RuntimeError: Given groups=1, weight of size 16 3 1 1, expected input[3, 1, 128, 128] to have 3 channels, but got 1 channels instead

[josephrocca]

RuntimeError: Given groups=1, weight of size 16 3 1 1, expected input[3, 1, 128, 128] to have 3 channels, but got 1 channels instead

[lucidrains]

@josephrocca this is very helpful actually! it is my goal to let everyone experience this phenomenon easily :)

[josephrocca]

Oh, I've just converted them to jpgs and it seems to be working now!

mogrify -format jpg ./twemoji/assets/72x72/*.png
rm ./twemoji/assets/72x72/*.png

[lucidrains]

ok, try again! ff09e94

[josephrocca]

All working fine now! Thanks again for this! :)

[josephrocca]

I tried resizing the twemoji images to 256x256 like so:

sudo apt-get install imagemagick
git clone https://github.com/twitter/twemoji.git
mogrify -resize 256x256 ./twemoji/assets/72x72/*.png
stylegan2_pytorch --data ./twemoji/assets/72x72

But that didn't help, although the error message was slightly different:

RuntimeError: Given groups=1, weight of size 16 3 1 1, expected input[3, 4, 128, 128] to have 3 channels, but got 4 channels instead

[josephrocca]

@lucidrains Perhaps an error message could be shown to tell people that they can't use png images? I assumed it would be as easy as (something like) this (stylegan2_pytorch.py#L133):

    def __getitem__(self, index):
        path = self.paths[index]
        img = Image.open(path)
        img.convert('RGB')
        img.save('__tmp395739487394.jpg')
        jpg = Image.open('__tmp395739487394.jpg')
        return self.transform(jpg)

But that still results in the same error for some reason (using the mnist_png):

RuntimeError: Given groups=1, weight of size 16 3 1 1, expected input[3, 1, 128, 128] to have 3 channels, but got 1 channels instead

and this on the twemoji dataset:

UserWarning: Palette images with Transparency expressed in bytes should be converted to RGBA images

[lucidrains]

@josephrocca I made another patch! upgrade the package!

[josephrocca]

Works for the mnist_png code now! Getting this for the twemoji example:

  0%|                                                          | 0/100000 [00:00<?, ?it/s]G: -3314.26 | D: 52.87 | GP: 103.12 | PL: 0.20
  0%|                                              | 15/100000 [00:10<18:33:42,  1.50it/s]Traceback (most recent call last):
  File "/home/user/miniconda/envs/py36/bin/stylegan2_pytorch", line 56, in <module>
    fire.Fire(train_from_folder)
  File "/home/user/miniconda/envs/py36/lib/python3.6/site-packages/fire/core.py", line 138, in Fire
    component_trace = _Fire(component, args, parsed_flag_args, context, name)
  File "/home/user/miniconda/envs/py36/lib/python3.6/site-packages/fire/core.py", line 471, in _Fire
    target=component.__name__)
  File "/home/user/miniconda/envs/py36/lib/python3.6/site-packages/fire/core.py", line 675, in _CallAndUpdateTrace
    component = fn(*varargs, **kwargs)
  File "/home/user/miniconda/envs/py36/bin/stylegan2_pytorch", line 51, in train_from_folder
    model.train()
  File "/home/user/miniconda/envs/py36/lib/python3.6/site-packages/stylegan2_pytorch/stylegan2_pytorch.py", line 459, in train
    image_batch = next(self.loader).cuda()
  File "/home/user/miniconda/envs/py36/lib/python3.6/site-packages/stylegan2_pytorch/stylegan2_pytorch.py", line 75, in cycle
    for i in iterable:
  File "/home/user/miniconda/envs/py36/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 345, in __next__
    data = self._next_data()
  File "/home/user/miniconda/envs/py36/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 856, in _next_data
    return self._process_data(data)
  File "/home/user/miniconda/envs/py36/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 881, in _process_data
    data.reraise()
  File "/home/user/miniconda/envs/py36/lib/python3.6/site-packages/torch/_utils.py", line 394, in reraise
    raise self.exc_type(msg)
RuntimeError: Caught RuntimeError in DataLoader worker process 13.
Original Traceback (most recent call last):
  File "/home/user/miniconda/envs/py36/lib/python3.6/site-packages/torch/utils/data/_utils/worker.py", line 178, in _worker_loop
    data = fetcher.fetch(index)
  File "/home/user/miniconda/envs/py36/lib/python3.6/site-packages/torch/utils/data/_utils/fetch.py", line 44, in fetch
    data = [self.dataset[idx] for idx in possibly_batched_index]
  File "/home/user/miniconda/envs/py36/lib/python3.6/site-packages/torch/utils/data/_utils/fetch.py", line 44, in <listcomp>
    data = [self.dataset[idx] for idx in possibly_batched_index]
  File "/home/user/miniconda/envs/py36/lib/python3.6/site-packages/stylegan2_pytorch/stylegan2_pytorch.py", line 140, in __getitem__
    return self.transform(img)
  File "/home/user/miniconda/envs/py36/lib/python3.6/site-packages/torchvision/transforms/transforms.py", line 70, in __call__
    img = t(img)
  File "/home/user/miniconda/envs/py36/lib/python3.6/site-packages/torchvision/transforms/transforms.py", line 322, in __call__
    return self.lambd(img)
  File "/home/user/miniconda/envs/py36/lib/python3.6/site-packages/stylegan2_pytorch/stylegan2_pytorch.py", line 112, in expand_to_rgb
    return tensor.expand(3, -1, -1)
RuntimeError: The expanded size of the tensor (3) must match the existing size (4) at non-singleton dimension 0.  Target sizes: [3, -1, -1].  Tensor sizes: [4, 128, 128]

  0%|                                              | 15/100000 [00:13<24:54:58,  1.11it/s]
Exception in thread Thread-2:
Traceback (most recent call last):
  File "/home/user/miniconda/envs/py36/lib/python3.6/threading.py", line 916, in _bootstrap_inner
    self.run()
  File "/home/user/miniconda/envs/py36/lib/python3.6/threading.py", line 864, in run
    self._target(*self._args, **self._kwargs)
  File "/home/user/miniconda/envs/py36/lib/python3.6/site-packages/torch/utils/data/_utils/pin_memory.py", line 25, in _pin_memory_loop
    r = in_queue.get(timeout=MP_STATUS_CHECK_INTERVAL)
  File "/home/user/miniconda/envs/py36/lib/python3.6/multiprocessing/queues.py", line 113, in get
    return _ForkingPickler.loads(res)
  File "/home/user/miniconda/envs/py36/lib/python3.6/site-packages/torch/multiprocessing/reductions.py", line 294, in rebuild_storage_fd
    fd = df.detach()
  File "/home/user/miniconda/envs/py36/lib/python3.6/multiprocessing/resource_sharer.py", line 57, in detach
    with _resource_sharer.get_connection(self._id) as conn:
  File "/home/user/miniconda/envs/py36/lib/python3.6/multiprocessing/resource_sharer.py", line 87, in get_connection
    c = Client(address, authkey=process.current_process().authkey)
  File "/home/user/miniconda/envs/py36/lib/python3.6/multiprocessing/connection.py", line 493, in Client
    answer_challenge(c, authkey)
  File "/home/user/miniconda/envs/py36/lib/python3.6/multiprocessing/connection.py", line 732, in answer_challenge
    message = connection.recv_bytes(256)         # reject large message
  File "/home/user/miniconda/envs/py36/lib/python3.6/multiprocessing/connection.py", line 216, in recv_bytes
    buf = self._recv_bytes(maxlength)
  File "/home/user/miniconda/envs/py36/lib/python3.6/multiprocessing/connection.py", line 407, in _recv_bytes
    buf = self._recv(4)
  File "/home/user/miniconda/envs/py36/lib/python3.6/multiprocessing/connection.py", line 379, in _recv
    chunk = read(handle, remaining)
ConnectionResetError: [Errno 104] Connection reset by peer

Exact replication:

sudo docker run --gpus all --rm -it anibali/pytorch:cuda-10.1 bash
pip install stylegan2_pytorch
git clone https://github.com/twitter/twemoji.git
stylegan2_pytorch --data ./twemoji/assets/72x72

This is not important at all though (very easy to convert to jpg) - only reporting to help make your package super easy for others to get started with. I'm worried that I'm taking your time away from more important things with these nit picks 😅