search

asavinov / intelligent-trading-bot

Intelligent Trading Bot: Automatically generating signals and trading based on machine learning and feature engineering
https://t.me/intelligent_trading_signals
MIT License
803 stars 179 forks source link

when i change the base_window such like your example new parameters #26

Closed ellisfan closed 1 year ago

ellisfan commented 1 year ago 
"base_window": 40320,
"averaging_windows": [1, 60, 360, 1440, 4320, 10080],
"area_windows": [60, 360, 1440, 4320, 10080],

and i change train_features:

"train_features": [
    "close_1", "close_60", "close_360", "close_1440", "close_4320", "close_10080",
    "close_std_60", "close_std_360", "close_std_1440", "close_std_4320", "close_std_10080",
    "volume_1", "volume_60", "volume_360", "volume_1440", "volume_4320", "volume_10080",
    "span_1", "span_60", "span_360", "span_1440", "span_4320", "span_10080",
    "trades_1", "trades_60", "trades_360", "trades_1440", "trades_4320", "trades_10080",
    "tb_base_1", "tb_base_60", "tb_base_360", "tb_base_1440", "tb_base_4320", "tb_base_10080",
    "close_area_60", "close_area_360", "close_area_1440", "close_area_4320", "close_area_10080",
    "close_trend_60", "close_trend_360", "close_trend_1440", "close_trend_4320", "close_trend_10080",
    "volume_trend_60", "volume_trend_360", "volume_trend_1440", "volume_trend_4320", "volume_trend_10080"
  ],

so do i need to change labels and score_aggregation?

iploshchik commented 1 year ago

you need to re-train your model, no need to change labels as I suppose