Colab notebook does not work

[geekmangnu]

Hi, because most people do not have suitable GPU colab file is important. Current colab notebook run gives following error below. It seems because of python version which currently is 3.7 and async is reserved word for python 3.7. Changing pose_model.py file by replacing async to non_blocking resolves it.

Traceback (most recent call last):
File "demo.py", line 10, in
opt = ValOptions().parse()
File "/content/Global-Flow-Local-Attention/options/base_options.py", line 74, in parse
opt = self.gather_options()
File "/content/Global-Flow-Local-Attention/options/base_options.py", line 60, in gather_options
model_option_set = model.get_option_setter(opt.model)
File "/content/Global-Flow-Local-Attention/model/init.py", line 25, in get_option_setter
model = find_model_using_name(model_name)
File "/content/Global-Flow-Local-Attention/model/init.py", line 10, in find_model_using_name
modellib = importlib.import_module(model_file_name)
File "/usr/lib/python3.7/importlib/init.py", line 127, in import_module
return _bootstrap._gcd_import(name[level:], package, level)
File "", line 1006, in _gcd_import
File "", line 983, in _find_and_load
File "", line 967, in _find_and_load_unlocked
File "", line 677, in _load_unlocked
File "", line 724, in exec_module
File "", line 860, in get_code
File "", line 791, in source_to_code
File "", line 219, in _call_with_frames_removed
File "/content/Global-Flow-Local-Attention/model/pose_model.py", line 103
self.input_P1 = input_P1.cuda(self.gpu_ids[0], async=True)
^
SyntaxError: invalid syntax

after this fix it gives another error which seems setup related as follow:

Loading data pairs ...
Loading data pairs finished ...
dataset [FashionDataset] of size 41 was created
val images = 41
Network [PoseGenerator] was created. Total number of parameters: 14.0474 million. To see the architecture, do print(network).
THCudaCheck FAIL file=/pytorch/aten/src/THC/generic/THCTensorMathPairwise.cu line=225 error=11 : invalid argument
Traceback (most recent call last):
File "demo.py", line 15, in
model = create_model(opt)
File "/content/Global-Flow-Local-Attention/model/init.py", line 32, in create_model
instance = model(opt)
File "/content/Global-Flow-Local-Attention/model/pose_model.py", line 64, in init
norm='instance', activation='LeakyReLU', extractor_kz=opt.kernel_size)
File "/content/Global-Flow-Local-Attention/model/networks/init.py", line 37, in define_g
return create_network(netG_cls, opt, **parameter_dic)
File "/content/Global-Flow-Local-Attention/model/networks/init.py", line 28, in create_network
net.init_weights(opt.init_type)
File "/content/Global-Flow-Local-Attention/model/networks/base_network.py", line 55, in init_weights
self.apply(init_func)
File "/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py", line 242, in apply
module.apply(fn)
File "/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py", line 242, in apply
module.apply(fn)
File "/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py", line 242, in apply
module.apply(fn)
[Previous line repeated 1 more time]
File "/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py", line 243, in apply
fn(self)
File "/content/Global-Flow-Local-Attention/model/networks/base_network.py", line 47, in init_func
init.orthogonal_(m.weight.data, gain=gain)
File "/usr/local/lib/python3.7/dist-packages/torch/nn/init.py", line 356, in orthogonal_
q, r = torch.qr(flattened)
RuntimeError: cuda runtime error (11) : invalid argument at /pytorch/aten/src/THC/generic/THCTensorMathPairwise.cu:225

[radhakrishr36]

I face the same issues ..How to resolve this error?

RuntimeError: cuda runtime error (11) : invalid argument at /pytorch/aten/src/THC/generic/THCTensorMathPairwise.cu:225

[RenYurui]

Sorry for the delay!
I think it is related to the CUDA version. The Colab CUDA has been improved for the latest.
I will fix this as soon as possible.

[geekmangnu]

Sorry for the delay!
I think it is related to the CUDA version. The Colab CUDA has been improved for the latest.
I will fix this as soon as possible.

Thank you looking forward to try on colab.

[whuzjw]

Sorry for the delay!
I think it is related to the CUDA version. The Colab CUDA has been improved for the latest.
I will fix this as soon as possible.

Looking forward to the new version

[amrzv]

ERROR: Could not find a version that satisfies the requirement torch==1.0.0 (from versions: 1.11.0, 1.12.0, 1.12.1, 1.13.0, 1.13.1, 2.0.0, 2.0.1, 2.1.0, 2.1.1, 2.1.2)
ERROR: No matching distribution found for torch==1.0.0