Simple Neural Networks gives zero loss after 1 epoch

I am trying to build a simple NN to classify my data into 5 classes labeled as [1,2,3,4,5]. The inpput data (provided as X_t [for training] amd X_v[for test] is read from am instrument that with 15 pixels (X features = 15). I don’t get any error when running the code but I am suspicious something is not right because the loss converges to zero after 1 epoch and the accuracy for test set becomes 100%. Here is my code:

import numpy as np
import torch
import torch.nn as nn
import torch.optim as optim
from import Dataset, TensorDataset, DataLoader
from torch.autograd import Variable
import torch.nn.functional as F

X_trn= torch.from_numpy(X_t).float()
y_trn = torch.from_numpy(y_t).float()

train_dataset = TensorDataset(X_trn, y_trn)
train_loader = DataLoader(train_dataset, batch_size=12, shuffle=True)

X_test = torch.from_numpy(X_v).float()
y_test = torch.from_numpy(y_v).float()

test_dataset = TensorDataset(X_test, y_test)
test_loader = DataLoader(test_dataset, batch_size=12)

class Net(nn.Module):
    def __init__(self):
        super(Net, self).__init__()
        self.fc1 = nn.Linear(15, 8)
        self.fc2 = nn.Linear(8, 5)
    def forward(self,x):
        x = F.relu(self.fc1(x))
        x = self.fc2(x)
        return F.log_softmax(x, dim=1)

optimzier = optim.SGD(net.parameters(), lr=0.001, momentum=0.9)
criterion = nn.NLLLoss()

epochs = 10
for epoch in range(epochs):
    for batch_idx, (data, target) in enumerate(train_loader):
        data, target = Variable(data), Variable(target)
        data = data.view(-1, 15)
        net_out = net(data)
        loss = criterion(net_out, torch.max(target, 1)[1])


val_loss = 0
correct = 0
with torch.no_grad():
    for data, target in test_loader:
        data, target = Variable(data), Variable(target)

        net_out = net(data)
        val_loss += criterion(net_out, torch.max(target,1)[1])
        pred =[1]

        correct += pred.eq(torch.max(target,1)[1]).sum()