import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim
from torchvision import datasets, transforms
from torch.optim.lr_scheduler import ExponentialLR

# Get CPU or GPU device for training
device = "cuda" if torch.cuda.is_available() else "cpu"
device = torch.device(device)

# Random seed for reproducibility
seed = 42
torch.manual_seed(seed)

# Save the model at the end?
save_model = False

# Batch sizes for training and testing
batch_size = 64
test_batch_size = 14

# Training epochs
n_epochs = 10

# Learning rate
learning_rate = 1.0

# Decay rate for adjusting the learning rate
gamma = 0.7

# How many batches before logging training status
log_interval = 10

# Number of target classes in the MNIST data
num_classes = 10

train_kwargs = {'batch_size': batch_size}
test_kwargs = {'batch_size': test_batch_size}

# CUDA settings
if torch.cuda.is_available():
    cuda_kwargs = {'num_workers': 1,
                   'pin_memory': True,
                   'shuffle': True}
    train_kwargs.update(cuda_kwargs)
    test_kwargs.update(cuda_kwargs)

# The scaled mean and standard deviation of the MNIST dataset (precalculated)
data_mean = 0.1307
data_std = 0.3081

# Convert input images to tensors and normalize
transform=transforms.Compose([
    transforms.ToTensor(),
    transforms.Normalize((data_mean,), (data_std,))
    ])

# Get the MNIST data from torchvision
dataset1 = datasets.MNIST('../data', train=True, download=True,
                    transform=transform)
dataset2 = datasets.MNIST('../data', train=False,
                    transform=transform)

# Define the data loaders that will handle fetching of data
train_loader = torch.utils.data.DataLoader(dataset1,**train_kwargs)
test_loader = torch.utils.data.DataLoader(dataset2, **test_kwargs)

# Define the architecture of the neural network
class Net(nn.Module):
    def __init__(self):
        super(Net, self).__init__()
        self.conv1 = nn.Conv2d(1, 32, kernel_size=3, stride=1, padding='valid')
        self.conv2 = nn.Conv2d(32, 64, kernel_size=3, stride=1, padding='valid')
        self.dropout1 = nn.Dropout(0.25)
        self.dropout2 = nn.Dropout(0.5)
        self.fc1 = nn.Linear(9216, 128)
        self.fc2 = nn.Linear(128, num_classes)

    def forward(self, x):
        x = self.conv1(x)
        x = F.relu(x)
        x = self.conv2(x)
        x = F.relu(x)
        x = F.max_pool2d(x, 2)
        x = self.dropout1(x)
        x = torch.flatten(x, 1)
        x = self.fc1(x)
        x = F.relu(x)
        x = self.dropout2(x)
        x = self.fc2(x)
        output = F.softmax(x, dim=1)
        return output

def train(model, device, train_loader, optimizer, epoch, log_interval):
    model.train()
    for batch_idx, (data, target) in enumerate(train_loader):
        data, target = data.to(device), target.to(device)
        optimizer.zero_grad()
        output = model(data)
        loss = F.cross_entropy(output, target)
        loss.backward()
        optimizer.step()
        if batch_idx % log_interval == 0:
            print('Train Epoch: {} [{}/{} ({:.0f}%)]\tLoss: {:.6f}'.format(
                epoch, batch_idx * len(data), len(train_loader.dataset),
                100. * batch_idx / len(train_loader), loss.item()))

def test(model, device, test_loader):
    model.eval()
    test_loss = 0
    correct = 0
    with torch.no_grad():
        for data, target in test_loader:
            data, target = data.to(device), target.to(device)
            output = model(data)
            # sum up batch loss
            test_loss += F.nll_loss(output, target, reduction='sum').item()
            # get the index of the max log-probability
            pred = output.argmax(dim=1, keepdim=True)
            correct += pred.eq(target.view_as(pred)).sum().item()

    test_loss /= len(test_loader.dataset)

    print('\nTest set: Average loss: {:.4f}, Accuracy: {}/{} ({:.0f}%)\n'.format(
        test_loss, correct, len(test_loader.dataset),
        100. * correct / len(test_loader.dataset)))

# Send the model to the device (CPU or GPU)
model = Net().to(device)

# Define the optimizer to user for gradient descent
optimizer = optim.Adadelta(model.parameters(), lr=learning_rate)

# Shrinks the learning rate by gamma every step_size
scheduler = ExponentialLR(optimizer, gamma=gamma)

# Train the model
for epoch in range(1, n_epochs + 1):
    train(model, device, train_loader, optimizer, epoch, log_interval)
    test(model, device, test_loader)
    scheduler.step()

if save_model:
    torch.save(model.state_dict(), "mnist_cnn_pytorch.ckpt")

def visualize_and_predict(model, device, data_loader):
    model.eval()
    with torch.no_grad():
        # Extract the first batch of images and labels
        data, target = next(iter(data_loader))
        # Select the first image and label
        img, label = data[0], target[0]
        
        # Visualize the image
        plt.imshow(img.squeeze(), cmap='gray')
        plt.title(f'Actual Label: {label.item()}')
        plt.show()

        # Run inference
        img = img.to(device)
        output = model(img.unsqueeze(0))  # Add batch dimension
        pred = output.argmax(dim=1, keepdim=True)

        print(f'Predicted Label: {pred.item()}')

visualize_and_predict(model, device, test_loader)