Option trading based on a strategy
AdaBoost model from sklearn package due to 32bit system on raspberry pi
pre processing of data should not output duplicates or non mature options models should be well documented
- name
- date
- details of data trained on
- features used
- performance
- use optuna package for bayesian search with TPE surrogate
The following Errors have been encountered on Raspberry PI when import numpy or matplotlib ImportError: /lib/arm-linux-gnueabihf/libc.so.6: version `GLIBC_2.28' not found
Solved: pip install numpy --global-option="-mfloat-abi=hard" --force-reinstall now solved by updating raspberry version from 9 to 10 (Buster)
Installing chromium webdriver and browser to make scraping work sudo apt-get install chromium-browser --yes apt-get install chromium-chromedriver --yes apt-get install xvfb --yes pip install PyVirtualDisplay xvfbwrapper selenium
Swap file raspberry pi, increased size according to these steps
Due to memory leakage of lxpanel it might be wise to restart it periodically
lxpanelctl restart
Some basic commands to list raspberry pi usage:
disk space: df -h
memory usage: free -h
processes: ps aux