/kaggle-titanic-practice

PROJECT STATUS: IN-PROGRESS

Primary LanguageJupyter NotebookMIT LicenseMIT

kaggle-titanic-practice

Practicing some data science... Status: in-progress

Plan

  • Use titanic kaggle dataset
  • use PyCaret (autoMl, model validation)
  • SHAP to explain model output? SHAP
  • Submission in the kaggle platform
  • Docker and/or binder to facilitate deployment

Current score

Score: 0.77033 Leaderboard: 9090

Welcome! Here is some of the information and subject available in the analysis:

  • Jupyter folder
    • 01_EDA -> exploratory analysis with first assumptions
      • TO DOs and versioning
    • 02_Process -> jupyter used to format and prepare file to submit

To start playing with this analysis, there are two things you need to do:

IN-PROGRESS

Historic/Versioning

Version Score Leaderboard File
v1 0.76555 9557 submission-12-11-2022.csv
v2 0.77033 9090 submission-12-12-2022.csv

Here are some ideas to contribute if you want (and also my TO DO list for this project):

IN-PROGRESS