entbappy/End-to-End-Chest-Cancer-Classification-using-MLflow-DVC — reverse-engineered prompt

Build me an end to end chest cancer image classification project that can train a model, track experiments, and serve predictions through a simple web app.

I want the user to be able to upload a chest scan image and get a clear prediction back. Please set up the training pipeline so it can be rerun cleanly, with config files for parameters, model settings, and paths. Use experiment tracking so training runs, metrics, and model versions are logged, and use a lightweight data pipeline tool so the steps can be reproduced.

Include a basic Flask style app with an upload page, prediction result, and a trained model loading flow. Also add the pieces needed to run it locally with Docker and make it ready for deployment through GitHub Actions to AWS using ECR and EC2.

Keep it simple and practical, with clear commands in the README for training, reproducing the pipeline, launching tracking, running the app, and deploying.