#import mylib_dataset as md
import mylib_rf as mrf
import feature_list
#import numpy as np
#import matplotlib.pyplot as plt
proc_time_start = datetime.datetime.now()
#start_date = md.get_datetime_from_string('2014-01-19')
#end_date = md.get_datetime_from_string('2018-05-1')
feature_dicts = feature_list.get_features_dicts()
#blockchain_indicators = feature_list.get_blockchain_indicators()
rf_obj = mrf.get_data_RF()
dataset_path, log_path, rootLogger = mrf.init_paths()
data = rf_obj.get_input_data(database_path=dataset_path,
feature_dicts=feature_dicts,
normalization_method='rescale',
dataset_type='dataset2',
normalization=None,
lookback=None)
test_train_data, h, l = rf_obj.get_test_and_train_data(label_window=4,
label_type='bool_up',
thresh_ratio=1,
cross_validation=None)
#rf_obj.normalize_features()
##################### get log for estimator
proc_time_start = datetime.datetime.now()
#start_date = md.get_datetime_from_string('2014-01-19')
#end_date = md.get_datetime_from_string('2018-05-1')
path_features_imp = '/home/catalin/git_workspace/disertatie/dict_perf_feats.pkl'
ordered_values_mean, ordered_values_var = md.get_feature_importances_mean(path_features_imp)
features = feature_list.get_feature_set()[0]
features.extend(ordered_values_mean['keys'][:30])
features = feature_list.get_features_list()
rf_obj = rf.get_data_RF()
dataset_path, log_path, rootLogger = rf.init_paths()
data = rf_obj.get_input_data(database_path = dataset_path,
feature_names = features,
preprocessing_constant = 0.9,
normalization_method = 'rescale',
skip_preprocessing = False,
#datetime_interval = {'start':start_date,'end':end_date},
datetime_interval = {},
blockchain_indicators = {},
lookback = 0 )
test_train_eth, h ,l = rf_obj.get_test_and_train_data(label_window = 12 , bool_labels = True)