- testGAN2.py has our training loop
- model_test_2.py is our model
- buildnewyalefaces.py and pngtonpy.py build our datasets (latter makes the npy file needed to run the training loop)
- custom_results.py is used to create and evaluate results from our testing set, input a string for the picture of your choice
- psnrgen.npy and psnrint.npy have psnr values of psnr for our psnrgen.npy
- evaluating.py has a function used to calculates psnr for our interpolated and generated images
- training_loss_g.npy and validation_loss_g.npy store our loss over the various epochs
- resultsPSNR.py if run will evaluate PSNR on the generated images in the results file as well as produce bicubically interpolated images and calculate their PSNR
- plot_values_epochs.py plots the PSNR and generator loss over time for the training and validation and bicubically interpolated images
- visualize can take a random image from the results folder (input the epoch number) and create a visualization for it
sjameel98/Enhance
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Folders and files
Name | Name | Last commit message | Last commit date | |
---|---|---|---|---|
Repository files navigation
About
A super resolution model for headshots based on ESRGan
Resources
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published