!python /content/DragGAN/visualizer_drag_gradio.py Problem

[seimuron1]

Hi. I have a problem to start the interface of DragGan AI after running command "!python /content/DragGAN/visualizer_drag_gradio.py" :

File under cache_dir (./checkpoints):
['stylegan2-cat-config-f.pkl', 'stylegan2_lions_512_pytorch.pkl', 'stylegan2-afhqcat-512x512.pkl', 'stylegan2-car-config-f.pkl', 'stylegan2_dogs_1024_pytorch.pkl', 'stylegan2-ffhq-512x512.pkl', 'stylegan2_elephants_512_pytorch.pkl', 'stylegan2_horses_256_pytorch.pkl']
Valid checkpoint file:
{'stylegan2-cat-config-f': './checkpoints/stylegan2-cat-config-f.pkl', 'stylegan2_lions_512_pytorch': './checkpoints/stylegan2_lions_512_pytorch.pkl', 'stylegan2-afhqcat-512x512': './checkpoints/stylegan2-afhqcat-512x512.pkl', 'stylegan2-car-config-f': './checkpoints/stylegan2-car-config-f.pkl', 'stylegan2_dogs_1024_pytorch': './checkpoints/stylegan2_dogs_1024_pytorch.pkl', 'stylegan2-ffhq-512x512': './checkpoints/stylegan2-ffhq-512x512.pkl', 'stylegan2_elephants_512_pytorch': './checkpoints/stylegan2_elephants_512_pytorch.pkl', 'stylegan2_horses_256_pytorch': './checkpoints/stylegan2_horses_256_pytorch.pkl'}
Loading "./checkpoints/stylegan2_lions_512_pytorch.pkl"... Done.
()
{'z_dim': 512, 'c_dim': 0, 'w_dim': 512, 'img_resolution': 512, 'img_channels': 3, 'mapping_kwargs': {'num_layers': 8, 'embed_features': None, 'layer_features': None, 'activation': 'lrelu', 'lr_multiplier': 0.01, 'w_avg_beta': 0.995}, 'synthesis_kwargs': {'channel_base': 32768, 'channel_max': 512, 'num_fp16_res': 0, 'conv_clamp': None, 'architecture': 'skip', 'resample_filter': [1, 3, 3, 1], 'use_noise': True, 'activation': 'lrelu'}}
Traceback (most recent call last):
File "/content/DragGAN/visualizer_drag_gradio.py", line 205, in
global_state = init_images(global_state)
File "/content/DragGAN/visualizer_drag_gradio.py", line 75, in init_images
state['renderer'].init_network(
File "/content/DragGAN/viz/renderer.py", line 224, in init_network
res.img_resolution = G.img_resolution
AttributeError: 'dict' object has no attribute 'img_resolution'

Also I was running the "!python scripts/download_model.py" command instead of "!sh scripts/download_model.sh".
Can someone help me? Im absolutely newbie in programming, just want to test this AI

Thanks

[rajnish159]

same issue😒😒

[woshixty]

me too

[woshixty]

I solved this problem!

• pip install draggan
• pip3 install --pre torch torchvision torchaudio --index-url https://download.pytorch.org/whl/nightly/cu118
• pip install Pillow==9.5.0
  After installing the above three software packages, you can run DragGAN normally.

[rajnish159]

pip3 install --pre torch torchvision torchaudio --index-url https://download.pytorch.org/whl/nightly/cu118
bro this is not working. please help me

[rajnish159]

I solved this problem!

• pip install draggan
  !pip3 install --pre torch torchvision torchaudio -f https://download.pytorch.org/whl/nightly/cu118/torch_nightly.html
• pip install Pillow==9.5.0
  After installing the above three software packages, you can run DragGAN normally.

thanks, this is properly working.

[OcheaDumi]

Hello Everybod,
This did not get solved with the installation, I still face the same error

docker run -p 7860:7860 -v "$PWD":/workspace/src -it draggan:latest bash
root@a7ef1370a172:/workspace# cd src && python visualizer_drag_gradio.py --listen
File under cache_dir (./checkpoints):
['stylegan2-ffhq-512x512.pkl', 'stylegan2_dogs_1024_pytorch.pkl', 'stylegan2-car-config-f.pkl', 'stylegan2_elephants_512_pytorch.pkl', 'stylegan2_lions_512_pytorch.pkl', 'stylegan2-cat-config-f.pkl', 'stylegan2-afhqcat-512x512.pkl', 'stylegan2_horses_256_pytorch.pkl']
Valid checkpoint file:
{'stylegan2-ffhq-512x512': './checkpoints/stylegan2-ffhq-512x512.pkl', 'stylegan2_dogs_1024_pytorch': './checkpoints/stylegan2_dogs_1024_pytorch.pkl', 'stylegan2-car-config-f': './checkpoints/stylegan2-car-config-f.pkl', 'stylegan2_elephants_512_pytorch': './checkpoints/stylegan2_elephants_512_pytorch.pkl', 'stylegan2_lions_512_pytorch': './checkpoints/stylegan2_lions_512_pytorch.pkl', 'stylegan2-cat-config-f': './checkpoints/stylegan2-cat-config-f.pkl', 'stylegan2-afhqcat-512x512': './checkpoints/stylegan2-afhqcat-512x512.pkl', 'stylegan2_horses_256_pytorch': './checkpoints/stylegan2_horses_256_pytorch.pkl'}
Loading "./checkpoints/stylegan2_lions_512_pytorch.pkl"... Done.
()
{'z_dim': 512, 'c_dim': 0, 'w_dim': 512, 'img_resolution': 512, 'img_channels': 3, 'mapping_kwargs': {'num_layers': 8, 'embed_features': None, 'layer_features': None, 'activation': 'lrelu', 'lr_multiplier': 0.01, 'w_avg_beta': 0.995}, 'synthesis_kwargs': {'channel_base': 32768, 'channel_max': 512, 'num_fp16_res': 0, 'conv_clamp': None, 'architecture': 'skip', 'resample_filter': [1, 3, 3, 1], 'use_noise': True, 'activation': 'lrelu'}}
Traceback (most recent call last):
File "/workspace/src/visualizer_drag_gradio.py", line 205, in
global_state = init_images(global_state)
File "/workspace/src/visualizer_drag_gradio.py", line 75, in init_images
state['renderer'].init_network(
File "/workspace/src/viz/renderer.py", line 224, in init_network
res.img_resolution = G.img_resolution
AttributeError: 'dict' object has no attribute 'img_resolution'
root@a7ef1370a172:/workspace/src# # before you build the docker container, make sure you have cloned this repo, and downloaded the pretrained model by python scripts/download_model.py.
docker build . -t draggan:latest
docker run -p 7860:7860 -v "$PWD":/workspace/src -it draggan:latest bash
cd src && python visualizer_drag_gradio.py --listen
bash: docker: command not found
bash: docker: command not found
bash: cd: src: No such file or directory
root@a7ef1370a172:/workspace/src# python visualizer_drag_gradio.py
File under cache_dir (./checkpoints):
['stylegan2-ffhq-512x512.pkl', 'stylegan2_dogs_1024_pytorch.pkl', 'stylegan2-car-config-f.pkl', 'stylegan2_elephants_512_pytorch.pkl', 'stylegan2_lions_512_pytorch.pkl', 'stylegan2-cat-config-f.pkl', 'stylegan2-afhqcat-512x512.pkl', 'stylegan2_horses_256_pytorch.pkl']
Valid checkpoint file:
{'stylegan2-ffhq-512x512': './checkpoints/stylegan2-ffhq-512x512.pkl', 'stylegan2_dogs_1024_pytorch': './checkpoints/stylegan2_dogs_1024_pytorch.pkl', 'stylegan2-car-config-f': './checkpoints/stylegan2-car-config-f.pkl', 'stylegan2_elephants_512_pytorch': './checkpoints/stylegan2_elephants_512_pytorch.pkl', 'stylegan2_lions_512_pytorch': './checkpoints/stylegan2_lions_512_pytorch.pkl', 'stylegan2-cat-config-f': './checkpoints/stylegan2-cat-config-f.pkl', 'stylegan2-afhqcat-512x512': './checkpoints/stylegan2-afhqcat-512x512.pkl', 'stylegan2_horses_256_pytorch': './checkpoints/stylegan2_horses_256_pytorch.pkl'}
Loading "./checkpoints/stylegan2_lions_512_pytorch.pkl"... Done.
()
{'z_dim': 512, 'c_dim': 0, 'w_dim': 512, 'img_resolution': 512, 'img_channels': 3, 'mapping_kwargs': {'num_layers': 8, 'embed_features': None, 'layer_features': None, 'activation': 'lrelu', 'lr_multiplier': 0.01, 'w_avg_beta': 0.995}, 'synthesis_kwargs': {'channel_base': 32768, 'channel_max': 512, 'num_fp16_res': 0, 'conv_clamp': None, 'architecture': 'skip', 'resample_filter': [1, 3, 3, 1], 'use_noise': True, 'activation': 'lrelu'}}
Traceback (most recent call last):
File "/workspace/src/visualizer_drag_gradio.py", line 205, in
global_state = init_images(global_state)
File "/workspace/src/visualizer_drag_gradio.py", line 75, in init_images
state['renderer'].init_network(
File "/workspace/src/viz/renderer.py", line 224, in init_network
res.img_resolution = G.img_resolution
AttributeError: 'dict' object has no attribute 'img_resolution'

Any ideas?

Thanks
I have tried to reinstall the docker all together, drivers, toolkit, the whole app from the beginning, it is the same outcome - if I run --gpus all it does not recognize gpu, if i run it without .... i get this error

[prajwalsingh]

Hi, if someone still has the problem, you can fix it by following:

Add/replace the following with code at line 70 in visualizer_drag_gradio

    if isinstance(global_state, gr.State):
        state = dnnlib.EasyDict(global_state.value)
    else:
        state = dnnlib.EasyDict(global_state)

    for key, item in state.items():
        if isinstance(item, dict):
            state[key] = dnnlib.EasyDict(item)

Add the following code before returning in function init_images (visualizer_drag_gradio):

global_state.value = state
return global_state

[dm4digitalmedia]

/unsubscribe

[Harry19s]

when i run python visualizer_drag_gradio.py
report issue: AttributeError: 'dict' object has no attribute 'img_resolution
how do i solve this issue?

I don't know why but I solved this issue by

pip install gradio==3.36.1

this was also useful for pydantic error

[kriswangchunyan]

when i run python visualizer_drag_gradio.py
report issue: AttributeError: 'dict' object has no attribute 'img_resolution
how do i solve this issue?

I don't know why but I solved this issue by

pip install gradio==3.36.1

this was also useful for pydantic error

thanks，i solved this error and running now