adelorenzo
/
stable
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

update backtesting to handle new output framework

This commit is contained in:
robcaulk 2022-07-03 17:34:44 +02:00
parent 8ac8d53c32
commit bd3a6ba2fe
2 changed files with 56 additions and 28 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

75
freqtrade/freqai/data_kitchen.py
View File

@ -48,15 +48,17 @@ class FreqaiDataKitchen:
self.data_dictionary: Dict[Any, Any] = {} self.data_dictionary: Dict[Any, Any] = {}
self.config = config self.config = config
self.freqai_config = config["freqai"] self.freqai_config = config["freqai"]
self.predictions: npt.ArrayLike = np.array([]) # self.predictions: npt.ArrayLike = np.array([])
self.do_predict: npt.ArrayLike = np.array([]) # self.do_predict: npt.ArrayLike = np.array([])
self.target_mean: npt.ArrayLike = np.array([]) # self.target_mean: npt.ArrayLike = np.array([])
self.target_std: npt.ArrayLike = np.array([]) # self.target_std: npt.ArrayLike = np.array([])
self.full_predictions: npt.ArrayLike = np.array([]) # self.full_predictions: npt.ArrayLike = np.array([])
self.full_do_predict: npt.ArrayLike = np.array([]) # self.full_do_predict: npt.ArrayLike = np.array([])
self.full_DI_values: npt.ArrayLike = np.array([]) # self.full_DI_values: npt.ArrayLike = np.array([])
self.full_target_mean: npt.ArrayLike = np.array([]) # self.full_target_mean: npt.ArrayLike = np.array([])
self.full_target_std: npt.ArrayLike = np.array([]) # self.full_target_std: npt.ArrayLike = np.array([])
self.full_df: DataFrame = DataFrame()
self.append_df: DataFrame = DataFrame()
self.data_path = Path() self.data_path = Path()
self.label_list: List = [] self.label_list: List = []
self.model_filename: str = "" self.model_filename: str = ""
@ -716,31 +718,56 @@ class FreqaiDataKitchen:
Append backtest prediction from current backtest period to all previous periods Append backtest prediction from current backtest period to all previous periods
""" """
ones = np.ones(len(predictions)) # ones = np.ones(len(predictions))
target_mean, target_std = ones * self.data["target_mean"], ones * self.data["target_std"] # target_mean, target_std = ones * self.data["target_mean"], ones * self.data["target_std"]
self.append_df = DataFrame()
for label in self.label_list:
self.append_df[label] = predictions[label]
self.append_df[f"{label}_mean"] = self.data["labels_mean"][label]
self.append_df[f"{label}_std"] = self.data["labels_std"][label]
self.full_predictions = np.append(self.full_predictions, predictions) self.append_df["do_predict"] = do_predict
self.full_do_predict = np.append(self.full_do_predict, do_predict)
if self.freqai_config.get("feature_parameters", {}).get("DI_threshold", 0) > 0: if self.freqai_config.get("feature_parameters", {}).get("DI_threshold", 0) > 0:
self.full_DI_values = np.append(self.full_DI_values, self.DI_values) self.append_df["DI_values"] = self.DI_values
self.full_target_mean = np.append(self.full_target_mean, target_mean)
self.full_target_std = np.append(self.full_target_std, target_std) if self.full_df.empty:
self.full_df = self.append_df
else:
self.full_df = pd.concat([self.full_df, self.append_df], axis=0)
# self.full_predictions = np.append(self.full_predictions, predictions)
# self.full_do_predict = np.append(self.full_do_predict, do_predict)
# if self.freqai_config.get("feature_parameters", {}).get("DI_threshold", 0) > 0:
# self.full_DI_values = np.append(self.full_DI_values, self.DI_values)
# self.full_target_mean = np.append(self.full_target_mean, target_mean)
# self.full_target_std = np.append(self.full_target_std, target_std)
return return
def fill_predictions(self, len_dataframe): def fill_predictions(self, dataframe):
""" """
Back fill values to before the backtesting range so that the dataframe matches size Back fill values to before the backtesting range so that the dataframe matches size
when it goes back to the strategy. These rows are not included in the backtest. when it goes back to the strategy. These rows are not included in the backtest.
""" """
filler = np.zeros(len_dataframe - len(self.full_predictions)) # startup_candle_count len_filler = len(dataframe) - len(self.full_df.index) # startup_candle_count
self.full_predictions = np.append(filler, self.full_predictions) filler_df = pd.DataFrame(
self.full_do_predict = np.append(filler, self.full_do_predict) np.zeros((len_filler, len(self.full_df.columns))), columns=self.full_df.columns
if self.freqai_config.get("feature_parameters", {}).get("DI_threshold", 0) > 0: )
self.full_DI_values = np.append(filler, self.full_DI_values)
self.full_target_mean = np.append(filler, self.full_target_mean) self.full_df = pd.concat([filler_df, self.full_df], axis=0, ignore_index=True)
self.full_target_std = np.append(filler, self.full_target_std)
to_keep = [col for col in dataframe.columns if not col.startswith("&")]
self.return_dataframe = pd.concat([dataframe[to_keep], self.full_df], axis=1)
self.append_df = DataFrame()
self.full_df = DataFrame()
# self.full_predictions = np.append(filler, self.full_predictions)
# self.full_do_predict = np.append(filler, self.full_do_predict)
# if self.freqai_config.get("feature_parameters", {}).get("DI_threshold", 0) > 0:
# self.full_DI_values = np.append(filler, self.full_DI_values)
# self.full_target_mean = np.append(filler, self.full_target_mean)
# self.full_target_std = np.append(filler, self.full_target_std)
return return

9
freqtrade/freqai/freqai_interface.py
View File

@ -201,6 +201,7 @@ class IFreqaiModel(ABC):
if not self.model_exists( if not self.model_exists(
metadata["pair"], dk, trained_timestamp=trained_timestamp.stopts metadata["pair"], dk, trained_timestamp=trained_timestamp.stopts
): ):
dk.find_features(dataframe_train)
self.model = self.train(dataframe_train, metadata["pair"], dk) self.model = self.train(dataframe_train, metadata["pair"], dk)
self.dd.pair_dict[metadata["pair"]]["trained_timestamp"] = trained_timestamp.stopts self.dd.pair_dict[metadata["pair"]]["trained_timestamp"] = trained_timestamp.stopts
dk.set_new_model_names(metadata["pair"], trained_timestamp) dk.set_new_model_names(metadata["pair"], trained_timestamp)
@ -210,12 +211,12 @@ class IFreqaiModel(ABC):
self.check_if_feature_list_matches_strategy(dataframe_train, dk) self.check_if_feature_list_matches_strategy(dataframe_train, dk)
preds, do_preds = self.predict(dataframe_backtest, dk) pred_df, do_preds = self.predict(dataframe_backtest, dk)
dk.append_predictions(preds, do_preds, len(dataframe_backtest)) dk.append_predictions(pred_df, do_preds, len(dataframe_backtest))
print("predictions", len(dk.full_predictions), "do_predict", len(dk.full_do_predict)) # print("predictions", len(dk.full_predictions), "do_predict", len(dk.full_do_predict))
dk.fill_predictions(len(dataframe)) dk.fill_predictions(dataframe)
return dk return dk