Loading repository data…
Adrijan-Petek / repository
A starter template for a self-learning crypto trading bot. Includes Python backend scripts for data ingestion, feature engineering, a simple ML model (Random Forest), a self-learning loop, backtesting utilities, sample reports, and a minimal web placeholder. This is a template- you'll need to add your own API keys
A transparent discovery signal based on current public GitHub metadata.
This score does not audit code, security, maintainers, documentation quality, or suitability. Verify the repository and its current documentation before adoption.
A starter template for a self-learning crypto trading bot.
Includes Python backend scripts for data ingestion, feature engineering, a simple ML model (Random Forest), a self-learning loop, backtesting utilities, sample reports, and a minimal web placeholder.
This is a template- you'll need to add your own API keys and tweak strategies before using with real money.
python -m venv venv
source venv/bin/activate
pip install -r requirements.txt
Edit .env (create from .env.example) with your keys and config.
Run a backtest (sample):
python -m src.bot --symbol BTC/USDT --timeframe 1h --mode backtest
python -m src.bot --symbol BTC/USDT --timeframe 1h --mode paper
crypto-self-learning-bot/
├─ src/
│ ├─ data.py
│ ├─ features.py
│ ├─ model.py
│ ├─ bot.py
├─ config/
│ └─ bot.config.json
├─ reports/
│ └─ sample-report.json
├─ web/
│ └─ README.md
├─ .github/workflows/daily-bot.yml
├─ .env.example
├─ requirements.txt
└─ README.md
A full-stack crypto strategy lab with a Python backtest engine and a production-style Next.js dashboard.
This project gives you a complete loop:
Selected from shared topics, language and repository description—not editorial ratings.
janishar /
A python project starter template for data-analytics and data-science.
83/100 healthccxt.SMA, EMA, RSI, MACD, returns).RandomForestClassifier).reports/sample-report.json (latest run),reports/backtest-history.jsonl (run archive),reports/latest-equity-curve.csv (chart-ready time series),reports/model.joblib (optional saved model).GET /api/reportGET /api/configGET /api/historyGET /api/equitysrc/data.py data loading (CSV + exchange)src/features.py indicators and feature columnssrc/model.py ML training/inference wrappersrc/bot.py CLI entrypoint and report pipelineconfig/bot.config.json default strategy/runtime settingsreports/ generated run artifactsweb/ Next.js dashboardpython3 -m venv .venv
./.venv/bin/pip install -r requirements.txt
Run a backtest:
./.venv/bin/python -m src.bot --mode backtest --save_model
The command updates:
reports/sample-report.jsonreports/backtest-history.jsonlreports/latest-equity-curve.csvreports/model.joblib (when --save_model is used)cd web
npm install
npm run dev
Open http://localhost:3000.
Production build:
npm run build
npm run start
Main command:
./.venv/bin/python -m src.bot --mode backtest [options]
Important options:
--symbol trading pair (default BTC/USDT)--timeframe candle timeframe (default 1h)--initial_balance starting capital (default 10000)--fee_rate transaction fee per trade (default 0.001)--predict_horizon target horizon in candles--data_mode sample or exchange--data_path custom CSV path when using sample mode--exchange_name exchange id for ccxt (default binance)--limit OHLCV row limit in exchange mode--report_dir artifact output directory (default reports)--save_model persist model to disk--model_path model output path (default reports/model.joblib)./.venv/bin/python -m src.bot \
--mode backtest \
--data_mode exchange \
--exchange_name binance \
--symbol BTC/USDT \
--timeframe 1h \
--limit 1200 \
--save_model
This software is for research and education. It is not investment advice. Crypto trading is high risk. Do not deploy with real capital without robust validation, controls, and monitoring.
See LICENSE.
(Resolve merge conflicts and fix bot syntax error)