[
https://issues.apache.org/jira/browse/BEAM-6557?focusedWorklogId=208090&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-208090
]
ASF GitHub Bot logged work on BEAM-6557:
----------------------------------------
Author: ASF GitHub Bot
Created on: 05/Mar/19 21:34
Start Date: 05/Mar/19 21:34
Worklog Time Spent: 10m
Work Description: davidcavazos commented on issue #7679: [BEAM-6557] Adds
notebooks for quickstarts and custom I/O
URL: https://github.com/apache/beam/pull/7679#issuecomment-469865972
@aaltay answering some of the questions inline.
> * How can we keep these up to date and in sync with quick start?
For now, they are manually copy-pasted. In the future, there are two options:
1. Generate the markdown pages from the `.ipynb` files and let the docs be
autogenerated from the notebooks (this is how TensorFlow does it). This
simplifies things by having a single source of truth and more consistency.
2. Insert snippet region tags into the code blocks. Then write a tool to
convert the `.ipynb` files into `.py` files, so the docs can grab the snippets
from those `.py` files. This would allow for automated testing too, but the
region tags will "uglify" the notebooks.
3. A combination of both approaches. Generating doc pages from the `.ipynb`
*and* creating the `.py` for testing only.
> * I believe the location needs to change. Yes there are written in python
but they are not all python examples. Perhaps examples/notebooks/get-started
would be a better place.
Sure, I can move them there.
> * for python a python 3 version could be added.
Yes! This would be great. I just quickly changed the runtime to Python 3 on
the Python wordcount and it works out of the box without any changes. Should we
start moving towards Python 3 on all snippets (this isn't a production app
anyways) and only leave Python 2 whenever we find a missing feature or
something failing? I think this would be a good idea (I might do this on a
follow up Pull Request).
> * what is the usage model for colab? Could anyone try these out without
restrictions?
Colab is free and anyone can use it free of charge. There are no
restrictions unless you are using GPUs or TPUs under long running jobs. More
info [here](https://research.google.com/colaboratory/faq.html).
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
Issue Time Tracking
-------------------
Worklog Id: (was: 208090)
Time Spent: 2h 20m (was: 2h 10m)
> Add IPython notebooks for quickstarts and custom I/O
> ----------------------------------------------------
>
> Key: BEAM-6557
> URL: https://issues.apache.org/jira/browse/BEAM-6557
> Project: Beam
> Issue Type: Improvement
> Components: website
> Reporter: David Cavazos
> Assignee: David Cavazos
> Priority: Minor
> Labels: triaged
> Time Spent: 2h 20m
> Remaining Estimate: 0h
>
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
