apache / apache/dolphinscheduler-sdk-python

[impv] Auto create the whole task workspace to make the task the same workspace when running locally and on the DolphinScheduler platform

Open
#59 0 comments 1 reaction 0 assignees View on GitHub
Dominant language
Python
Stars
71
Forks
26
PR merge metrics
No merged PRs in 30d

Description

For example, If I have the workspace like this: https://github.com/pytorch/examples/tree/main/word_language_model
image

We can submit the task
```python
# [start workflow]
from pydolphinscheduler.core import Workflow
from pydolphinscheduler.core.resource import Resource, LocalResource
from pydolphinscheduler.tasks import Shell

with Workflow(
name="auto_resources_example",
) as workflow:
# [start use_exists_resources]
task_use_resource = Shell(
name="example",
command=f"python main.py --cuda",
auto_resource=True
)
# [end use_exists_resources]

workflow.run()
# [end workflow]
```

Then, all the files will be uploaded to the resource center.
And there is another best way to track the file version, If this is the git repository, we only record the git repository, version, and diff.
Then, we can reproduce the workspace in dolphinscheduler. (Inspired by ClearML)
So, After developing a project locally, we can directly submit it to DolphinScheduler and run it directly. Except for some inconsistent environment variables.

Contributor guide

Open the contributing guide

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.