My notes on Machine Learning as I work on projects. I realized that I was looking up the same things too frequently so decided to make notes and logs.
- General Workflow: git, conda, wsl, vscode notes
- Data preprocessing: pandas, numpy notes
- Training: pytorch, tensorflow, hyperparamater tuning, cluster notes
- Domain Specific: time series, object detection, nlp notes
create environment with name env
python3 -m venv env
create environment with latest python
conda create -n mlbase python
update conda itself
conda update conda
save conda environment (clean)
conda env export --from-history | grep -v "prefix" > environment.yml
keyboard shortcuts
Ctrl + Alt + Up/Down Add Cursors
Shift + Alt + Up/Down Copy Line Up/Down
Alt + Up/Down Move Line(s) Up/Down
Ctrl + D Add Cursor next to match
Ctrl + U Cursor Undo
Ctrl + Shift + \ Go to Bracket
Ctrl + ] Indent Line Forward
Ctrl + W Close Tab