Summary: collection of useful Python functions and modules which I commonly use in the context of machine learning workflows
Design decisions:
- single module dependecy: each module should not depend on other models inside of the repository
- each utility function should be self contained
ffmpeg -f lavfi -i testsrc=duration=5:size=300x300:rate=30 -vf "drawtext=fontfile=/usr/share/fonts/truetype/dejavu/DejaVuSans-Bold.ttf: text='%{n}': fontcolor=white: fontsize=h/12: x=(w-text_w)/2: y=(h-text_h)/2: box=1: boxcolor=black@0.5" -c:v mpeg4 -qscale:v 1 -y centered_numbered.avi