(
|
|
self,
|
|
pair: str,
|
|
timeframe: str,
|
|
data: pd.DataFrame,
|
|
candle_type: CandleType
|
|
) -> None:
|
|
"""
|
|
Append data to existing data structures
|
|
:param pair: Pair
|
|
:param timeframe: Timeframe this ohlcv data is for
|
|
:param data: Data to append.
|
|
:param candle_type: Any of the enum CandleType (must match trading mode!)
|
|
"""
|
|
raise NotImplementedError()
|
|
|
|
def trades_store(self, pair: str, data: TradeList) -> None:
|
|
"""
|
|
Store trades data (list of Dicts) to file
|
|
:param pair: Pair - used for filename
|
|
:param data: List of Lists containing trade data,
|
|
column sequence as in DEFAULT_TRADES_COLUMNS
|
|
"""
|
|
key = self._pair_trades_key(pair)
|
|
|
|
pd.DataFrame(data, columns=DEFAULT_TRADES_COLUMNS).to_hdf(
|
|
self._pair_trades_filename(self._datadir, pair), key,
|
|
mode='a', complevel=9, complib='blosc',
|
|
format='table', data_columns=['timestamp']
|
|
)
|
|
|
|
def trades_append(self, pair: str, data: TradeList):
|
|
"""
|
|
Append data to existing files
|
|
:param pair: Pair - used for filename
|
|
:param data: List of Lists containing trade data,
|
|
column sequence as in DEFAULT_TRADES_COLUMNS
|
|
"""
|
|
raise NotImplementedError()
|
|
|
|
def _trades_load(self, pair: str, timerange: Optional[TimeRange] = None) -> TradeList:
|
|
"""
|
|
Load a pair from h5 file.
|
|
:param pair: Load trades for this pair
|
|
:param timerange: Timerange to load trades for - currently not implemented
|
|
:return: List of trades
|
|
"""
|
|
key = self._pair_trades_key(pair)
|
|
filename = self._pair_trades_filename(self._datadir, pair)
|
|
|
|
if not filename.exists():
|
|
return []
|
|
where = []
|
|
if timerange:
|
|
if timerange.starttype == 'date':
|
|
where.append(f"timestamp >= {timerange.startts * 1e3}")
|
|
if timerange.stoptype == 'date':
|
|
where.append(f"timestamp < {timerange.stopts * 1e3}")
|
|
|
|
trades: pd.DataFrame = pd.read_hdf(filename, key=key, mode="r", where=where)
|
|
trades[['id', 'type']] = trades[['id', 'type']].replace({np.nan: None})
|
|
return trades.values.tolist()
|
|
|
|
@classmethod
|
|
def _get_file_extension(cls):
|
|
return "h5"
|
|
|
|
@classmethod
|
|
def _pair_ohlcv_key(cls, pair: str, timeframe: str) -> str:
|
|
# Escape futures pairs to avoid warnings
|
|
pair_esc = pair.replace(':', '_')
|
|
return f"{pair_esc}/ohlcv/tf_{timeframe}"
|
|
|
|
@classmethod
|
|
def _pair_trades_key(cls, pair: str) -> str:
|
|
return f"{pair}/trades"
|