Rahul Rajesh, Zhuang Xinjie, Chitrangna Bhatt
https://higgsml.lal.in2p3.fr/files/2014/04/documentation_v1.8.pdf
Estimate the likelihood that a given event's signature was the result of a Higgs boson.
-
linear_models
(directory) : this folder contains all the implementation details for the models used. It is written in an object-oriented fashion for ease of use.linear_reg_gd.py
: implementation of linear regression using gradient descentlinear_reg_lsq.py
: implementation of linear regression using normal equationlogistic_reg.py
: implementation of logistic regression using gradient equationridge_reg_lsq.py
: implementation of ridge regression using normal equation
-
preprocess
(directory) : this folder contains all the implementation details for the various preprocessing teachniques used.imputer.py
: class that handles imputation for missing or undefined valuesscaler.py
: class that handles normalization of feature matrix
-
implementations.py
: method implementations required for the project. Note this file relises on thelinear_models
directory -
proj1_helpers.py
: various helper methods for the project -
project1.ipynb
: jupyter notebook showcasing the various steps carried out to solve this problem -
run.py
: generates csv file for test set - used for submission to platform -
report.pdf
: final report for project
Requirements: Python3, Numpy, Matplotlib
- Specify input path for data-files in
run.py
- Specify output path for prediction in
run.py
- In your terminal, run
python run.py