Skip to content

In this project, we aim at solving the Higgs Boson classification.

Notifications You must be signed in to change notification settings

sMamooler/Higgs-Boson-Classification

Repository files navigation

Detecting the Higgs Boson by analyzing proton collisions

In this project, we aim at solving the Higgs Boson classification, a problem posed by the CERN.

Folders/files

In the this folder are the following files: |'run.py'| A main script containing our best method chosen to solve this classification problem : Ridge Regression In order to run this scrpit you need to have the test.csv and train.csv in this very same folder.

|'implementations.py'| A file containing our implementation of 6 machine learning regression and classification algorithms

  • least squares
  • least squares Gradient Descent
  • least squares Stochastic Gradient Descent
  • Ridge Regression
  • Logistic regression
  • Regularized logistic regression

|'proj1_helpers'| A file containing all other additional necessary functions used for

  • loading the data
  • preprocessing
  • splitting the data
  • batch iteration

|'cross_validation'| A file containing all required functions for cross validation

  • build_k_indices
  • k_fold_cross_validation
  • ridge_reg_cross_validation
  • reg_log_regression_cross_validation

Requirements to run the project

The following 'Python 3' packages are necessary for running our project : 'numpy'

Our results on AICrowd challenge

Team name : 'Outliers' Our team on is accessible with the following link : https://www.aicrowd.com/challenges/epfl-machine-learning-higgs/teams/Outliers Our best result is :

  • Categorical accuracy of 0.830
  • F1 score of 0.740

Authors

Chabenat Eugénie : eugenie.chabenat@epfl.ch Djambazovska Sara : sara.djambazovka@epfl.ch Mamooler Sepideh : sepideh.mamooler@epfl.ch

About

In this project, we aim at solving the Higgs Boson classification.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages