conda env create -f environment.yml
The required models can be downloaded from the following links:
This repository contains the code for the following:
- Loading the .pt model using PyTorch
- Converting the model to ONNX format using PyTorch
- Model inference using ONNX model
- Web app for model inference using ONNX model
The web app can be found in app/
directory.
This is a sample flask application that uses Bootstrap for the Frontend UI and PyTorch for model inference.
conda activate pytorch_inference
python app_chest_xray.py
Go to 127.0.0.1:4555 on your browser.
The test images can be found in chest_xray_kaggle
folder.
The test images have been obtained from Kaggle Chest X-Ray Competition.
You should get a screen shown below.
Choose Browse and upload an image. Click Upload to get the output