RU translation: /tutorials/keras/

[0101011]

1. For convenience I've created a separate site/ru folder (site/ru/tutorials/keras/) for all translations to be committed there

2. Feel free to change the structure as needed

[googlebot]

Thanks for your pull request. It looks like this may be your first contribution to a Google open source project (if not, look below for help). Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).

📝 Please visit https://cla.developers.google.com/ to sign.

Once you've signed (or fixed any issues), please reply here (e.g. I signed it!) and we'll verify it.

---

What to do if you already signed the CLA

Individual signers

• It's possible we don't have your GitHub username or you're using a different email address on your commit. Check your existing CLA data and verify that your email is set on your git commits.

Corporate signers

• Your company has a Point of Contact who decides which employees are authorized to participate. Ask your POC to be added to the group of authorized contributors. If you don't know who your Point of Contact is, direct the Google project maintainer to go/cla#troubleshoot (Public version).
• The email used to register you as an authorized contributor must be the email used for the Git commit. Check your existing CLA data and verify that your email is set on your git commits.
• The email used to register you as an authorized contributor must also be attached to your GitHub account.

[0101011]

Signed the CLA

[googlebot]

CLAs look good, thanks!

[lamberta]

Thanks, Andrew! That was quicker than I anticipated 😀
I'll need to have some process dicussions as well as find a reviewer, but this is a great start. I'm out of the office until next week but will follow up when I get back.

Just curious, how difficult was the translation process? I take it you were already familiar with Jupyter notebooks and TensorFlow? As we build out a community process, is this something you're interested in contributing to?

[0101011]

Hi Billy,

1. I am not an X-Man, but quite familiar with TensorFlow and Jupyter and eager to contribute more:
   Is there a path I should follow for translations (tutorial folder, what next?)

2. Translation process is quite interesting tbh:
   I've got the image of 16-20 years-old student reading this tutorial and tried to use the best suited language to make it understandable.

• But sure do find a reviewer: I would appreciate some feedback! :)

[ezhulenev]

@lamberta can we share a "style guide" we used for other translations? @0101011 we basically have a small guide with "common terminology" that should not be translated, my guess that 90% of it will be directly applicable to russian translation

[0101011]

@lamberta @ezhulenev Style guides would be handy:

1. Yesterday I had a random thought of adding bite-sized "cheat sheets" to docs and colabs for readers to catch up quickly

2. Btw, cheat sheets are all the rage now :)

[lamberta]

Thanks for your work on this guys, it's been a pretty educational experience :)

After talking with Eugene, it does seem like there are technical words in English that do not have a natural translation, and we suspect many of these words are common across many target languages (based on feedback from some Chinese reviewers, as well). We don't have a master cheat sheet of "do not translate" words, but I just created a wiki page so maybe we can start adding to that: https://github.com/tensorflow/docs/wiki/Translation-notes

Andrew, for the cheat sheets, do you mean like a glossary of translations for ML terms? That could be interesting. The Machine Learning Crash Course maintains an ML glossary, and even some translations, but it doesn't look like Russian is there.

[0101011]

@lamberta

1. The wiki guide for translation is great: we'll add the most arguable terms over the course of the next week

2. Downside: I can't PR to wiki of TF/docs, so we will drop you a list here as it's done

3. We can implement cheat sheets like this into docs and colabs as little snippets or blocks (think of college textbooks) with short explanations of newly meet words (can be done with vanilla markdown):

My guess it'd be quite nifty for Korean, Chinese, Russian and other major target languages (btw, I saw an Arabic ML community on Twitter, ages 19 - 27: we can reach out to them for translations too)

4. Alternative is a separate glossary as ipynb doc, but I'll come back with a more practical example later

5. Here's a cool wiki page to get a sense of the approximate number of words in languages and why sometimes it is a problem :)

[lamberta]

Hi Andrew,
We recently updated the Keras regression tutorial. The format is mostly the same but the content has changed to make the problem a little more straightforward. Would you mind updating your translation to reflect the current tensorflow.org? This section should now be relatively stable.

Yes, feel free to add your list here and I can update that wiki page. I'm not sure if we're going to keep that, but it should do for now. I'd prefer we standardize around the MLCC glossary but it doesn't look like there's a Russian version :(

Hold off on the cheatsheets, it will be easier to maintain these translations if they follow the en/ versions.

Thanks!

[0101011]

@lamberta agree, maintaining should be stable

Got some fixes coming in 10 hours, I’ll be back 👍

[lamberta]

This is great work, Andrew, thanks so much!
I'm going to merge this here, but we still have some infrastructure work to do before we can pull the notebook in, render it, then display on https://tensorflow.org/?hl=ru
I can ping this bug to let you know when we get that working, probably after the winter break.

I've created a docs@tensorflow.org list to start to organize some of the community a little. Still early days but hopefully this will be a better way for contributors to communicate instead of just PR comments ;)

Thanks!

[0101011]

@lamberta Hi Billy,

1. Sure, no rush: I’m planning to do some final proofreading tomorrow just in case

2. I’ll subscribe to the mailing list in a minute to see how we can make use of it (discuss and developers lists are buzzing, so I guess that’ll work :)

Have a great weekend!

[lamberta]

Sounds good. Feel free to create another pull request if you have more edits. Thanks!