Simple Logistic Regression using Matrix Factorization in PyTorch
This is a simple Python project that implements logistic regression using matrix factorization in PyTorch. It can be used for collaborative filtering tasks such as recommendation systems.
- Python 3.6 or higher
- PyTorch 1.8 or higher
- Numpy
- Pandas
- Clone this repository to your local machine.
logistic_mf.py
contains the core logic to train the model on a toy dataset (tiny_data).- You can also use
logistic_mf_helper.ipynb
ortest_logistic_mf.ipynb
to interact with the model in a Jupyter notebook. - You can modify the hyperparameters and the dataset in the
logistic_mf.py
file.