profile/dynamic plugins should work in Colab #1913
Labels
Comments
wchargin
added a commit
that referenced
this issue
Feb 28, 2019
Summary: The profile dashboard now shows a “disabled” message on the frontend when we detect that we’re running in a Colab notebook. See #1913. This commit does _not_ cause the profile dashboard to be marked inactive by the main TensorBoard UI. We might want to do that, but it’s less straightforward: currently, the backend is responsible for telling the frontend which plugins are active, and the backend can’t know whether we’re running in Colab. As a quick hack, we could add special-case logic in `tf-tensorboard.html` to check for the profile plugin specifically. As a longer-term solution, we could let the `tf_tensorboard.Dashboard` interface specify an `isActive` callback, defaulting to `() => true`. ![Screenshot of the “Profiling isn’t yet supported in Colab” message][s] [s]: https://user-images.githubusercontent.com/4317806/53533696-a6497d80-3ab0-11e9-935d-56e857a687a8.png The implementation includes a small refactor to the profile dashboard to make it easier to introduce a new state. Test Plan: Evaluate `!!(window.TENSORBOARD_ENV || {}).IN_COLAB` in both standalone TensorBoard and Colab `%tensorboard`, and note that each evaluates to the appropriate boolean. Run TensorBoard locally with the profile plugin’s demo data, and note that its behavior is unchanged. To trigger the loading state, it suffices to load the dashboard, then throttle your network to “Slow 3G”, then change the selected tool. To trigger the “no data” state, relaunch TensorBoard pointing at a different log directory. Then, test the Colab behavior by either pointing TensorBoard at a log directory that does not contain profile data, then executing ```js window.TENSORBOARD_ENV = window.TENSORBOARD_ENV || {}; window.TENSORBOARD_ENV["IN_COLAB"] = true; ``` in the main TensorBoard pane before switching to the inactive profile dashboard. Note that this is the same setup JavaScript as is injected into the Colab output frame by `notebook.py`. wchargin-branch: profile-disable-colab
wchargin
added a commit
that referenced
this issue
Feb 28, 2019
Summary: The profile dashboard now shows a “disabled” message on the frontend when we detect that we’re running in a Colab notebook. See #1913. This commit does _not_ cause the profile dashboard to be marked inactive by the main TensorBoard UI. We might want to do that, but it’s less straightforward: currently, the backend is responsible for telling the frontend which plugins are active, and the backend can’t know whether we’re running in Colab. As a quick hack, we could add special-case logic in `tf-tensorboard.html` to check for the profile plugin specifically. As a longer-term solution, we could let the `tf_tensorboard.Dashboard` interface specify an `isActive` callback, defaulting to `() => true`. ![Screenshot of the “Profiling isn’t yet supported in Colab” message][s] [s]: https://user-images.githubusercontent.com/4317806/53541886-b83b1880-3ad0-11e9-9501-299f5e671d7a.png The implementation includes a small refactor to the profile dashboard to make it easier to introduce a new state. Test Plan: Evaluate `!!(window.TENSORBOARD_ENV || {}).IN_COLAB` in both standalone TensorBoard and Colab `%tensorboard`, and note that each evaluates to the appropriate boolean. Run TensorBoard locally with the profile plugin’s demo data, and note that its behavior is unchanged. To trigger the loading state, it suffices to load the dashboard, then throttle your network to “Slow 3G”, then change the selected tool. To trigger the “no data” state, relaunch TensorBoard pointing at a different log directory. Then, test the Colab behavior by either pointing TensorBoard at a log directory that does not contain profile data, then executing ```js window.TENSORBOARD_ENV = window.TENSORBOARD_ENV || {}; window.TENSORBOARD_ENV["IN_COLAB"] = true; ``` in the main TensorBoard pane before switching to the inactive profile dashboard. Note that this is the same setup JavaScript as is injected into the Colab output frame by `notebook.py`. wchargin-branch: profile-disable-colab
|
Now that #1914 has merged, I’m demoting this to a feature request. |
wchargin
added a commit
to wchargin/tensorboard
that referenced
this issue
Mar 5, 2019
Summary: The profile dashboard now shows a “disabled” message on the frontend when we detect that we’re running in a Colab notebook. See tensorflow#1913. This commit does _not_ cause the profile dashboard to be marked inactive by the main TensorBoard UI. We might want to do that, but it’s less straightforward: currently, the backend is responsible for telling the frontend which plugins are active, and the backend can’t know whether we’re running in Colab. As a quick hack, we could add special-case logic in `tf-tensorboard.html` to check for the profile plugin specifically. As a longer-term solution, we could let the `tf_tensorboard.Dashboard` interface specify an `isActive` callback, defaulting to `() => true`. ![Screenshot of the “Profiling isn’t yet supported in Colab” message][s] [s]: https://user-images.githubusercontent.com/4317806/53541886-b83b1880-3ad0-11e9-9501-299f5e671d7a.png The implementation includes a small refactor to the profile dashboard to make it easier to introduce a new state. Test Plan: Evaluate `!!(window.TENSORBOARD_ENV || {}).IN_COLAB` in both standalone TensorBoard and Colab `%tensorboard`, and note that each evaluates to the appropriate boolean. Run TensorBoard locally with the profile plugin’s demo data, and note that its behavior is unchanged. To trigger the loading state, it suffices to load the dashboard, then throttle your network to “Slow 3G”, then change the selected tool. To trigger the “no data” state, relaunch TensorBoard pointing at a different log directory. Then, test the Colab behavior by either pointing TensorBoard at a log directory that does not contain profile data, then executing ```js window.TENSORBOARD_ENV = window.TENSORBOARD_ENV || {}; window.TENSORBOARD_ENV["IN_COLAB"] = true; ``` in the main TensorBoard pane before switching to the inactive profile dashboard. Note that this is the same setup JavaScript as is injected into the Colab output frame by `notebook.py`. wchargin-branch: profile-disable-colab
Merged
wchargin
added a commit
that referenced
this issue
Mar 6, 2019
Summary: The profile dashboard now shows a “disabled” message on the frontend when we detect that we’re running in a Colab notebook. See #1913. This commit does _not_ cause the profile dashboard to be marked inactive by the main TensorBoard UI. We might want to do that, but it’s less straightforward: currently, the backend is responsible for telling the frontend which plugins are active, and the backend can’t know whether we’re running in Colab. As a quick hack, we could add special-case logic in `tf-tensorboard.html` to check for the profile plugin specifically. As a longer-term solution, we could let the `tf_tensorboard.Dashboard` interface specify an `isActive` callback, defaulting to `() => true`. ![Screenshot of the “Profiling isn’t yet supported in Colab” message][s] [s]: https://user-images.githubusercontent.com/4317806/53541886-b83b1880-3ad0-11e9-9501-299f5e671d7a.png The implementation includes a small refactor to the profile dashboard to make it easier to introduce a new state. Test Plan: Evaluate `!!(window.TENSORBOARD_ENV || {}).IN_COLAB` in both standalone TensorBoard and Colab `%tensorboard`, and note that each evaluates to the appropriate boolean. Run TensorBoard locally with the profile plugin’s demo data, and note that its behavior is unchanged. To trigger the loading state, it suffices to load the dashboard, then throttle your network to “Slow 3G”, then change the selected tool. To trigger the “no data” state, relaunch TensorBoard pointing at a different log directory. Then, test the Colab behavior by either pointing TensorBoard at a log directory that does not contain profile data, then executing ```js window.TENSORBOARD_ENV = window.TENSORBOARD_ENV || {}; window.TENSORBOARD_ENV["IN_COLAB"] = true; ``` in the main TensorBoard pane before switching to the inactive profile dashboard. Note that this is the same setup JavaScript as is injected into the Colab output frame by `notebook.py`. wchargin-branch: profile-disable-colab
Merged
stephanwlee
changed the title
|
Retroactively cleaning issues. #2798 should fix the issue and the fix should be available in tb-nightly (not 2.0.1). |
|
This is awesome!! Thanks for the fix! |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
The profile dashboard currently does not work in Colab, because the
trace viewer is embedded via an iframe, which cannot traverse Colab’s
service worker tunnel for security reasons (service workers never proxy
iframe content).
Ideally, the profile dashboard should work in Colab, which we could by
including the trace viewer directly rather than using an iframe.* At
minimum, we should display a helpful error instead of just failing to
load the content.
* @stephanwlee says that this is not trivial, because even though the
trace viewer uses web components as well, it’s built via a different
mechanism that’s specific to Chrome, and so we’d need some translation
layer for the dependency structure (possibly manual).
Edit: 2019-06-27
Dynamic plugins also employ iframe to render the plugin content which
is not compatible with Colab. We have plans to use new Colab's API,
proxyPort, to mitigate the issue but it is not ready for consumption yet.The text was updated successfully, but these errors were encountered: