https://github.com/gclouduniverse/notebook_training

Execute Jupyter notebooks in Google Cloud AI training jobs
https://github.com/gclouduniverse/notebook_training

Last synced: 5 days ago
JSON representation

Execute Jupyter notebooks in Google Cloud AI training jobs

• Host: GitHub
• URL: https://github.com/gclouduniverse/notebook_training
• Owner: gclouduniverse
• License: mit
• Created: 2020-03-10T20:34:58.000Z (almost 5 years ago)
• Default Branch: master
• Last Pushed: 2022-12-08T03:48:03.000Z (about 2 years ago)
• Last Synced: 2024-12-02T21:47:28.844Z (about 1 month ago)
• Language: Python
• Size: 32.2 KB
• Stars: 7
• Watchers: 5
• Forks: 2
• Open Issues: 10
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# Cloud AI Notebook Training

This package allows to run a Jupyter notebook at Google Cloud AI Platform Training Jobs.

Install tool with pip by using `pip install gcloud-notebook-training`

## Syntax

```

gcloud-notebook-training [-h] --input-notebook INPUT_NOTEBOOK

              [--project-id PROJECT_ID]

              [--output-notebook OUTPUT_NOTEBOOK]

              [--job-id JOB_ID]

              [--region REGION]

              [--worker-machine-type WORKER_MACHINE_TYPE]

              [--bucket-name BUCKET_NAME]

              [--max-running-time MAX_RUNNING_TIME]

              [--container-uri CONTAINER_URI]

              [--accelerator-type ACCELERATOR_TYPE]

              [--service-account SERVICE_ACCOUNT]

```

The only required parameter is --input-notebook.

This parameter accepts local path or GCS path.

For example:

```

gcloud-notebook-training --input-notebook /local/path/to/notebook.ipynb

```

or

```

gcloud-notebook-training --input-notebook gs://bucket_name/notebook.ipynb

```

The `output-notebook` path can be specified explicitly.

Same as the `input-notebook`, this parameter accepts local path or GCS path.

If the `output-notebook` is not specified, it will be implied as `input-notebook` plus 'output' suffix.

If `output-notebook` is a GCS path, then the tool will only submit the training job and exit.

If `output-notebook` is a local path, then the tool will wait until the training job succeeds, and then will download the output notebook.

`container-uri` parameter specifies the container used by training job.

If this parameter is not specified, the tool will try to pull this information from the notebook metadata.