My code for fixed-effect Zero-Inflated Poisson Regression model.
This code was writted as part of my work on the paper Prediction of Sparse User-Item Consumption Rates with Zero-Inflated Poisson Regression. Please cite the paper if you're using any parts of the code.
I've done a lot of work to make it super efficient and I believe there are more comments in the code than actual code, but I'm sure more could have been done.
To make it easier to work with the code, I added to IPython notebooks to the repository.
- The first one creates a synthetic data that I've used for uni-testing, but you can use it to understand the formatting of the data.
- The second one is a complete run example of the model using the synthetic data showing both sanity uni-tests and empirical evaluations.
- numpy and scipy
- pandas
- cython
- You also need to run the setup.py. Type in the cmd line 'python setup.py build_ext --inplace' in the project directory.