-
Notifications
You must be signed in to change notification settings - Fork 19.5k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Fix Issue #11384 - Explaining how to load external backends in the documentation #11405
Conversation
Added documentation for loading an external backend
Documentation added to "Switching from one backend to another" section of documentation
Not sure why the most recent commit failed the travis CI testing, as the commit just before that managed to pass. The latest commit passes all tests except for Python 3.6 KERAS_BACKEND=cntk PYTHONWARNINGS=ignore. If anyone has any solutions please let me know |
Currently, the travis build has a random timout which we have no information about. It's been there for a while and we have no idea how to fix it. So it can happend that the build fails randomly with the message "no output has been recieved since 10 minutes". |
@farizrahman4u I'll let you review it if you don't mind. I think you're better fitted for the task. |
@gabrieldemarmiesse Its supposed to be a module from a package rather than a standalone file. Like |
docs/templates/backend.md
Outdated
@@ -45,6 +45,22 @@ KERAS_BACKEND=tensorflow python -c "from keras import backend" | |||
Using TensorFlow backend. | |||
``` | |||
|
|||
In keras it is possible to load more backends than `"tensorflow"`, `"theano"`, and `"cntk"`. keras can use external backends as well, and this can be performed by changing the `keras.json` configuration file, and the `"backend"` setting. Suppose you have a python file called mxnet_backend.py that you wanted to use as your external backend. The `keras.json` configuration file would be changed as follows: |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
MXNet is not a good example, because there is, in fact, a Keras MXNet which is a fork and cannot be used as Keras backend.
@farizrahman4u do you have an example of backend which is working is ready to be loaded with this method? |
@gabrieldemarmiesse The one for which I made this change is not open source (yet) :( |
Ok I suppose that we can suppose that we have a backend called @JoshuaRM Can you replace mxnet by this dummy backend |
@farizrahman4u Could you review? I pushed a commit so I can't review this PR. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Perfect.
… in the documentation (keras-team#11405) * Update backend.md Added documentation for loading an external backend * Added documentation for using an external backend * Added documentation for loading external backends * Added documentation to correct section Documentation added to "Switching from one backend to another" section of documentation * Used a dummy name for the module containing the backend.
Summary
Created additional documentation for the "Switching from one backend to another" section of the documentation, including an example of an edited keras.json configuration file.
Related Issues
Solves issue #11384
PR Overview