This repository contains a collection of deep learning projects I have worked on to practice building and training neural network models for different applications.
The repository contains the following projects:
-
Cat vs Dog Image Classifier: A convolutional neural network to classify images of cats and dogs. Implements custom model and transfer learning with VGG16.
-
MobileNet Image Classification: Uses a compact MobileNet model to classify images of common objects.
-
Sign Language Digit Recognition: Classifies images of hand signs representing digits 0-9 using a CNN.
- Artistic Style Transfer: Blends the style of one image with the content of another using a pre-trained deep learning model.
- Medical Trial Analysis: Analyzes simulated medical trial data using neural networks to predict side effects based on patient age.
The goal of these projects is to get hands-on practice with deep learning concepts like:
- Building and training convolutional neural networks
- Using transfer learning
- Processing image data and creating data pipelines
- Evaluating model performance
Completing these fundamental projects helped me gain experience with different network architectures and strategies for tackling image-related tasks.
The projects are self-contained in their own directories with a dedicated README explaining the problem, data, model architecture and training process.
Jupyter notebooks are included documenting the model implementation and training code. Pre-trained models and sample images are provided to easily test out the models.