Computer Vision & Deep Learning

Overview

A short Colab/Jupyter notebook from a computer vision and deep learning course that begins defining a convolutional neural network in PyTorch. It sets up the imports and the opening layers of a CNN class intended for image input, representing the starting scaffold of a classifier rather than a finished model.

Key Achievements

• Defines a CNN class subclassing torch.nn.Module in PyTorch
• Specifies convolutional layers (nn.Conv2d) starting from 3-channel image input and 2x2 max-pooling layers (nn.MaxPool2d)
• Imports the core PyTorch deep-learning stack (torch, torch.nn, torch.nn.functional)

Approach

The notebook uses PyTorch's object-oriented module API, subclassing nn.Module and declaring layers (Conv2d, MaxPool2d) in the constructor. It is written in a Google Colab environment as the initial architecture-definition step of a convolutional network for image data.

Tools & Technologies

Results

No results are documented: the notebook contains no forward pass, training loop, dataset, or cell outputs. It is an incomplete, early-stage architecture scaffold rather than a trained or evaluated model.