speaks for itself
conda install with conda create --name cryt310 --file requirements2.txt
install with pip3 install -r requirements2.txt
uses python 3.10
install ta lib with conda install -c conda-forge ta-lib
pip install requests schedule pandas mplfinance numpy
conda create -n cryt310 python=3.10
conda activate cryt310
pip install -q numpy requests schedule pandas mplfinance notebook
pip install python-binance
conda install -y -c conda-forge ta-lib
python testimports.py
python test_trades.py
-
fix entry for times when entry was before initialisation ( need to check )
-
look for tickers at 5m interval with more than 9% (to 6%) change. next ticks might be high.
- see STMXUSDT at 2023/7/23 12:30
-
restructure the code so that its more modular. the followings should be kept seperate
- download/get data, identify signals, act on signals
- this way, (2) and (3) can be backtested
-
✅ rewrite
run_trades_ver2.py
to scale up to ~750 ticker/interval pairs per 5 minutes -
✅ fix run_trades_ver2, dont Upslow on initial run?
-
✅ reduce status calls to discord ping.
-
✅ limit number of positions? using binanceexceptions
-
✅ some code to validate the timing of prev candlestick
-
✅ a script to start/restart all runs
-
✅ run multiple signals at the same time.
-
✅ try binance apis for trading
-
✅ update historical data to recent months
-
✅ report reason for exit, either TP,SL,Exitsignal
- from 9_0_5
- do predictions for top 10 tickers at 5m resolution
- then use hourly/30m-ly 24hr change to find datasets to validate this analysis/prediction
- get_data to return more live data then just 1000
- find tickerpairs present on MT4 trading platform.
- optimise parameters for tickerpairs.
conda activate cryt310 cd Documents\Github\cryptotradr python aver6_run_trades.py
python aver5_run_trades.py 0 40 5m python aver5_run_trades.py 80 120 30m