RuntimeError: view size is not compatible with input tensor's size and stride (at least one dimension spans across two contiguous subspaces). Use .reshape(...) instead.

[dcbiton]

2024-09-02 15:42:54,465 [] [INFO] Loading the model
2024-09-02 15:42:55,222 [] [INFO] Model has been loaded
  0%|                                                                         | 0/10 [00:00<?, ?it/s]INPZ torch.Size([1, 488, 488, 128])
MASK torch.Size([1, 488, 488])
  0%|                                                                         | 0/10 [00:00<?, ?it/s]
Traceback (most recent call last):
  File "/Users/dionessa/Documents/cryptochrome_codes/alphaflow/predict.py", line 183, in <module>
    main()
  File "/opt/anaconda3/envs/alphaflow/lib/python3.9/site-packages/torch/autograd/grad_mode.py", line 27, in decorate_context
    return func(*args, **kwargs)
  File "/Users/dionessa/Documents/cryptochrome_codes/alphaflow/predict.py", line 129, in main
    prots, outputs = model.inference(batch, as_protein=False, noisy_first=args.noisy_first,
  File "/Users/dionessa/Documents/cryptochrome_codes/alphaflow/alphaflow/model/wrapper.py", line 363, in inference
    output = self.model(batch)
  File "/opt/anaconda3/envs/alphaflow/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1130, in _call_impl
    return forward_call(*input, **kwargs)
  File "/Users/dionessa/Documents/cryptochrome_codes/alphaflow/alphaflow/model/alphafold.py", line 223, in forward
    inp_z = self._get_input_pair_embeddings(
  File "/Users/dionessa/Documents/cryptochrome_codes/alphaflow/alphaflow/model/alphafold.py", line 130, in _get_input_pair_embeddings
    inp_z = self.input_pair_stack(inp_z, mask, chunk_size=None)
  File "/opt/anaconda3/envs/alphaflow/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1130, in _call_impl
    return forward_call(*input, **kwargs)
  File "/Users/dionessa/Documents/cryptochrome_codes/alphaflow/alphaflow/model/input_stack.py", line 276, in forward
    t, = checkpoint_blocks(
  File "/opt/anaconda3/envs/alphaflow/lib/python3.9/site-packages/openfold/utils/checkpointing.py", line 85, in checkpoint_blocks
    return exec(blocks, args)
  File "/opt/anaconda3/envs/alphaflow/lib/python3.9/site-packages/openfold/utils/checkpointing.py", line 72, in exec
    a = wrap(block(*a))
  File "/opt/anaconda3/envs/alphaflow/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1130, in _call_impl
    return forward_call(*input, **kwargs)
  File "/Users/dionessa/Documents/cryptochrome_codes/alphaflow/alphaflow/model/input_stack.py", line 122, in forward
    self.tri_att_end(
  File "/opt/anaconda3/envs/alphaflow/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1130, in _call_impl
    return forward_call(*input, **kwargs)
  File "/opt/anaconda3/envs/alphaflow/lib/python3.9/site-packages/openfold/model/triangular_attention.py", line 114, in forward
    x = self.layer_norm(x)
  File "/opt/anaconda3/envs/alphaflow/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1130, in _call_impl
    return forward_call(*input, **kwargs)
  File "/opt/anaconda3/envs/alphaflow/lib/python3.9/site-packages/openfold/model/primitives.py", line 211, in forward
    out = nn.functional.layer_norm(
  File "/opt/anaconda3/envs/alphaflow/lib/python3.9/site-packages/torch/nn/functional.py", line 2503, in layer_norm
    return torch.layer_norm(input, normalized_shape, weight, bias, eps, torch.backends.cudnn.enabled)
RuntimeError: view size is not compatible with input tensor's size and stride (at least one dimension spans across two contiguous subspaces). Use .reshape(...) instead.