neerajsharma9195 / count-sketch-feature-selection

count-sketch-feature-selection

Project Structure:

1. src/independent_study: contains code related to independent study experiments

   • dumps: folder contains results dump
   • logistic_regression: code for logistic regression (Standard Datasets, Synthetic Dataset)
   • dataset_generation: code to generate synthetic dataset
   • image_df.ipynb: jupyter notebook contains code to generate graphs for all 5 models and creates html page

2. src/sketches: contains code for all variations of count sketches and top-K heap

3. src/processing: code to preprocess data and converting in required format

4. src/classifiers: code from prev semester ML project

5. src/results: results from prev semester ML project