Scribd
Upload
3 views4 pages

Appendix Modules Report

The document outlines a series of modules for a machine learning project, focusing on data preprocessing, DataLoader creation, CNN model architecture, training loop, and evaluation. Each module includes a description of its purpose and relevant code snippets for implementation. The overall aim is to prepare input images, efficiently load data, define a CNN for classification, train the model, and evaluate its performance.

Uploaded by

siddharthamaurya63
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
3 views4 pages

Appendix Modules Report

The document outlines a series of modules for a machine learning project, focusing on data preprocessing, DataLoader creation, CNN model architecture, training loop, and evaluation. Each module includes a description of its purpose and relevant code snippets for implementation. The overall aim is to prepare input images, efficiently load data, define a CNN for classification, train the model, and evaluate its performance.

Uploaded by

siddharthamaurya63
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd

Appendix B: Project Snapshots and

Description
B.2 Module 2: Data Preprocessing and Transformation
Description:
This module prepares input images before training. It includes resizing, normalization, and
augmentation.
These steps improve model generalization and robustness.

Code:
from torchvision import transforms

img_size = 224

train_tf = [Link]([
[Link](),
[Link]((img_size, img_size)),
[Link](),
[Link](10),
[Link](),
[Link](mean=[0.5,0.5,0.5], std=[0.5,0.5,0.5])
])

test_tf = [Link]([
[Link](),
[Link]((img_size, img_size)),
[Link](),
[Link](mean=[0.5,0.5,0.5], std=[0.5,0.5,0.5])
])

B.3 Module 3: DataLoader Creation


Description:
This module loads data efficiently using batching and shuffling for training.

Code:
from [Link] import DataLoader

train_ds = ImgDataset(train_r_path, train_f_path, train_tf)


test_ds = ImgDataset(test_r_path, test_f_path, test_tf)

train_loader = DataLoader(train_ds, batch_size=32, shuffle=True)


test_loader = DataLoader(test_ds, batch_size=32, shuffle=False)

B.4 Module 4: Model Architecture (CNN Model)


Description:
Defines CNN architecture for feature extraction and classification.

Code:
import torch
import [Link] as nn
import [Link] as F

class CNNModel([Link]):
def __init__(self):
super(CNNModel, self).__init__()

self.conv1 = nn.Conv2d(3, 32, 3, padding=1)


self.conv2 = nn.Conv2d(32, 64, 3, padding=1)
self.conv3 = nn.Conv2d(64, 128, 3, padding=1)

[Link] = nn.MaxPool2d(2, 2)

self.fc1 = [Link](128 * 28 * 28, 128)


self.fc2 = [Link](128, 1)

def forward(self, x):


x = [Link]([Link](self.conv1(x)))
x = [Link]([Link](self.conv2(x)))
x = [Link]([Link](self.conv3(x)))

x = [Link]([Link](0), -1)
x = [Link](self.fc1(x))
x = [Link](self.fc2(x))

return x

B.5 Module 5: Training Loop


Description:
Handles model training using forward pass, loss computation, and backpropagation.

Code:
import [Link] as optim

model = CNNModel().to(device)
criterion = [Link]()
optimizer = [Link]([Link](), lr=0.0003)

epochs = 12

for epoch in range(epochs):


[Link]()
running_loss = 0

for imgs, labels in train_loader:


imgs, labels = [Link](device), [Link]().to(device)

optimizer.zero_grad()

outputs = model(imgs).squeeze()
loss = criterion(outputs, labels)

[Link]()
[Link]()

running_loss += [Link]()

print(f"Epoch {epoch+1}, Loss: {running_loss/len(train_loader):.4f}")

B.6 Module 6: Evaluation and Prediction


Description:
Evaluates model performance using confusion matrix and classification metrics.

Code:
from [Link] import confusion_matrix, classification_report

[Link]()
all_preds = []
all_labels = []

with torch.no_grad():
for imgs, labels in test_loader:
imgs = [Link](device)

outputs = model(imgs).squeeze()
preds = (outputs > 0.5).int().cpu()

all_preds.extend([Link]())
all_labels.extend([Link]())
print(confusion_matrix(all_labels, all_preds))
print(classification_report(all_labels, all_preds))

You might also like