ketrint / Code4ML

Code4ML code blocks extractions

This is a source code for CODE4ML dataset construction.

Prerequisites

In order to load data form Kaggle one should have a kaggle.json file with the username and key specified.

An example of collecting code snippets from Kaggle can be found [here].

Overview

This is an official repository for code snippets from Kaggle kernels collecting.

You can find the instructions above.

Kaggle kernels links collection

1. Collection of kernels links to kernel_lists directory.

mkdir kernel_lists

python collect_kernels_from_competitions.py

The script does the following:

• Collects the links to the Kaggle competitions to .csv table
• Runs competition_kernels.sh, which collect kernels information of every competition
• Collects .csv files with the Kaggle kernels links to kernel_lists directory

Output: Kaggle kernels links .csv files

2. Combining kernels links tables into one .csv table

python unite_kernel_lists.py

Input:

Output: .csv table with the links to the Kaggle kernels

Kaggle kernels parsing

python code_blocks_extraction.py

Input: .csv table with the links to the Kaggle kernels

Output: .csv table with the following columns: "kernel_id", "code_block", "code_block_id".