shshim0513 / sd-viton Goto Github PK

Hello, thank you for your excellent efforts. I'm currently training the model on the dress code dataset. However, I'm encountering challenges with convergence. Could you provide insights into how long the model was trained on the VITON-HD dataset for reference?

Hi @SHShim0513 ,

For each of these parameters, what models should be selected for them?
--tocg_checkpoint {condition generator ckpt}
--gen_checkpoint {image generator ckpt}

Thank you.

Do anyone know how to fix this issue?

File "/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py", line 2152, in load_state_dict
raise RuntimeError('Error(s) in loading state_dict for {}:\n\t{}'.format(
RuntimeError: Error(s) in loading state_dict for SPADEGenerator:
Missing key(s) in state_dict: "up_4.conv_0.weight_orig", "up_4.conv_0.weight", "up_4.conv_0.weight_u", "up_4.conv_0.bias", "up_4.conv_0.weight_orig", "up_4.conv_0.weight_u", "up_4.conv_0.weight_v", "up_4.conv_1.weight_orig", "up_4.conv_1.weight", "up_4.conv_1.weight_u", "up_4.conv_1.bias", "up_4.conv_1.weight_orig", "up_4.conv_1.weight_u", "up_4.conv_1.weight_v", "up_4.conv_s.weight_orig", "up_4.conv_s.weight", "up_4.conv_s.weight_u", "up_4.conv_s.weight_orig", "up_4.conv_s.weight_u", "up_4.conv_s.weight_v", "up_4.norm_0.noise_scale", "up_4.norm_0.conv_shared.0.weight", "up_4.norm_0.conv_shared.0.bias", "up_4.norm_0.conv_gamma.weight", "up_4.norm_0.conv_gamma.bias", "up_4.norm_0.conv_beta.weight", "up_4.norm_0.conv_beta.bias", "up_4.norm_1.noise_scale", "up_4.norm_1.conv_shared.0.weight", "up_4.norm_1.conv_shared.0.bias", "up_4.norm_1.conv_gamma.weight", "up_4.norm_1.conv_gamma.bias", "up_4.norm_1.conv_beta.weight", "up_4.norm_1.conv_beta.bias", "up_4.norm_s.noise_scale", "up_4.norm_s.conv_shared.0.weight", "up_4.norm_s.conv_shared.0.bias", "up_4.norm_s.conv_gamma.weight", "up_4.norm_s.conv_gamma.bias", "up_4.norm_s.conv_beta.weight", "up_4.norm_s.conv_beta.bias".
size mismatch for conv_img.weight: copying a param with shape torch.Size([3, 64, 3, 3]) from checkpoint, the shape in current model is torch.Size([4, 32, 3, 3]).
size mismatch for conv_img.bias: copying a param with shape torch.Size([3]) from checkpoint, the shape in current model is torch.Size([4]).

python3 test_generator.py --occlusion --test_name test1 --tocg_checkpoint /usr/src/SD-VITON/tocg.pth --gpu_ids 0 --gen_checkpoint /usr/src/SD-VITON/toig.pth --dataroot /usr/src/SD-VITON/data --data_list /usr/src/SD-VITON/data/1.txt --composition_mask

Start to test %s!
Network [SPADEGenerator] was created. Total number of parameters: 100.5 million. To see the architecture, do print(network).
Traceback (most recent call last):
File "/usr/src/SD-VITON/test_generator.py", line 266, in
main()
File "/usr/src/SD-VITON/test_generator.py", line 260, in main
test(opt, test_loader, board, tocg, generator)
File "/usr/src/SD-VITON/test_generator.py", line 115, in test
for inputs in test_loader.data_loader:
File "/usr/local/lib/python3.10/site-packages/torch/utils/data/dataloader.py", line 630, in next
data = self._next_data()
File "/usr/local/lib/python3.10/site-packages/torch/utils/data/dataloader.py", line 1345, in _next_data
return self._process_data(data)
File "/usr/local/lib/python3.10/site-packages/torch/utils/data/dataloader.py", line 1371, in _process_data
data.reraise()
File "/usr/local/lib/python3.10/site-packages/torch/_utils.py", line 694, in reraise
raise exception
RuntimeError: Caught RuntimeError in DataLoader worker process 0.
Original Traceback (most recent call last):
File "/usr/local/lib/python3.10/site-packages/torch/utils/data/_utils/worker.py", line 308, in _worker_loop
data = fetcher.fetch(index)
File "/usr/local/lib/python3.10/site-packages/torch/utils/data/_utils/fetch.py", line 51, in fetch
data = [self.dataset[idx] for idx in possibly_batched_index]
File "/usr/local/lib/python3.10/site-packages/torch/utils/data/utils/fetch.py", line 51, in
data = [self.dataset[idx] for idx in possibly_batched_index]
File "/usr/src/SD-VITON/cp_dataset_test.py", line 164, in getitem
parse_map = parse_map.scatter(0, parse, 1.0)
RuntimeError: Index tensor must have the same number of dimensions as self tensor

Thank you for your great work!

Does this source code support inferencing on custom data?

Hello, author! I used test_generator.py to run and the generated images look like this:

Did I do something wrong?

Is there any chance that your code work on full body with pants and cloth ?

Second question is how can we use with custom images such as human and garments?

Thanks.

I wanted to compare the results of your model checkpoint with others. However, when I use the unpaired data settings, it still seeks clothing on the model (human), as in the paired dataset. Could you please guide me on how to make it work for unpaired datasets?

Hello, I'm trying to run the model with my own data, and I've done all the preprocessing required. (except for the original cloth of my own image, which I don't have, but I edited the code cause I suppose this is not needed.)

However, my image resolution is 2160x2160 and the cloth image in VITON-HD is 768x1024, which gives the error parse_arm = (np.array(mask_arm) / 255) * (parse_array == parse_id).astype(np.float32) ValueError: operands could not be broadcast together with shapes (1024,768) (2160,2160). So I want to ask whether the model supports input of any resolution?

Looking forward to your assist, thanks a lot!

Hey authors,

May I know when will you upload the code for the paper.

Thanks