gabrielhamel / pneumonia-detection

Little CNN network who take chest xray and found if the patient have pneumonia and which kind

Zoidberg 2.0

Installation

0. Clone the project and go into

   git clone [email protected]:EpitechMscPro2020/zoidberg2.0_2020_35.git
   cd zoidberg2.0_2020_35

1. Download the dataset and place chest_Xray folder at the repository root.

2. Create python environment. You must have venv or conda on your machine

   # venv
   python3 -m venv env
   # or with conda
   conda create -n env python=3

3. Install dependencies

   source env/bin/activate
   pip install --upgrade pip
   pip install -r requirements.txt

4. Launch jupyter web browser

   jupyter-notebook

Notebooks

1. Dataset
   • Prepare data before training, it is mandatory before execution of another notebook
   • It will create a new file dataset.hdf5
2. CNN
   • Train AI on previous created dataset with CNN model

Binary

We made a GUI to use our already trained model in ./model/cnn. You just need to upload image with upload button. The GUI will show the classification label and her prediction accuracy

Doesn't need notebooks execution or any dataset

You must have enabled the virtual env (goto installation 2.)

python prgm/ui/main.py