import json import os import copy import numpy as np import pandas as pd from pandas import DataFrame from joblib import dump from joblib import load from sklearn.model_selection import train_test_split from sklearn.metrics.pairwise import pairwise_distances import datetime from typing import Any, Dict, List, Tuple import pickle as pk from freqtrade.configuration import TimeRange SECONDS_IN_DAY = 86400 class DataHandler: """ Class designed to handle all the data for the IFreqaiModel class model. Functionalities include holding, saving, loading, and analyzing the data. author: Robert Caulk, rob.caulk@gmail.com """ def __init__(self, config: Dict[str, Any], dataframe: DataFrame, data: List): self.full_dataframe = dataframe (self.training_timeranges, self.backtesting_timeranges) = self.split_timerange( config['freqai']['full_timerange'], config['freqai']['train_period'], config['freqai']['backtest_period']) self.data = data self.data_dictionary = {} self.config = config self.freq_config = config['freqai'] self.predictions = np.array([]) self.do_predict = np.array([]) self.target_mean = np.array([]) self.target_std = np.array([]) def save_data(self, model: Any) -> 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 os.path.exists(self.model_path): os.mkdir(self.model_path) save_path = self.model_path + self.model_filename # Save the trained model dump(model, save_path+"_model.joblib") self.data['model_path'] = self.model_path self.data['model_filename'] = self.model_filename self.data['training_features_list'] = list(self.data_dictionary['train_features'].columns) # store the metadata with open(save_path+"_metadata.json", 'w') as fp: json.dump(self.data, fp, default=self.np_encoder) # save the train data to file so we can check preds for area of applicability later self.data_dictionary['train_features'].to_pickle(save_path+"_trained_df.pkl") return def load_data(self) -> 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 """ model = load(self.model_path+self.model_filename+"_model.joblib") with open(self.model_path+self.model_filename+"_metadata.json", 'r') as fp: self.data = json.load(fp) if self.data.get('training_features_list'): self.training_features_list = [*self.data.get('training_features_list')] self.data_dictionary['train_features'] = pd.read_pickle(self.model_path+ self.model_filename+"_trained_df.pkl") self.model_path = self.data['model_path'] self.model_filename = self.data['model_filename'] if self.config['freqai']['feature_parameters']['principal_component_analysis']: self.pca = pk.load(open(self.model_path+self.model_filename+"_pca_object.pkl","rb")) return model def make_train_test_datasets(self, filtered_dataframe: DataFrame, labels: DataFrame) -> None: ''' Given the dataframe for the full history for training, split the data into training and test data according to user specified parameters in configuration file. :filtered_dataframe: cleaned dataframe ready to be split. :labels: cleaned labels ready to be split. ''' if self.config['freqai']['feature_parameters']['weight_factor'] > 0: weights = self.set_weights_higher_recent(len(filtered_dataframe)) else: weights = np.ones(len(filtered_dataframe)) (train_features, test_features, train_labels, test_labels, train_weights, test_weights) = train_test_split( filtered_dataframe[:filtered_dataframe.shape[0]], labels, weights, **self.config['freqai']['data_split_parameters'] ) return self.build_data_dictionary( train_features,test_features, train_labels,test_labels, train_weights,test_weights) def filter_features(self, unfiltered_dataframe: DataFrame, training_feature_list: List, labels: DataFrame = None, training_filter: bool=True) -> Tuple[DataFrame, DataFrame]: ''' Filter the unfiltered dataframe to extract the user requested features and properly remove all NaNs. Any row with a NaN is removed from training dataset or replaced with 0s in the prediction dataset. However, prediction dataset do_predict will reflect any row that had a NaN and will shield user from that prediction. :params: :unfiltered_dataframe: the full dataframe for the present training period :training_feature_list: list, the training feature list constructed by self.build_feature_list() according to user specified parameters in the configuration file. :labels: the labels for the dataset :training_filter: boolean which lets the function know if it is training data or prediction data to be filtered. :returns: :filtered_dataframe: dataframe cleaned of NaNs and only containing the user requested feature set. :labels: labels cleaned of NaNs. ''' filtered_dataframe = unfiltered_dataframe.filter(training_feature_list, axis=1) drop_index = pd.isnull(filtered_dataframe).any(1) # get the rows that have NaNs, if training_filter: # we don't care about total row number (total no. datapoints) in training, we only care about removing any row with NaNs drop_index_labels = pd.isnull(labels) filtered_dataframe = filtered_dataframe[(drop_index==False) & (drop_index_labels==False)] # dropping values labels = labels[(drop_index==False) & (drop_index_labels==False)] # assuming the labels depend entirely on the dataframe here. print('dropped',len(unfiltered_dataframe)-len(filtered_dataframe), 'training data points due to NaNs, ensure you have downloaded all historical training data') self.data['filter_drop_index_training'] = drop_index else: # we are backtesting so we need to preserve row number to send back to strategy, so now we use do_predict to avoid any prediction based on a NaN drop_index = pd.isnull(filtered_dataframe).any(1) self.data['filter_drop_index_prediction'] = drop_index filtered_dataframe.fillna(0, inplace=True) # replacing all NaNs with zeros to avoid issues in 'prediction', but any prediction that was based on a single NaN is ultimately protected from buys with do_predict drop_index = ~drop_index self.do_predict = np.array(drop_index.replace(True,1).replace(False,0)) print('dropped',len(self.do_predict) - self.do_predict.sum(),'of',len(filtered_dataframe), 'prediction data points due to NaNs. These are protected from prediction with do_predict vector returned to strategy.') return filtered_dataframe, labels def build_data_dictionary(self, train_df: DataFrame, test_df: DataFrame, train_labels: DataFrame, test_labels: DataFrame, train_weights: Any, test_weights: Any) -> Dict: self.data_dictionary = {'train_features': train_df, 'test_features': test_df, 'train_labels': train_labels, 'test_labels': test_labels, 'train_weights': train_weights, 'test_weights': test_weights} return self.data_dictionary def standardize_data(self, data_dictionary: Dict) -> None: ''' Standardize all data in the data_dictionary according to the training dataset :params: :data_dictionary: dictionary containing the cleaned and split training/test data/labels :returns: :data_dictionary: updated dictionary with standardized values. ''' # standardize the data by training stats train_mean = data_dictionary['train_features'].mean() train_std = data_dictionary['train_features'].std() data_dictionary['train_features'] = (data_dictionary['train_features'] - train_mean) / train_std data_dictionary['test_features'] = (data_dictionary['test_features'] - train_mean) / train_std train_labels_std = data_dictionary['train_labels'].std() train_labels_mean = data_dictionary['train_labels'].mean() data_dictionary['train_labels'] = (data_dictionary['train_labels'] - train_labels_mean) / train_labels_std data_dictionary['test_labels'] = (data_dictionary['test_labels'] - train_labels_mean) / train_labels_std for item in train_std.keys(): self.data[item+'_std'] = train_std[item] self.data[item+'_mean'] = train_mean[item] self.data['labels_std'] = train_labels_std self.data['labels_mean'] = train_labels_mean return data_dictionary def standardize_data_from_metadata(self, df: DataFrame) -> DataFrame: ''' Standardizes a set of data using the mean and standard deviation from the associated training data. :params: :df: Dataframe to be standardized ''' for item in df.keys(): df[item] = (df[item] - self.data[item+'_mean']) / self.data[item+'_std'] return df def split_timerange(self, tr: Dict, train_split: int=28, bt_split: int=7) -> list: ''' Function which takes a single time range (tr) and splits it into sub timeranges to train and backtest on based on user input tr: str, full timerange to train on train_split: the period length for the each training (days). Specified in user configuration file bt_split: the backtesting length (dats). Specified in user configuration file ''' train_period = train_split * SECONDS_IN_DAY bt_period = bt_split * SECONDS_IN_DAY full_timerange = TimeRange.parse_timerange(tr) timerange_train = copy.deepcopy(full_timerange) timerange_backtest = copy.deepcopy(full_timerange) tr_training_list = [] tr_backtesting_list = [] first = True while True: if not first: timerange_train.startts = timerange_train.startts + bt_period timerange_train.stopts = timerange_train.startts + train_period # if a full training period doesnt fit, we stop if timerange_train.stopts > full_timerange.stopts: break first = False start = datetime.datetime.utcfromtimestamp(timerange_train.startts) stop = datetime.datetime.utcfromtimestamp(timerange_train.stopts) tr_training_list.append(start.strftime("%Y%m%d")+'-'+stop.strftime("%Y%m%d")) ## associated backtest period timerange_backtest.startts = timerange_train.stopts timerange_backtest.stopts = timerange_backtest.startts + bt_period start = datetime.datetime.utcfromtimestamp(timerange_backtest.startts) stop = datetime.datetime.utcfromtimestamp(timerange_backtest.stopts) tr_backtesting_list.append(start.strftime("%Y%m%d")+'-'+stop.strftime("%Y%m%d")) return tr_training_list, tr_backtesting_list def slice_dataframe(self, tr: str, df: DataFrame) -> DataFrame: """ Given a full dataframe, extract the user desired window :params: :tr: timerange string that we wish to extract from df :df: Dataframe containing all candles to run the entire backtest. Here it is sliced down to just the present training period. """ timerange = TimeRange.parse_timerange(tr) start = datetime.datetime.fromtimestamp(timerange.startts, tz=datetime.timezone.utc) stop = datetime.datetime.fromtimestamp(timerange.stopts, tz=datetime.timezone.utc) df = df.loc[df['date'] >= start, :] df = df.loc[df['date'] <= stop, :] return df def principal_component_analysis(self) -> None: """ Performs Principal Component Analysis on the data for dimensionality reduction and outlier detection (see self.remove_outliers()) No parameters or returns, it acts on the data_dictionary held by the DataHandler. """ from sklearn.decomposition import PCA # avoid importing if we dont need it n_components = self.data_dictionary['train_features'].shape[1] pca = PCA(n_components=n_components) pca = pca.fit(self.data_dictionary['train_features']) n_keep_components = np.argmin(pca.explained_variance_ratio_.cumsum() < 0.999) pca2 = PCA(n_components=n_keep_components) self.data['n_kept_components'] = n_keep_components pca2 = pca2.fit(self.data_dictionary['train_features']) print('reduced feature dimension by',n_components-n_keep_components) print("explained variance",np.sum(pca2.explained_variance_ratio_)) train_components = pca2.transform(self.data_dictionary['train_features']) test_components = pca2.transform(self.data_dictionary['test_features']) self.data_dictionary['train_features'] = pd.DataFrame(data=train_components, columns = ['PC'+str(i) for i in range(0,n_keep_components)], index = self.data_dictionary['train_features'].index) self.data_dictionary['test_features'] = pd.DataFrame(data=test_components, columns = ['PC'+str(i) for i in range(0,n_keep_components)], index = self.data_dictionary['test_features'].index) self.data['n_kept_components'] = n_keep_components self.pca = pca2 if not os.path.exists(self.model_path): os.mkdir(self.model_path) pk.dump(pca2, open(self.model_path + self.model_filename+"_pca_object.pkl","wb")) return None def compute_distances(self) -> float: print('computing average mean distance for all training points') pairwise = pairwise_distances(self.data_dictionary['train_features'],n_jobs=-1) avg_mean_dist = pairwise.mean(axis=1).mean() print('avg_mean_dist',avg_mean_dist) return avg_mean_dist def remove_outliers(self,predict: bool) -> None: """ Remove data that looks like an outlier based on the distribution of each variable. :params: :predict: boolean which tells the function if this is prediction data or training data coming in. """ lower_quantile = self.data_dictionary['train_features'].quantile(0.001) upper_quantile = self.data_dictionary['train_features'].quantile(0.999) if predict: df = self.data_dictionary['prediction_features'][(self.data_dictionary['prediction_features']lower_quantile)] drop_index = pd.isnull(df).any(1) self.data_dictionary['prediction_features'].fillna(0,inplace=True) drop_index = ~drop_index do_predict = np.array(drop_index.replace(True,1).replace(False,0)) print('remove_outliers() tossed',len(do_predict)-do_predict.sum(),'predictions because they were beyond 3 std deviations from training data.') self.do_predict += do_predict self.do_predict -= 1 else: filter_train_df = self.data_dictionary['train_features'][(self.data_dictionary['train_features']lower_quantile)] drop_index = pd.isnull(filter_train_df).any(1) self.data_dictionary['train_features'] = self.data_dictionary['train_features'][(drop_index==False)] self.data_dictionary['train_labels'] = self.data_dictionary['train_labels'][(drop_index==False)] self.data_dictionary['train_weights'] = self.data_dictionary['train_weights'][(drop_index==False)] # do the same for the test data filter_test_df = self.data_dictionary['test_features'][(self.data_dictionary['test_features']lower_quantile)] drop_index = pd.isnull(filter_test_df).any(1) #pdb.set_trace() self.data_dictionary['test_labels'] = self.data_dictionary['test_labels'][(drop_index==False)] self.data_dictionary['test_features'] = self.data_dictionary['test_features'][(drop_index==False)] self.data_dictionary['test_weights'] = self.data_dictionary['test_weights'][(drop_index==False)] return def build_feature_list(self, config: dict) -> int: """ Build the list of features that will be used to filter the full dataframe. Feature list is construced from the user configuration file. :params: :config: Canonical freqtrade config file containing all user defined input in config['freqai] dictionary. """ features = [] for tf in config['freqai']['timeframes']: for ft in config['freqai']['base_features']: for n in range(config['freqai']['feature_parameters']['shift']+1): shift='' if n>0: shift = '_shift-'+str(n) features.append(ft+shift+'_'+tf) for p in config['freqai']['corr_pairlist']: features.append(p.split("/")[0]+'-'+ft+shift+'_'+tf) print('number of features',len(features)) return features def check_if_pred_in_training_spaces(self) -> None: """ Compares the distance from each prediction point to each training data point. It uses this information to estimate a Dissimilarity Index (DI) and avoid making predictions on any points that are too far away from the training data set. """ print('checking if prediction features are in AOA') distance = pairwise_distances(self.data_dictionary['train_features'], self.data_dictionary['prediction_features'],n_jobs=-1) do_predict = np.where(distance.min(axis=0) / self.data['avg_mean_dist'] < self.config['freqai']['feature_parameters']['DI_threshold'],1,0) print('Distance checker tossed',len(do_predict)-do_predict.sum(), 'predictions for being too far from training data') self.do_predict += do_predict self.do_predict -= 1 def set_weights_higher_recent(self, num_weights: int) -> int: """ Set weights so that recent data is more heavily weighted during training than older data. """ weights = np.zeros(num_weights) for i in range(1, len(weights)): weights[len(weights) - i] = np.exp(-i/ (self.config['freqai']['feature_parameters']['weight_factor']*num_weights)) return weights def append_predictions(self, predictions, do_predict, len_dataframe): """ Append backtest prediction from current backtest period to all previous periods """ ones = np.ones(len_dataframe) s_mean, s_std = ones*self.data['s_mean'], ones*self.data['s_std'] self.predictions = np.append(self.predictions,predictions) self.do_predict = np.append(self.do_predict,do_predict) self.target_mean = np.append(self.target_mean,s_mean) self.target_std = np.append(self.target_std,s_std) return def fill_predictions(self, len_dataframe): """ 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. """ filler = np.zeros(len_dataframe -len(self.predictions)) # startup_candle_count self.predictions = np.append(filler,self.predictions) self.do_predict = np.append(filler,self.do_predict) self.target_mean = np.append(filler,self.target_mean) self.target_std = np.append(filler,self.target_std) return def np_encoder(self, object): if isinstance(object, np.generic): return object.item()