Project Organization

├── README.md               <- The top-level README for developers using this project.
├── data
│   ├── train.csv           <- Training dataset provided by Kaggle.
│   ├── test.csv            <- Test dataset provided by Kaggle.
│   └── ground-truth.csv    <- Ground truth dataset which contains actual label for both training and test data.
│
├── img                     <- Images inserted to notebook
│
├── 1.0-wsh-titanic-exploratory-data-analysis.ipynb
│                           <- Jupyter notebooks for exploratory study and feature engineering
│ 
├── 2.0-wsh-titanic-modelling-and-evaluation.ipynb
│                           <- Jupyter notebooks for model training and parameter tuning
│
├── requirements.txt        <- The requirements file for reproducing the analysis environment, e.g.
│                              generated with `pip freeze > requirements.txt`
│
├── helps.py                <- Helper script for data preprocessing and parameter tuning
│ 
└── submission.csv          <- Prediction of labels for test data submitted to Kaggle.

Install and start Jupyter Lab

$ sudo -H pip3 uninstall -y jupyterlab && sudo -H pip3 install jupyterlab
$ jupyter-lab

Follow the guidelines on https://plot.ly/python/getting-started/ to install extensions for the use of plotly with Jupyter Lab.

Install lightgbm on macOS system

$ brew install cmake
$ brew install gcc@7
$ git clone --recursive https://github.com/Microsoft/LightGBM ; cd LightGBM
$ mkdir build ; cd build
$ cmake -DCMAKE_CXX_COMPILER=g++-7 -DCMAKE_C_COMPILER=gcc-7 ..
$ make -j
$ cp -r LightGBM/python-package/lightgbm python3.7/site-packages/
$ cp LightGBM/lib_lightgbm.so python3.7/site-packages/lightgbm/

Recommend Projects