9 Ways to Get Help with Deep Learning in Keras

Keras is a Python deep learning library that can use the efficient Theano or TensorFlow symbolic math libraries as a backend.

Keras is so easy to use that you can develop your first Multilayer Perceptron, Convolutional Neural Network, or LSTM Recurrent Neural Network in minutes.

You may have technical questions when you get started using Keras. You may need a little help.

In this post, you will discover 9 places where you can ask questions and get help when developing deep learning models with Keras.

Let’s dive in.

How to Best Use These Resources

Knowing where to get help is the first step, but you need to know how to get the most out of these resources.

Below are some tips that you can use:

  • Boil your question down to the simplest form. E.g. not “my model does not work” or “how does x work“.
  • Search for answers before asking questions.
  • Provide the smallest possible working example to demonstrate your issue.

1. Keras Users Google Group

Keras Users Google Group

Keras Users Google Group

Perhaps the most direct place to ask questions of the Keras community is the Keras Users group on Google groups (old usenet).

You do not need to get the emails; you can participate online, which I recommend.

2. Keras Slack Channel

Keras Slack Channel

Keras Slack Channel

Perhaps the best place for direct chatting about Keras and related issues is the Keras slack channel.

This has basically replaced IM and IIRC.

You must submit an application to join first.

3. Keras on Gitter

Keras Gitter Group

Keras Gitter Group

Another good place to chat about Keras is on the Keras Gitter.

This is just like slack and still has a good amount of people on it.

Dive in.

4. Keras tag on StackOverflow

Keras tag on StackOverflow

Keras tag on StackOverflow

StackOverflow is a question-and-answer site dedicated to programming.

There are a ton of questions on or related to deep learning and Keras.

I would recommend searching and asking questions that use the “keras” tag.

5. Keras tag on CrossValidated

Keras tag on CrossValidated

Keras tag on CrossValidated

CrossValidated is a question-and-answer site dedicated to machine learning.

There are still a ton of questions on and related to Keras, but they may have a more theoretical bend and less of a focus on code and programming.

Again, I recommend searching and asking questions that use the “keras” tag.

6. Keras tag on DataScience

Keras tag on DataScience Stack Exchange

Keras tag on DataScience Stack Exchange

The StackExchange network that backs StackOverflow and CrossValidated also has a site dedicated to DataScience.

It is still in the beta program, and is not an official member of the network yet. Nevertheless, there are a good number of Keras questions on the site.

Unlike the other two sites, the questions may have a more process orientation.

Again, I recommend searching and asking questions that use the “keras” tag.

7. Keras Topic on Quora

Keras topic on Quora

Keras topic on Quora

Quora is a large general purpose question-and-answer site.

Although it is general purpose, there are a lot of questions on technical topics, including Keras and deep learning.

The focus of the questions are more text-based and expository. You may get more detail on the techniques rather than implementation detail.

I would recommend searching and asking questions in the “Keras” topic.

8. Keras Github Issues

Keras Github Issues

Keras Github Issues

Keras is an open source project that is hosted on GitHub.

GitHub provides a basic issue management system and the Keras project has a lot of issues reported. Although issues are supposed to be limited to technical issues with the code, there is a surprising amount of good discussion on Keras in the issues.

I would recommend searching through the Keras issues but only posting an issue if you have found a bug or have a new feature request (see the guidelines).

9. Keras on Twitter

Keras hash tag on Twitter

Keras hash tag on Twitter

You may get fast answers to one-off Keras questions on twitter.

I would recommend using the Twitter search and searching for “keras” first.

I would also recommend searching and posting using the “keras” hashtag.

You may also get insight by following the creator and lead developer behind Keras, François Chollet.

Other Keras Destinations

Below are other Keras sites where you may find help.

Summary

In this post, you discovered 9 different places around the web where you can get answers and ask questions on Keras for deep learning.

Have you used any of these resources?
How did you do?

Are there are places where you like to get help with Keras?
Please let me know in the comments below.

Frustrated With Your Progress In Deep Learning?

Deep Learning with Python

 What If You Could Develop A Network in Minutes

…with just a few lines of Python

Discover how in my new Ebook: Deep Learning With Python

It covers self-study tutorials and end-to-end projects on topics like:
Multilayer PerceptronsConvolutional Nets and Recurrent Neural Nets, and more…

Finally Bring Deep Learning To
Your Own Projects

Skip the Academics. Just Results.

Click to learn more.

2 Responses to 9 Ways to Get Help with Deep Learning in Keras

  1. Gerrit Govaerts August 2, 2017 at 5:11 pm #

    You forgot François Chollet’s own Keras Blog .He has a very interesting article on it about how he sees AI evolving into AGI : https://blog.keras.io/the-future-of-deep-learning.html

Leave a Reply