Clutter Image Rating - pytorch

Pytorch implementation of Automatic Assessment of Hoarding Clutter from Images Using Convolutional Neural Networks.

1)Training

1. Required libraries are listed in requirements.txt. Run pip install -r requirements.txt using python 3.6 to install them.
2. Follow the steps in train.ipynb

2) Inference

1. Required libraries are listed in requirements_inf.txt. Run pip install -r requirements_inf.txt using python 3.6 to install them.
2. Change model_path and image_path in inference.py with the desired paths.
3. Run python inference.py to see the estimated CIR score of the image