RuntimeError: Expected 4-dimensional input for 4-dimensional weight 64, but got 1-dimensional input of size [32] instead

I would appreciate to know how this error can be fixed.
This is the code:

%matplotlib inline
import matplotlib.pyplot as plt
import numpy as np

import torch
import torch.nn as nn
import torch.nn.functional as F

from torchvision.datasets import ImageFolder, MNIST
from torchvision import transforms
from torch import autograd
from torch.autograd import Variable
from torchvision.utils import make_grid
from tensorboardX import SummaryWriter
batch_size = 32
cuda = True if torch.cuda.is_available() else False
d= 128
transform = transforms.Compose([
transforms.ToTensor(), transforms.Normalize([0.5], [0.5])])
data_loader = torch.utils.data.DataLoader(MNIST('data', train=True, download=True, transform=transform),
                                          batch_size=batch_size, shuffle=True)
class Generator(nn.Module):
    # initializers
    def __init__(self, d=128):
        super(Generator, self).__init__()
        self.label_emb = nn.Embedding(10, 10)
        self.deconv1_1 = nn.ConvTranspose2d(100, d*2, 4, 1, 0)
        self.deconv1_1_bn = nn.BatchNorm2d(d*2)
        self.deconv1_2 = nn.ConvTranspose2d(10, d*2, 4, 1, 0)
        self.deconv1_2_bn = nn.BatchNorm2d(d*2)
        self.deconv2 = nn.ConvTranspose2d(d*4, d*2, 4, 2, 1)
        self.deconv2_bn = nn.BatchNorm2d(d*2)
        self.deconv3 = nn.ConvTranspose2d(d*2, d, 4, 2, 1)
        self.deconv3_bn = nn.BatchNorm2d(d)
        self.deconv4 = nn.ConvTranspose2d(d, 1, 4, 2, 1)

    # forward method
    def forward(self, input, label):
        x = F.relu(self.deconv1_1_bn(self.deconv1_1(input)))
        y = F.relu(self.deconv1_2_bn(self.deconv1_2(label)))
        x = torch.cat([x, y], 1)
        x = F.relu(self.deconv2_bn(self.deconv2(x)))
        x = F.relu(self.deconv3_bn(self.deconv3(x)))
        x = F.tanh(self.deconv4(x))
        # x = F.relu(self.deconv4_bn(self.deconv4(x)))
        # x = F.tanh(self.deconv5(x))

        return x

class Discriminator(nn.Module):
    # initializers
    def __init__(self, d=128):        
        super(Discriminator, self).__init__()
        self.conv1_1 = nn.Conv2d(1, 64, 4, 2, 1)
        self.conv1_2 = nn.Conv2d(10, 64, 4, 2, 1)
        self.conv2 = nn.Conv2d(d, d*2, 4, 2, 1)
        self.conv2_bn = nn.BatchNorm2d(d*2)
        self.conv3 = nn.Conv2d(d*2, d*4, 4, 2, 1)
        self.conv3_bn = nn.BatchNorm2d(d*4)
        self.conv4 = nn.Conv2d(d * 4, 1, 4, 1, 0)
        
    # forward method
    def forward(self, input, label):
        x = F.leaky_relu(self.conv1_1(input), 0.2)
        y = F.leaky_relu(self.conv1_2(label), 0.2)
        x = torch.cat([x, y], 1)
        x = F.leaky_relu(self.conv2_bn(self.conv2(x)), 0.2)
        x = F.leaky_relu(self.conv3_bn(self.conv3(x)), 0.2)
        x = F.sigmoid(self.conv4(x))

        return x

generator = Generator()
discriminator = Discriminator()
if cuda:
    generator.cuda()
    discriminator.cuda()
criterion = nn.BCELoss()
d_optimizer = torch.optim.Adam(discriminator.parameters(), lr=1e-4)
g_optimizer = torch.optim.Adam(generator.parameters(), lr=1e-4)

def generator_train_step(batch_size, discriminator, generator, g_optimizer, criterion):
    g_optimizer.zero_grad()
    z = Variable(torch.randn(batch_size, 100)).cuda()
    fake_labels = Variable(torch.LongTensor(np.random.randint(0, 10, batch_size))).cuda()
    fake_images = generator(z, fake_labels)
    validity = discriminator(fake_images, fake_labels)
    g_loss = criterion(validity, Variable(torch.ones(batch_size)).cuda())
    g_loss.backward()
    g_optimizer.step()
    return g_loss.item()
def discriminator_train_step(batch_size, discriminator, generator, d_optimizer, criterion, real_images, labels):
    d_optimizer.zero_grad()

    # train with real images
    real_validity = discriminator(real_images, labels)
    real_loss = criterion(real_validity, Variable(torch.ones(batch_size)).cuda())
    
    # train with fake images
    z = Variable(torch.randn(batch_size, 100)).cuda()
    fake_labels = Variable(torch.LongTensor(np.random.randint(0, 10, batch_size))).cuda()
    fake_images = generator(z, fake_labels)
    fake_validity = discriminator(fake_images, fake_labels)
    fake_loss = criterion(fake_validity, Variable(torch.zeros(batch_size)).cuda())
    
    d_loss = real_loss + fake_loss
    d_loss.backward()
    d_optimizer.step()
    return d_loss.item()
num_epochs = 30
n_critic = 5
display_step = 50
for epoch in range(num_epochs):
    print('Starting epoch {}...'.format(epoch), end=' ')
    for i, (images, labels) in enumerate(data_loader):
        
        step = epoch * len(data_loader) + i + 1
        real_images = Variable(images).cuda()
        labels = Variable(labels).cuda()
        generator.train()
        
        d_loss = discriminator_train_step(len(real_images), discriminator,
                                          generator, d_optimizer, criterion,
                                          real_images, labels)
        

        g_loss = generator_train_step(batch_size, discriminator, generator, g_optimizer, criterion)
        
        writer.add_scalars('scalars', {'g_loss': g_loss, 'd_loss': d_loss}, step)  
        
        if step % display_step == 0:
            generator.eval()
            z = Variable(torch.randn(9, 100)).cuda()
            labels = Variable(torch.LongTensor(np.arange(9))).cuda()
            sample_images = generator(z, labels).unsqueeze(1)
            grid = make_grid(sample_images, nrow=3, normalize=True)
            writer.add_image('sample_image', grid, step)
    print('Done!')

Could you add print statements in your forward methods and print the shape of the used tensors?
Some of them are creating a shape mismatch.
Usually you get this error, if you pass wrong input tensor to e.g. a nn.Conv2d layer:

conv = nn.Conv2d(3, 6, 3, 1, 1)
x = torch.randn(100)
out = conv(x)
> RuntimeError: Expected 4-dimensional input for 4-dimensional weight 6 3 3 3, but got 1-dimensional input of size [100] instead

This layer expects a 4-dimensional input in [batch_size, channels, height, width].

If you have trouble isolating the issue, please post the complete stack trace.