search

NotJoeMartinez / APDL_Quality_Control

Computer Vision Quality control for the Advanced Particle Detector Laboratory at Texas Tech University
0 stars 1 forks source link

Update preprocessing for new model #12

Closed NotJoeMartinez closed 3 years ago

NotJoeMartinez commented 3 years ago

For some reason the image preprocessing for the "trainable machine" model only works on jpegs. Convert images to jpg before splitting the dataset, or find out how to make the preprocessing work on pngs.

NotJoeMartinez commented 3 years ago

Because I have already written the testing sets to only accept jpeg images I should probably just train the new models on jpegs. This will involve running the mogrify -format jpg *.* script on the main dataset. Hopefully the augment_imgs.py script will accept jpegs.