gulzainali98

did you find the solution

@pcshih can you please mention the problem you are running into?

@pcshih which file to run from the three files?

@parz1val1028 did you figure it out?

this is how i am testing the models : python main.py --is_train=False --load_path="path to saved models"

I also have the similar problem and my results seem to quite similar to some of the results mentioned here i.e I got 31.257 PSNR on noise=30 on set8.

even when i run the trained model the results are not completely similar. I get 31.60 something vs 31.68 mentioned in the paper.

Here is the constant file. ``` # Copyright (c) Meta Platforms, Inc. and affiliates. All Rights Reserved. # # This source code is licensed under the MIT license found in...

@punitkoura I have checked the constants.py file. this was missing in my constants "--ddp-backend fully_sharded", as compared to provided here:https://github.com/facebookresearch/metaseq/blob/main/metaseq/cli/README.md Not sure if this is the reason for the error.

yes i have changed it. Here is my updated constant file that i am running right now. ``` import os MAX_SEQ_LEN = 2048 BATCH_SIZE = 2048 # silly high bc...