huanqingxu

'relu_slope': 0.2, 'seed': 1, 'validation_every_epoch': 1, 'volume_shape': [144, 192, 256], 'weight_decay': 0, 'x_test': '', 'x_train': '', 'x_valid': '', 'y_test': '', 'y_train': '', 'y_valid': ''} *************************************** The experiment name is unet_experiment...

i got the same, how's going

qr = np.einsum('bgci,cij->bgij', q, q_embedding) File "", line 6, in einsum File "/home/lab549/anaconda3/envs/medt/lib/python3.6/site-packages/numpy/core/einsumfunc.py", line 1356, in einsum return c_einsum(*operands, **kwargs) ValueError: operands could not be broadcast together with remapped shapes...

1000, 1000, 3) (256, 256, 1) torch.Size([500, 8, 1, 500]) torch.Size([1, 64, 64]) torch.Size([500, 8, 1, 500]) Traceback (most recent call last): File "/media/lab549/Data/Medical-Transformer-main/train.py", line 140, in output = model(X_batch)...

well done

Hello, developer! Thank you very much for your shared code. is it possible to share the preprocessed data?