The SagesGymCNN strategy is designed for backtesting trading strategies. It has several key components:
populate_indicators function: This function takes raw data from an exchange and adds various technical analysis (TA) indicators to the DataFrame. It is recommended to only include the necessary indicators to optimize memory and CPU usage.
populate_buy_trend function: Based on the TA indicators, this function populates the buy signal for the given DataFrame.
It uses a reinforcement learning model (rl_model) to predict the action.
If the predicted buy action is greater than 0.50, the 'buy' column in the DataFrame is set to 1, indicating a buy signal. Similarly, the 'sell' column is set based on the predicted sell action. populate_sell_trend function: Similar to populate_buy_trend, this function populates the sell signal based on the TA indicators. It sets the 'sell' column in the DataFrame accordingly. preprocess function: This function preprocesses the indicator data by applying transformations or feature extraction techniques. The processed data is then fed into the models for prediction. The main part of the strategy involves iterating over the indicators and predicting the actions for each window of data. The actions are stored in the action_output DataFrame. The strategy utilizes a convolutional neural network (SagesGymCNN) for making predictions based on the indicators and reinforces the buy and sell signals. The strategy aims to provide a framework for backtesting trading strategies and evaluating their performance.
populate_buy_trend function: Based on the TA indicators, this function populates the buy signal for the given DataFrame.
It uses a reinforcement learning model (rl_model) to predict the action.
If the predicted buy action is greater than 0.50, the 'buy' column in the DataFrame is set to 1, indicating a buy signal. Similarly, the 'sell' column is set based on the predicted sell action. populate_sell_trend function: Similar to populate_buy_trend, this function populates the sell signal based on the TA indicators. It sets the 'sell' column in the DataFrame accordingly. preprocess function: This function preprocesses the indicator data by applying transformations or feature extraction techniques. The processed data is then fed into the models for prediction. The main part of the strategy involves iterating over the indicators and predicting the actions for each window of data. The actions are stored in the action_output DataFrame. The strategy utilizes a convolutional neural network (SagesGymCNN) for making predictions based on the indicators and reinforces the buy and sell signals. The strategy aims to provide a framework for backtesting trading strategies and evaluating their performance.
stoploss: -0.99 timeframe: 1h hash(sha256): 389d5c2e3d61badd7eed4d5da37b03471405b0389decd3aef03037b8b0bcd65b indicators: sell date open close buy
No similar strategies found. (based on used indicators)
last change: 2024-04-29 00:24:16