frederisk/kaggle Action

Use kaggle to implement simple CI, which supports GPU.

Trust Signals

name	description	required	default
api_token	kaggle api token	—	—
username	kaggle username	—	—
key	kaggle key	—	—
id	The URL slug of your kernel	—	—
title	The title of the kernel	yes	—
code_file	The path to your kernel source code	yes	—
language	The language your kernel is written in	—	python
kernel_type	The type of kernel	—	script
is_private	Whether or not the kernel should be private	—	true
enable_gpu	Whether or not the kernel should run on a GPU	—	true
enable_tpu	Whether or not the kernel should run on a TPU	—	false
enable_internet	Whether or not the kernel should be able to access the internet	—	true
dataset_sources	A list of dataset sources	—	—
competition_sources	A list of competition sources	—	—
kernel_sources	A list of kernel sources	—	—
keywords	A list of keywords to apply to the kernel	—	—
docker_image	Which docker image to run with	—	—
docker_image_pinning_type	Which docker image to use for executing new versions going forward	—	—
machine_shape	The type of accelerator being used with the kernel	—	—
timeout_seconds	Limit the run time of a kernel to the given number of seconds. The global maximum time will not be exceeded	—	—
fetch_time_seconds	The time to wait for the kernel to finish fetching data	—	3

no outputs