imatge-upc / salgan

SalGAN: Visual Saliency Prediction with Generative Adversarial Networks
https://imatge-upc.github.io/salgan
MIT License
368 stars 106 forks source link

looking for modelWeights0080.npz #2

Closed josephwinston closed 7 years ago

josephwinston commented 7 years ago

I'm running the test script to produce the saliency maps as shown in the README (python 03-predict.py)

I cannot find in the instructions where to obtain the model weights. I see https://imatge.upc.edu/web/sites/default/files/resources/1720/saliency/2017-salgan/gen_modelWeights0090.npz and https://imatge.upc.edu/web/sites/default/files/resources/1720/saliency/2017-salgan/discrim_modelWeights0090.npz but the prefix of either gen_ or discrim_ is not present in the call to load_weights in 03-predict nor the epic of 80.

junting commented 7 years ago

Hello @josephwinston,

Thank you for your interest! I found some bug in the code, and they are fixed now! PS: The prefix of gen_ or discrim_ can be set with the input argument path load_weights(model.net['output'], path='gen_', epochtoload=90) But anyway, there are some other minor bugs too, so please update the code with the latest changes and you can try to run the test script with the provided test image to compare the result.

Best, Junting