torch shape error

[fire537]

Hello, after setting up the environment locally, I tested using the fasta files in the Data directory without any issues. However, when I run it with my own fasta files, I encounter some errors：
Traceback (most recent call last):
File "./Active_sites/main.py", line 14, in
predict(model_class, args)
File "/tools/GraphEC/Active_sites/utils.py", line 84, in predict
outputs = [model(data.X, data.node_feat, data.edge_index, data.seq, data.batch).sigmoid() for model in models]
File "/tools/GraphEC/Active_sites/utils.py", line 84, in
outputs = [model(data.X, data.node_feat, data.edge_index, data.seq, data.batch).sigmoid() for model in models]
File "/conda_envs/esmfold/lib/python3.7/site-packages/torch/nn/modules/module.py", line 1194, in _call_impl
return forward_call(*input, **kwargs)
File "/tools/GraphEC/Active_sites/model.py", line 151, in forward
h_V = torch.cat([h_V, h_V_geo], dim=-1)
RuntimeError: Sizes of tensors must match except in dimension 1. Expected size 1 but got size 213 for tensor number 1 in the list.

[YidongSong]

This error occurred when utilizing the torch.cat function. please verify the data dimensions.

[morant12]

我也在处理NEW392第355个id时出现了相同的问题，我通过补充了一个维度全为1再进行拼接。我在使用预测NEW392过程中遇到了line 279, in _get_direction_orientation
t = F.normalize(X[:, [0, 2, 3, 4]] - X_Ca.unsqueeze(1), dim=-1)
RuntimeError: CUDA error: device-side assert triggered
CUDA kernel errors might be asynchronously reported at some other API call,so the stacktrace below might be incorrect.
For debugging consider passing CUDA_LAUNCH_BLOCKING=1.