pytorch-image-comp-rnn
pytorch-image-comp-rnn copied to clipboard
1zb
PyTorch implementation of Full Resolution Image Compression with Recurrent Neural Networks
Dear @1zb , Greetings! **New Version of RNN with GDN** [Paper] "Islam, Khawar, L. Minh Dang, Sujin Lee, and Hyeonjoon Moon. "Image Compression With Recurrent Neural Network and Generalized Divisive...
hi 1zb, you did a great code! i tried to train your model. but i am confused with the training dataset. I saw the paper use 216 million images. but...
I tried to train a model on one cifar class. The model was trained, but, after training, an error appeared in the encoder. >> python3.7 encoder.py --model checkpoint/encoder_iter_00000000.pth --input ../dset/0/apple_s_000027.png...
does this implementation contain the entropy coding implementation too? the BinarryRNN part?
Can you give a network diagram of this code? thank you！
First, congratulations for the great code. I'm doing undergraduate work and would like to base it on your work. Can I use it? If so, I would also like to...
I am trying to reproduce this work too. Now I am training the model. However, I find the convergence is quite slow. I trained on a TITAN X for about...
Thanks for the great repository! We have a project that partially uses a modification of your code. We're wondering if you'd mind if we release the code publicly. Of course...
Can anybody provide the code for implementing the paper with variants of lstm as,,,associative lstm and GRU
Shouldn't this line https://github.com/1zb/pytorch-image-comp-rnn/blob/980f2ff6733c37ca0f4ecce9685a78fd873b2b3c/train.py#L155 be `res = res0 - output`, where `res0 = patches - 0.5`? Rather than `res = res - output` ? The residual at each iteration should...
