The MLStrategy is a backtesting strategy implemented in the Freqtrade bot framework. It aims to generate buy and sell signals for trading based on machine learning predictions. Here's a breakdown of its key components:
minimal_roi: This dictionary defines the target returns for different time intervals.
For example, it expects a 0% return in 40 minutes, a 1% return in 30 minutes, a 2% return in 20 minutes, and a 4% return in 0 minutes (immediate exit).
stoploss: This parameter sets the maximum acceptable loss for a trade, triggering a sell signal if the price drops by 10% from the entry point.
ticker_interval: This value represents the time interval between consecutive data points in the trading data. In this case, it is set to 5 minutes. slippage: This parameter accounts for the difference between the expected price and the actual price during trade execution. Here, it is set to 0.01 (1%). ML_parse_ticker_dataframe: This method preprocesses the ticker history data by setting the index and converting the date format to datetime. populate_indicators: This method adds technical analysis (TA) indicators to the DataFrame. It applies the Heikin-Ashi chart type to the data, converting it to Heikin-Ashi candles (ha_open, ha_close, ha_high, ha_low). populate_buy_trend: This method populates the "buy" column in the DataFrame based on machine learning predictions. It assigns a value of 1 to the "buy" column when the "Prediction" column is equal to 1. populate_sell_trend: This method populates the "sell" column in the DataFrame based on machine learning predictions. It assigns a value of 1 to the "sell" column when the "Prediction" column is equal to 0. Overall, this strategy utilizes machine learning predictions to generate buy and sell signals based on the provided indicators and predefined ROI and stop-loss parameters.
For example, it expects a 0% return in 40 minutes, a 1% return in 30 minutes, a 2% return in 20 minutes, and a 4% return in 0 minutes (immediate exit).
stoploss: This parameter sets the maximum acceptable loss for a trade, triggering a sell signal if the price drops by 10% from the entry point.
ticker_interval: This value represents the time interval between consecutive data points in the trading data. In this case, it is set to 5 minutes. slippage: This parameter accounts for the difference between the expected price and the actual price during trade execution. Here, it is set to 0.01 (1%). ML_parse_ticker_dataframe: This method preprocesses the ticker history data by setting the index and converting the date format to datetime. populate_indicators: This method adds technical analysis (TA) indicators to the DataFrame. It applies the Heikin-Ashi chart type to the data, converting it to Heikin-Ashi candles (ha_open, ha_close, ha_high, ha_low). populate_buy_trend: This method populates the "buy" column in the DataFrame based on machine learning predictions. It assigns a value of 1 to the "buy" column when the "Prediction" column is equal to 1. populate_sell_trend: This method populates the "sell" column in the DataFrame based on machine learning predictions. It assigns a value of 1 to the "sell" column when the "Prediction" column is equal to 0. Overall, this strategy utilizes machine learning predictions to generate buy and sell signals based on the provided indicators and predefined ROI and stop-loss parameters.
Unable to parse Traceback (Logfile Exceeded Limit)
stoploss: -0.1 timeframe: 5m hash(sha256): bcbc0fed43fce94c14b61282c485b9440bdd7eab0fe23584a363ee977c35f164 indicators: high ha_high Prediction ha_low self close ha_close date open low ha_open
No similar strategies found. (based on used indicators)
last change: 2024-07-27 08:43:35