gavindiaz/drift
1
0
Fork 0
mirror of https://github.com/webclinic017/drift.git synced 2026-08-22 23:38:08 +00:00
Code Issues Packages Projects Releases Wiki Activity

feat(Data): added create_target_cum_forward_returns, now using it for FF & LSTM models, with success

Browse Source
This commit is contained in:
Mark Aron Szulyovszky
2021-11-14 17:25:32 +01:00
parent 478a15a7cb
commit ead7db3a85
2 changed files with 11 additions and 9 deletions
Download Patch File Download Diff File Expand all files Collapse all files
model_ff.py
+6 -5
View File
@@ -1,21 +1,22 @@
#%% Import all the stuff, load data, define constants #%% Import all the stuff, load data, define constants
from load_data import load_files from load_data import load_files, create_target_cum_forward_returns
import pandas as pd import pandas as pd
from tensorflow import keras from tensorflow import keras
from utils.normalize import normalize from utils.normalize import normalize
import tensorflow as tf import tensorflow as tf
from utils.visualize import visualize_loss from utils.visualize import visualize_loss
data = load_files('data/', add_features=False, log_returns=False) data = load_files('data/', add_features=True, log_returns=False)
data.reset_index(drop=True, inplace=True) data.reset_index(drop=True, inplace=True)
data = data[[column for column in data.columns if not column.endswith('volume')]] data = data[[column for column in data.columns if not column.endswith('volume')]]
# data = data[["ETH_returns", "BTC_returns"]] data = data[["BTC_returns", "BTC_mom_10", "BTC_mom_20", "BTC_mom_30", "BTC_mom_60", "BTC_vol_10", "BTC_vol_20", "BTC_vol_60", "day_month", "day_week", "month"]]
ticker_to_predict = 'ETH_returns' ticker_to_predict = 'BTC_returns'
data = create_target_cum_forward_returns(data, ticker_to_predict, 10)
learning_rate = 0.002 learning_rate = 0.002
batch_size = 64 batch_size = 64
epochs = 100 epochs = 500
split_fraction = 0.715 split_fraction = 0.715
train_split = int(split_fraction * int(data.shape[0])) train_split = int(split_fraction * int(data.shape[0]))
model_lstm.py
+5 -4
View File
@@ -1,5 +1,5 @@
#%% Import all the stuff, load data, define constants #%% Import all the stuff, load data, define constants
from load_data import load_files from load_data import load_files, create_target_cum_forward_returns
import pandas as pd import pandas as pd
from tensorflow import keras from tensorflow import keras
from utils.normalize import normalize from utils.normalize import normalize
@@ -7,16 +7,17 @@ import tensorflow as tf
from utils.visualize import visualize_loss from utils.visualize import visualize_loss
from sklearn.preprocessing import StandardScaler from sklearn.preprocessing import StandardScaler
data = load_files('data/', add_features=True, log_returns=True) data = load_files('data/', add_features=True, log_returns=False)
data.reset_index(drop=True, inplace=True) data.reset_index(drop=True, inplace=True)
data = data[[column for column in data.columns if not column.endswith('volume')]] data = data[[column for column in data.columns if not column.endswith('volume')]]
data = data[["BTC_returns", "BTC_mom_10", "BTC_mom_20", "BTC_mom_30", "BTC_vol_10", "BTC_mom_20", "BTC_vol_20"]] data = data[["BTC_returns", "BTC_mom_10", "BTC_mom_20", "BTC_mom_30", "BTC_mom_60", "BTC_vol_10", "BTC_vol_20", "BTC_vol_60", "day_month", "day_week", "month"]]
ticker_to_predict = 'BTC_returns' ticker_to_predict = 'BTC_returns'
data = create_target_cum_forward_returns(data, ticker_to_predict, 10)
learning_rate = 0.002 learning_rate = 0.002
batch_size = 64 batch_size = 64
epochs = 100 epochs = 500
split_fraction = 0.715 split_fraction = 0.715
train_split = int(split_fraction * int(data.shape[0])) train_split = int(split_fraction * int(data.shape[0]))