size error

[badreddinemerabet]

when i tried to run the example from the youtube video "https://www.youtube.com/watch?v=-Grfxkg3-DI" I got this error when I try to run the training cell
Epoch 1:

RuntimeError Traceback (most recent call last)
in <cell line: 46>()
47 # train
48 print(f"Epoch {epoch}:")
---> 49 train(dataflow, model, device, optimizer)
50 print(optimizer.param_groups[0]['lr'])
51

13 frames
/content/torchquantum/torchquantum/functional/gate_wrapper.py in apply_unitary_bmm(state, mat, wires)
129 if len(mat.shape) > 2:
130 # both matrix and state are in batch mode
--> 131 new_state = mat.bmm(permuted)
132 else:
133 # matrix no batch, state in batch mode

RuntimeError: Expected size for first two dimensions of batch2 tensor to be: [10, 2] but got: [1, 2].

i didn't change anything

[01110011011101010110010001101111]

Try https://github.com/mit-han-lab/torchquantum/tree/main/examples/quanvolution (with #147 to increase accuracy); I’ll ask about updating the link!

[badreddinemerabet]

thank you, it worked

[badreddinemerabet]

Define transformations for the dataset

transform = transforms.Compose([
transforms.Grayscale(num_output_channels=1),
transforms.Resize((28, 28)),
transforms.ToTensor(),
transforms.Normalize((0.5,), (0.5,))
])

Load your dataset

data_dir = 'OAM28' # Replace with the path to your dataset
full_dataset = datasets.ImageFolder(root=data_dir, transform=transform)

Split dataset into train and test

train_size = int(0.9 * len(full_dataset))
test_size = len(full_dataset) - train_size
train_dataset, test_dataset = random_split(full_dataset, [train_size, test_size])

Create data loaders

batch_size = 16
train_loader = DataLoader(train_dataset, batch_size=batch_size, shuffle=True, num_workers=8, pin_memory=True)
test_loader = DataLoader(test_dataset, batch_size=batch_size, shuffle=False, num_workers=8, pin_memory=True)

dataflow = {'train': train_loader, 'test': test_loader}