Feature: Environments #46
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Adds instructions to the README on how to download data using freqtrade.
Adds several technical indicators, candlestick patterns, and normalized returns to the feature engineering process. Includes Weighted Bollinger Bands, MACD, and various bullish/bearish candlestick patterns. Improves return calculations and normalizes annual returns using daily volatility. Adds `can_short` attribute and `informative_pairs` method. Refactors RSI calculation to use a 30-period lookback. Removes raw price and volume features.
Improves the reward function by incorporating volatility normalization, dynamic penalties for long trades, and logarithmic scaling for exit rewards. Removes TODO comments and unused code blocks. Adds a new `most_recent_return` function to calculate tick-to-tick returns. This enhances the agent's learning process by providing a smoother, more informative reward signal, encouraging more strategic trading behavior.
Adds new RL agents `TradeFlowSRAgent` and `TradeFlowRSIWithSRAgent` that leverage support/resistance levels for enhanced trading decisions. `TradeFlowSRAgent` focuses on SR levels, while `TradeFlowRSIWithSRAgent` integrates RSI with SR indicators. Modifies the `config.test.json` to support the new agents and adjust trading parameters. Removes unused `most_recent_return` function from `TradeFlowAgent`. Updates reward calculation to factor in trade duration, volatility, and proximity to support/resistance levels. Includes penalties for inactivity and holding positions too long. Improves reward logic for entry and exit actions.
Decomposes complex reward function into smaller, more manageable methods. This improves readability and makes it easier to understand the logic behind reward calculations. The core reward logic remains unchanged. Renames `TradeFlowAgent` to `TradeFlowSimpleAgent` to better reflect its purpose.
… to reflect its new location.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR introduces enhancements and new functionality focused on improving the management and flexibility of trading environments within the
Trade_Flowapplication. Key updates include:Environment Management:
Reinforcement Learning Integration:
Task Management Improvements:
TaskManagerclass to manage processes and threads, enabling smooth orchestration of tasks like backtesting or live data streaming.Code Quality and Testing:
pytestfor validating environments and agent performance.Additional Features:
Objective:
These changes aim to streamline the development and deployment of advanced trading systems by providing a robust and extensible environment framework, making it easier to experiment with, validate, and deploy trading strategies.