search

Doodleverse / segmentation_gym

A neural gym for training deep learning models to carry out geoscientific image segmentation. Works best with labels generated using https://github.com/Doodleverse/dash_doodler
MIT License
45 stars 11 forks source link

Added Comments for Improved Readability #69

Closed 2320sharon closed 2 years ago

2320sharon commented 2 years ago

I added comments to seg_images_in_folder.py that clarify the purpose of variables such as weights,sample_direc, and more. These comments also explain the logic behind why certain files are loaded in, what is happening in each segment in the code, and the dynamically created variables from

 for k in config.keys():
    exec(k+'=config["'+k+'"]')

I also included comments for some of the functions, models, and variables loaded in from doodleverse_utils as it was difficult to discern where functions like do_seg were defined.

2320sharon commented 2 years ago

On lines 160 and 174 respectively the comments read #346,564 #242,812. I am assuming this is the target size for 'simple_unet' and 'satunet' right?

dbuscombe-usgs commented 2 years ago

Thank you I will manually add your comments to avoid git merge headaches

dbuscombe-usgs commented 2 years ago

I incorporated all these excellent comments in my latest push to main, https://github.com/Doodleverse/segmentation_gym/commit/7f9d9550d43f0d24e2a3a5968b9bd85eabb57dd4