Untitled

import torch
import torch.nn as nn
import torch.optim as optim
from torchvision import datasets, transforms, models
import os
import matplotlib.pyplot as plt
from torchvision.models import ResNet18_Weights
from tqdm import tqdm

data_transforms = {
    'train': transforms.Compose([
        transforms.Resize((416, 416)),
        transforms.ToTensor(),
        transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
    ]),
    'val': transforms.Compose([
        transforms.Resize((416, 416)),
        transforms.ToTensor(),
        transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
    ]),
}

data_dir = r"C:\Users\Emman\Desktop\corn\dataset\dataset"

image_datasets = {x: datasets.ImageFolder(os.path.join(data_dir, x), data_transforms[x]) for x in ['train', 'val']}
dataloaders = {x: torch.utils.data.DataLoader(image_datasets[x], batch_size=12, shuffle=True, num_workers=12) for x in ['train', 'val']}
dataset_sizes = {x: len(image_datasets[x]) for x in ['train', 'val']}
class_names = image_datasets['train'].classes
num_classes = len(class_names)


model = models.resnet18(weights=ResNet18_Weights.DEFAULT)

for name, param in model.named_parameters():
    if "fc" in name:
        param.requires_grad = True
    else:
        param.requires_grad = False

model.fc = nn.Sequential(
    nn.Linear(model.fc.in_features, 512),
    nn.ReLU(),
    nn.Linear(512, 3)
)

# Loss and optimizer
criterion = nn.CrossEntropyLoss()
optimizer = optim.SGD(model.parameters(), lr=0.001, momentum=0.9)

device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
model = model.to(device)

# Training function
def train_model(num_epochs=10):
    train_loss_history = []
    val_loss_history = []
    train_acc_history = []
    val_acc_history = []

    class_correct = {class_name: 0 for class_name in class_names}
    class_total = {class_name: 0 for class_name in class_names}

    for epoch in range(num_epochs):
        print(f'Epoch {epoch + 1}/{num_epochs}')

        for phase in ['train', 'val']:
            if phase == 'train':
                model.train()
            else:
                model.eval()

            running_loss = 0.0
            running_corrects = 0

            class_correct = {class_name: 0 for class_name in class_names}
            class_total = {class_name: 0 for class_name in class_names}

            with tqdm(total=dataset_sizes[phase], desc=f"{phase.capitalize()} Progress", unit="img") as pbar:
                for inputs, labels in dataloaders[phase]:
                    inputs, labels = inputs.to(device), labels.to(device)

                    optimizer.zero_grad()

                    with torch.set_grad_enabled(phase == 'train'):
                        outputs = model(inputs)
                        _, preds = torch.max(outputs, 1)
                        loss = criterion(outputs, labels)

                        if phase == 'train':
                            loss.backward()
                            optimizer.step()

                    running_loss += loss.item() * inputs.size(0)
                    running_corrects += torch.sum(preds == labels.data)

                    # Track per-class accuracy
                    for i in range(len(labels)):
                        label = labels[i]
                        pred = preds[i]
                        class_total[class_names[label]] += 1
                        if label == pred:
                            class_correct[class_names[label]] += 1

                    pbar.update(inputs.size(0))

            epoch_loss = running_loss / dataset_sizes[phase]
            epoch_acc = running_corrects.double() / dataset_sizes[phase]

            print(f'{phase} Loss: {epoch_loss:.4f} Acc: {epoch_acc:.4f}')

            if phase == 'train':
                train_loss_history.append(epoch_loss)
                train_acc_history.append(epoch_acc.item())
            else:
                val_loss_history.append(epoch_loss)
                val_acc_history.append(epoch_acc.item())

            # Print per-class accuracy after each phase
            print(f'Per-Class Accuracy for {phase}:')
            for class_name in class_names:
                acc = class_correct[class_name] / class_total[class_name] * 100 if class_total[class_name] > 0 else 0
                print(f'Class {class_name}: {acc:.2f}%')

    print("Training complete!")
    return train_loss_history, val_loss_history, train_acc_history, val_acc_history

def plot_loss_accuracy(train_loss, val_loss, train_acc, val_acc):
    plt.figure(figsize=(12, 6))

    plt.subplot(1, 2, 1)
    plt.plot(train_loss, label='Train Loss', color='blue', marker='o')
    plt.plot(val_loss, label='Val Loss', color='orange', marker='o')
    plt.xlabel('Epoch')
    plt.ylabel('Loss')
    plt.title('Loss over Epochs')
    plt.legend()
    plt.grid()

    plt.subplot(1, 2, 2)
    plt.plot(train_acc, label='Train Accuracy', color='green', marker='o')
    plt.plot(val_acc, label='Val Accuracy', color='red', marker='o')
    plt.xlabel('Epoch')
    plt.ylabel('Accuracy')
    plt.title('Accuracy over Epochs')
    plt.legend()
    plt.grid()

    plt.tight_layout()
    plt.show()

# Main
if __name__ == "__main__":
    train_loss, val_loss, train_acc, val_acc = train_model(num_epochs=10)

    # Save the model
    torch.save(model.state_dict(), 'corn_leaf_model.pth')
    print("Model saved to 'corn_leaf_model.pth'")

    plot_loss_accuracy(train_loss, val_loss, train_acc, val_acc)