Alibi deployment example - Anchor Tabular #629
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
jklaise
suggested changes
Jun 17, 2019
There was a problem hiding this comment.
clf.fit(preprocessor.transform(X_train), Y_train)--> needs to be y_train- can we use the latest s2-wrapper? (0.8)
- don't forget to port-forward ambassador and supply port (localhost:port, or otherwise specify nodeport is needed?)
- was getting timeouts with the default manifest config, so added a higher timeout, but the explanation just hangs for several minutes on the given test instance before timing out again - need to investigate what's going on
resp = sc_pred.predict--> needs to beresp = sc.predicty_validis not definedsc_xaishould besc?
|
Thanks for the comments @jklaise! I've addressed them now 👍 |
|
As an update, @jklaise has gone through this PR, but has encountered issues when running the explainer due to version missmatch between the local and remote environemnts. I have worked with @jklaise to solve this, but I am not able to replicate this issue, even with a clean conda install. @gsunner or @ryandawsonuk could have a look and see if it works properly for you? |
jklaise
approved these changes
Jul 5, 2019
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Alibi Anchor Tabular Seldon Deployment Example
This PR contains an example of an Alibi Anchor Tabular deployment, building upon the CogX Demo showcasing the income classifier example. The example can be found through the README in the base folder, as well as directly through the Jupyter Notebook provided.
The objective of this example is to serve as a basis of an end-to-end example that works with Kubeflow, as well as with further showcase with Seldon Deploy without much customisation. Furthermore, this has been very useful to also extend the existing Kubeflow NLP End-to-end example to showcase a full end-to-end explainer demo on the Anchor Text.
The architecture of the example is the following: