/jitpc

Primary LanguagePython

jitpc

Setup

Anaconda

To recreate the environment; run in a terminal:

Windows

conda env create -f environment-windows.yml

Any OS

conda env create -f environment.yml

or

conda create --name <env-name> python=3.8

conda activate <env-name>

conda install -c conda-forge imbalanced-learn=0.9.1

conda install pandas=1.2.5

Then copy and paste this project's [src, res] directories into your newly created anaconda environment.

Usage

New ML classification models can be trained, and existing models can be used to predict labels for new samples.
Pre-existing models are stored in the res/model directory, datasets are stored in the res/data directory, and classification reports are stored in the res/report directory.

Train

To train a new model; run in a terminal:

python src/train.py

To list all arguments:

python src/train.py -h

Classify

To classify samples using an existing model; run in a terminal:

python src/classify.py

To list all arguments:

python src/classify.py -h