Mark Aron Szulyovszky 1eaba0c221 fix(Evaluate): ignore empty data at evaluation time, add backtesting metrics (sharpe, etc), fixed crash when predicting 0.0 (#23) ...

* fix(Evaluate): ignore empty data at evaluation time, so we don't inflate the model's performance

* refactor(Pipeline): pass in data_loader arguments to the pipeline

* feat(Evaluation): added sharpe, sortino, etc

* fix: Took out the method to fill NaN numbers with 0s. This way in evaluation we can ignore NaN values.

* fix: Fix of the fix added fillna back. Either we root out NaN lines in the very beginning or we stick with the method you created.

Co-authored-by: Daniel Szemerey <szemy2@gmail.com>

2021-12-15 21:11:12 +01:00

..

evaluate.py

fix(Evaluate): ignore empty data at evaluation time, add backtesting metrics (sharpe, etc), fixed crash when predicting 0.0 (#23)

2021-12-15 21:11:12 +01:00

get_prices.py

feat(Forecasting): pytorch-forecasting scaffolding is now working, added narrow data format, fixed missing time column index name (#5)

2021-11-16 10:00:26 +01:00

normalize.py

feat(Model): now successfully training the basic LSTM model

2021-11-11 18:29:05 +01:00

rolling.py

feat(Models): added MAE & RMSE metrics, fixed NaN & Inf values in data,

2021-11-14 23:50:25 +01:00

sktime.py

feat(Models): added a basic sktime model and missing USD crypto currency pairs (#7)

2021-11-17 22:28:34 +01:00

technical_indicators.py

feat(Data): added new derived features + asset pairs for crypto tickers (#6)

2021-11-17 12:07:49 +01:00

typing.py

fix(WalkForward): major bug where we passed in "window of windows of data" is resolved, refactored walk_forward_train_test() and load_data()

2021-12-14 21:07:03 +01:00

visualize.py

feat(Model): now successfully training the basic LSTM model

2021-11-11 18:29:05 +01:00

walk_forward.py

fix(Evaluate): ignore empty data at evaluation time, add backtesting metrics (sharpe, etc), fixed crash when predicting 0.0 (#23)

2021-12-15 21:11:12 +01:00