pix2pix
pix2pix copied to clipboard
phillipi
Image-to-image translation with conditional adversarial nets
Hello, I'm sorry if this is a very obvious question, but I was wondering if it is possible to test on a CPU? I have successfully managed to train on...
New iMAC with new fresh install and having some problems installing pixtopix / torch (had to downgrade the CLT to enable running install-deps and other commands) and just when I...
Is it possible to somehow use the U-net architecture with images of size 480 x 640? In its current implementation the U-net seems to only work with images of size...
``` running "find" on each class directory, and concatenate all those filenames into a single file containing all image paths for a given class trainCache /Users/swinghu/prog/torch/pix2pix/cache/_Users_swinghu_prog_torch_pix2pix_datasets_facades_train_trainCache.t7 Creating train metadata serial...
`DATA_ROOT=/home……/datasets/facades name=facades_generation which_direction=AtoB th train.lua` i get the following error: ``` transferring to gpu…… done cannnot open /home……/models/facades_generation/opt.txt in mode w at /home……/torch/pkg/torch lib/TH/THDiskFile.c:670 ``` what's wrong? can anybody help...
It seems that less results was showed for bw2color task than other tasks in the paper and the project page. If result looks promise at most cases, I would like...
Running current implementation with 16 bit images automatically scales down output images to 8 bit. I there a way to get 16 bit output images ?
I'm attempting to process large amounts of images at a time (7000+) with pix2pix, but I am finding that Images from the /val folder are getting lost by the time...
createRealFake() is called 3 times in the main loop, one for the optimization, one for display and one for saving to the disk. I'm wondering why the last two are...
Because some operation is difficult to define in nngraph，I want to make an nn.sequential to realize U-net network architecture? what about use the nn.ConcatTable() to do it ? Anybody have...
