projet_ift712 is the semester project for the course IFT712 at the science faculty of Université de Sherbrooke. IFT712 is a course given by the computer science department. The goal of this project was to develop a code that was able to use 6 different machine learning techniques to recognize pathologies from several chest X-ray images from the kaggle dataset Random Sample of NIH Chest X-ray Dataset. The dataset is composed of 5606 images with a resolution of 1024 x 1024 px. Images are stored in PNG format.
To download the source code on your desktop, launch git clone https://github.com/felixdumais/projet_ift712.git in a Linux terminal. Once everything is downloaded, all the source code should be in a folder called projet_ift712.
Once the project is downloaded, create a new virtual environment. In the terminal write virtualenv ~/.venv_nih_ift712. Activate your new virtual environment with source ~/.venv_nih_ift712/bin/activate. Make sure you have the newest version of pip with the command pip install --upgrade pip.
Once everything is up to date, install all the packages listed in the file requirements.txt in your new virtual environment with the command pip install -r requirements.txt.
Make sure that the kaggle API is correctly installed and that the file kaggle.json, containing your kaggle credentials, is saved in the right folder. To see how to install the kaggle API correctly, go to the section Usage of Kaggle API of the README.md. To see if the kaggle API is correctly installed, go in the section Running the tests of the README.md.
Make sure that your working directory is /projet_ift712/src.
You can now parse arguments in the terminal to launch the file main_ift712.py. To better understand the arguments, launch python main_ift712.py -h.
All the source code is in the directory src.
This project uses Kaggle API to download the dataset. Ensure you have Python 3 and the package manager pip installed.
If the API is not already installed, run the following command to access the Kaggle API using the command line:
pip install kaggle (You may need to do pip install --user kaggle on Mac/Linux. This is recommended if problems come up during the installation process.) Installations done through the root user (i.e. sudo pip install kaggle) will not work correctly unless you understand what you're doing. Even then, they still might not work. User installs are strongly recommended in the case of permissions errors.
If you run into a kaggle: command not found error, ensure that your python binaries are on your path. You can see where kaggle is installed by doing pip uninstall kaggle and seeing where the binary is. For a local user install on Linux, the default location is ~/.local/bin. On Windows, the default location is $PYTHON_HOME/Scripts.
To use the Kaggle API, sign up for a Kaggle account at https://www.kaggle.com. Then go to the 'Account' tab of your user profile (https://www.kaggle.com/<username>/account) and select 'Create API Token'. This will trigger the download of kaggle.json, a file containing your API credentials. Place this file in the location ~/.kaggle/kaggle.json (on Windows in the location C:\Users\<Windows-username>\.kaggle\kaggle.json - you can check the exact location, sans drive, with echo %HOMEPATH%). You can define a shell environment variable KAGGLE_CONFIG_DIR to change this location to $KAGGLE_CONFIG_DIR/kaggle.json (on Windows it will be %KAGGLE_CONFIG_DIR%\kaggle.json).
For your security, ensure that other users of your computer do not have read access to your credentials. On Unix-based systems you can do this with the following command:
chmod 600 ~/.kaggle/kaggle.json
To verify that the kaggle API, run the file test/kaggle_API_test.py in your terminal with the command python test/kaggle_API_test.py. If the API does not found the file kaggle.json this error appears OSError: Could not find kaggle.json. Make sure it's located in ~/.kaggle/kaggle.json. Or use the environment method. Go to the section Usage of Kaggle API to see where to install kaggle.json file.
Report that explains the project
To see the management page used for this project, login on Zenhub with your GitHub account.
- Félix Dumais (14053686)
- Nicolas Fontaine (15203272)
- Joelle Fréchette-Viens (15057894)