Discriminator of GAN performing poorly for network anomaly detection

Intermediate
1

I took only positive samples and trained the generator and discriminator. The discriminator predicts anomlay or normal with an accuracy of 52 percent. I would appreciate some help to improve performance: This is the code for it:

train_data_length = df.shape[0]
train_labels = torch.zeros(train_data_length)
train_data = torch.tensor(df.values)
train_set = [
    (train_data[i], train_labels[i]) for i in range(train_data_length)
]

batch_size = 40
train_loader = torch.utils.data.DataLoader(
    train_set, batch_size=batch_size, shuffle=True
)

class Discriminator(nn.Module):
    def __init__(self):
        super().__init__()
        self.model = nn.Sequential(
            nn.Linear(118, 512),
            nn.ReLU(),
            nn.Dropout(0.3),
            nn.Linear(512, 256),
            nn.ReLU(),
            nn.Dropout(0.3),
            nn.Linear(256, 128),
            nn.ReLU(),
            nn.Dropout(0.3),
            nn.Linear(128, 64),
            nn.ReLU(),
            nn.Dropout(0.3),
            nn.Linear(64, 1),
            nn.Sigmoid(),
        )

    def forward(self, x):
        output = self.model(x)
        return output

discriminator = Discriminator()

class Generator(nn.Module):
    def __init__(self):
        super().__init__()
        self.model = nn.Sequential(
            nn.Linear(118, 256),
            nn.ReLU(),
            nn.Linear(256, 512),
            nn.ReLU(),
            nn.Linear(512, 118),
        )

    def forward(self, x):
        x = x.view(x.size(0), 118)
        output = self.model(x)
        return output

generator = Generator()

lr = 0.0001
num_epochs = 100
loss_function = nn.BCELoss()

optimizer_discriminator = torch.optim.Adam(discriminator.parameters(), lr=lr)
optimizer_generator = torch.optim.Adam(generator.parameters(), lr=lr)

for epoch in range(num_epochs):
    for n, (real_samples, _) in enumerate(train_loader):
        # Data for training the discriminator
        real_samples_labels = torch.ones((batch_size, 1))
        latent_space_samples = torch.randn((batch_size, 118))
        generated_samples = generator(latent_space_samples)
        generated_samples_labels = torch.zeros((batch_size, 1))
        all_samples = torch.cat((real_samples, generated_samples))
        all_samples_labels = torch.cat(
            (real_samples_labels, generated_samples_labels)
        )

        # Training the discriminator
        discriminator.zero_grad()
        all_samples = all_samples.float()
        output_discriminator = discriminator(all_samples)
        loss_discriminator = loss_function(
            output_discriminator, all_samples_labels)
        loss_discriminator.backward()
        optimizer_discriminator.step()

        # Data for training the generator
        latent_space_samples = torch.randn((batch_size, 118))

        # Training the generator
        generator.zero_grad()
        generated_samples = generator(latent_space_samples)
        output_discriminator_generated = discriminator(generated_samples)
        loss_generator = loss_function(
            output_discriminator_generated, real_samples_labels
        )
        loss_generator.backward()
        optimizer_generator.step()

        # Show loss
        if epoch % 10 == 0 and n == batch_size - 1:
            print(f"Epoch: {epoch} Loss D.: {loss_discriminator}")
            print(f"Epoch: {epoch} Loss G.: {loss_generator}")

pred = discriminator(test_data)
y_pred = pred.detach().numpy()

from sklearn.metrics import accuracy_score
threshold = 0.5
predicted_labels = (y_pred >= threshold).astype(int)

# Calculate accuracy
accuracy = accuracy_score(y, predicted_labels)

print(f'Accuracy: {accuracy}')

Thank you for your help!