问即使没有使用图像，PyTorch也需要在DataLoader中具有类似图像的维度

EN

      column_1  column_2  or  and  xor
0            1         1   1    1    0
1            0         1   1    0    1
2            0         1   1    0    1
3            0         1   1    0    1
4            1         0   1    0    1
...        ...       ...  ..  ...  ...
9995         1         1   1    1    0
9996         1         0   1    0    1
9997         1         0   1    0    1
9998         0         1   1    0    1
9999         0         1   1    0    1

# Imports
from pandas import read_csv, DataFrame
import torch
from torch.utils.data import Dataset, DataLoader
import torch.nn as torch_nn
import torch.nn.functional as torch_functional
import torchvision
import random


# Classes
class CustomDataset(Dataset):
    def __init__(self, data, target_column_name, transform=None):
        self.dataframe = data
        self.x = data[['column_1', 'column_2']].values
        self.y = data[[target_column_name]].values
        self.n_samples = len(data)
        self.transform = transform

    def __len__(self):
        return self.n_samples

    def __getitem__(self, index):
        x = self.x[index]
        y = self.y[index]

        if not self.transform == None:
            return (self.transform(x), self.transform(y))
        return (x, y)

class CustomNet(torch_nn.Module):
    def __init__(self):
        super().__init__()
        self.fully_connected_input_layer = torch_nn.Linear(2, 5)
        self.fully_connected_hidden_layer_1 = torch_nn.Linear(5, 5)
        self.fully_connected_hidden_layer_2 = torch_nn.Linear(5, 5)
        self.fully_connected_output_layer = torch_nn.Linear(5, 2)

    def forward(self, data):
        data = torch_functional.relu(self.fully_connected_input_layer(data))
        data = torch_functional.relu(self.fully_connected_hidden_layer_1(data))
        data = torch_functional.relu(self.fully_connected_hidden_layer_2(data))
        data = torch_functional.log_softmax(
            self.fully_connected_output_layer(data), 
            dim=1
        )
        # data = data.squeeze(1)

        return data


# Global variables
NUMBER_OF_OBSERVATIONS = 10000
N_TRAIN_OBSERVATIONS = 7000
BATCH_SIZE = 4
N_EPOCHS = 3
random.seed(42)


# Generating logical gate data
## Generating two columns with random 0s and 1s.
df_data = DataFrame({
    'column_1': random.choices([0, 1], k=NUMBER_OF_OBSERVATIONS),
    'column_2': random.choices([0, 1], k=NUMBER_OF_OBSERVATIONS)
})
## Adding the logic gate results of the previously generated two columns.
df_data.loc[:,'or'] = (df_data['column_1'] == 1) | (df_data['column_2'] == 1)
df_data.loc[:,'or'] = df_data.loc[:,'or'].astype(int)
df_data.loc[:,'and'] = (df_data['column_1'] == 1) & (df_data['column_2'] == 1).astype(int)
df_data.loc[:,'and'] = df_data.loc[:,'and'].astype(int)
df_data.loc[:,'xor'] = ((df_data['column_1'] == 1) & (df_data['column_2'] != 1)) | ((df_data['column_1'] != 1) & (df_data['column_2'] == 1)).astype(int)
df_data.loc[:,'xor'] = df_data.loc[:,'xor'].astype(int)
print(df_data.info())
print(df_data)

# Instantiating a CustomDataSet object and a DataLoader object.
dataset_data = CustomDataset(df_data, target_column_name='xor', transform=torchvision.transforms.ToTensor())
dataset_data_train, dataset_data_test = torch.utils.data.random_split(
    dataset_data,
    [N_TRAIN_OBSERVATIONS, len(dataset_data)-N_TRAIN_OBSERVATIONS]
)
dataloader_data_train = DataLoader(
    dataset=dataset_data_train, 
    batch_size=BATCH_SIZE, 
    shuffle=True
)
dataloader_data_test = DataLoader(
    dataset=dataset_data_test, 
    batch_size=BATCH_SIZE, 
    shuffle=True
)

# Instantiating the neural network
custom_net = CustomNet()

# Running one epoch
for epoch in range(N_EPOCHS):
    for data in dataloader_data_train:
        X, y = data[0].float(), data[1].float()
        net.zero_grad()
        output = net(X)
        print(output)
        print(y)
        loss = torch_functional.nll_loss(output, y)
        loss.backward()
        optimizer.step()
        break

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-2-f2833deb4b11> in <module>
     95 # Running one epoch
     96 for epoch in range(N_EPOCHS):
---> 97     for data in dataloader_data_train:
     98         X, y = data[0].float(), data[1].float()
     99         net.zero_grad()

~/anaconda3/lib/python3.7/site-packages/torch/utils/data/dataloader.py in __next__(self)
    343 
    344     def __next__(self):
--> 345         data = self._next_data()
    346         self._num_yielded += 1
    347         if self._dataset_kind == _DatasetKind.Iterable and \

~/anaconda3/lib/python3.7/site-packages/torch/utils/data/dataloader.py in _next_data(self)
    383     def _next_data(self):
    384         index = self._next_index()  # may raise StopIteration
--> 385         data = self._dataset_fetcher.fetch(index)  # may raise StopIteration
    386         if self._pin_memory:
    387             data = _utils.pin_memory.pin_memory(data)

~/anaconda3/lib/python3.7/site-packages/torch/utils/data/_utils/fetch.py in fetch(self, possibly_batched_index)
     42     def fetch(self, possibly_batched_index):
     43         if self.auto_collation:
---> 44             data = [self.dataset[idx] for idx in possibly_batched_index]
     45         else:
     46             data = self.dataset[possibly_batched_index]

~/anaconda3/lib/python3.7/site-packages/torch/utils/data/_utils/fetch.py in <listcomp>(.0)
     42     def fetch(self, possibly_batched_index):
     43         if self.auto_collation:
---> 44             data = [self.dataset[idx] for idx in possibly_batched_index]
     45         else:
     46             data = self.dataset[possibly_batched_index]

~/anaconda3/lib/python3.7/site-packages/torch/utils/data/dataset.py in __getitem__(self, idx)
    255 
    256     def __getitem__(self, idx):
--> 257         return self.dataset[self.indices[idx]]
    258 
    259     def __len__(self):

<ipython-input-2-f2833deb4b11> in __getitem__(self, index)
     26 
     27         if not self.transform == None:
---> 28             return (self.transform(x), self.transform(y))
     29         return (x, y)
     30 

~/anaconda3/lib/python3.7/site-packages/torchvision/transforms/transforms.py in __call__(self, pic)
     90             Tensor: Converted image.
     91         """
---> 92         return F.to_tensor(pic)
     93 
     94     def __repr__(self):

~/anaconda3/lib/python3.7/site-packages/torchvision/transforms/functional.py in to_tensor(pic)
     43 
     44     if _is_numpy(pic) and not _is_numpy_image(pic):
---> 45         raise ValueError('pic should be 2/3 dimensional. Got {} dimensions.'.format(pic.ndim))
     46 
     47     if isinstance(pic, np.ndarray):

ValueError: pic should be 2/3 dimensional. Got 1 dimensions.