Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Completion of Chest X-Ray Images (Pneumonia) Detection Demo Using Ivy Framework #93

Open
wants to merge 2 commits into
base: main
Choose a base branch
from

Conversation

leebissessar5
Copy link

Changes Proposed in this Pull Request:

  • Pneumonia Detection with Ivy: This notebook demonstrates pneumonia detection in chest X-ray images using the Ivy machine learning framework.
  • Data Preprocessing with PyTorch: The preprocessing of images is handled by PyTorch's built-in transformation modules, illustrating Ivy's interoperability with other libraries.
  • Model Definition with Ivy.Module: The classifier is built using Ivy's Module class, providing a clear and structured way to define the model's architecture.
  • Training and Validation: These are executed with an informative progress bar to monitor the experiment's advancement.
  • Evaluation: A classification report and a confusion matrix are used to quantify the model's performance.
  • Pretrained Weights: Weights for a pretrained model (achieving ~80% test accuracy) are provided.

Impact of Change:

  • Ivy Demonstration: This notebook serves as a practical example of using Ivy for medical image classification, highlighting its capabilities and ease of use.
  • Flexibility: The integration with PyTorch modules underscores Ivy's versatility and its ability to work with familiar tools in the deep learning ecosystem.

How to Test:

  1. Setup: Follow the instructions in the notebook to install dependencies and set up the environment.
  2. Run: Execute the notebook cells sequentially to train the model, evaluate its performance, and visualize results.

Additional Notes:

  • Dataset: The dataset, sourced from Kaggle, exhibits a class imbalance skewed towards pneumonia cases. Recommendations for addressing this are provided within the notebook, including techniques like oversampling and more advanced architectures.
  • Model Size and Training Speed: A deliberately small model was chosen to:
    • Accommodate the computational resources of Colab's free tier.
    • Enable faster training times.
    • Allow for experimentation on less powerful GPUs.
    • Emphasize the primary goal of showcasing Ivy's functionality.

Closes #91

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Chest X-Ray Images (Pneumonia) Detection Demo(Intermediate)
2 participants