README.md

Push Kaggle Kernel

A GitHub Action to push Kaggle kernels.

Usage

1. Add KAGGLE_USERNAME and KAGGLE_KEY as secrets in your repository.
2. Define your workflow.

Example

name: Upload

on:
  push:
    branches:
      - master

jobs:
  upload:
    name: Upload
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v2

      - name: Create a toy python script
        run: |
          echo "print('hello world')" > script.py

      - uses: harupy/push-kaggle-kernel@master
        env:
          # Do not leak your credentials.
          KAGGLE_USERNAME: ${{ secrets.KAGGLE_USERNAME }}
          KAGGLE_KEY: ${{ secrets.KAGGLE_KEY }}
        with:
          id: ${{ secrets.KAGGLE_USERNAME }}/${{ github.sha }}
          title: ${{ github.sha }}
          code_file: ./script.py
          language: python
          kernel_type: script
          # Do not share high-scoring kernels.
          is_private: false
          competition_sources: |
            titanic

Inputs

Name	Description	Required	Default	Options
id	ID of kernel (must have the format: {username}/{slug}).	true	-
title	Title of kernel (must be at least five characters).	true	-
code_file	Path to kernel to push (relative from the repo root).	true	-
language	Language that kernel is written in.	true	-	["python", "r", "rmarkdown"]
kernel_type	Type of kernel.	true	-	["script", "notebook"]
is_private	Whether or not kernel should be private.	false	"true"	["true", "false"]
enable_gpu	Whether or not kernel should run on a GPU.	false	"false"	["true", "false"]
enable_internet	Whether or not kernel should be able to access the internet.	false	"false"	["true", "false"]
dataset_sources	A list of data sources that kernel should use.	false	""
competition_sources	A list of competition data sources that kernel should use.	false	""
kernel_sources	A list of kernel data sources that kernel should use.	false	""