[module][langchain_mlrun] First version#963
[module][langchain_mlrun] First version#963guy1992l wants to merge 53 commits intomlrun:developmentfrom
Conversation
[Build] Fix html links, add <function>.html as source in documentation
[XGB-Custom] Fix test artifact key name
[XGB-Serving][XGB-Test][XGB-Trainer] Fix tests - artifact key
* [Build] Install python 3.9 when testing * [Build] Update python version in CI * .
* [Build] Build with python 3.9 * .
* [Noise-reduction] Add new function to hub * fix test * added multiprocessing and silence removal to function
This reverts commit 8872798.
* delete EOS functions * bring back validate_great_expectations * bring back load_dataset
* adjust batch infer v2 * update docs in NB
* [text to audio generator] Replaced bark with openai tts models * [text to audio generator] Fix base url env var * fix version * Add speech engine * after review
* [Build] Fix html links, Add <function>.html as source in documentation * Update CI temporarily and update index * [XGB-Custom] Fix test artifact key name * [XGB-Serving][XGB-Test][XGB-Trainer] Fix tests - artifact key * [Build] Install python 3.9 when testing (mlrun#618) * [Build] Update python version in CI (mlrun#620) * [Build] Install python 3.9 when testing * [Build] Update python version in CI * . * Revert "[Build] Update python version in CI (mlrun#620)" (mlrun#621) This reverts commit 0cd1f15. * Revert "[Build] Install python 3.9 when testing (mlrun#618)" (mlrun#619) This reverts commit 3301415. * [Build] Build with python 3.9 (mlrun#622) * [Build] Build with python 3.9 * . * Update requirements.txt
* fix feature_selection * fix feature_selection * fix feature_selection nb * update yaml name * fix test * fix test
* fix arbitrary file vulnerability * fix arbitrary file vulnerability * fix test
* add traversal test * add traversal test * add traversal test
|
Check out this pull request on See visual diffs & provide feedback on Jupyter Notebooks. Powered by ReviewNB |
Eyal-Danieli
left a comment
There was a problem hiding this comment.
very nice, had few comments
| :returns: An MLRun model endpoint monitoring client. | ||
| """ | ||
| if mlrun.mlconf.is_ce_mode(): | ||
| return _KafkaMLRunEndPointClient( |
There was a problem hiding this comment.
missing monitoring_broker and monitoring_topic params
There was a problem hiding this comment.
its a "hachana lemazgan"... there is a TODO basically anywhere related to Kafka usage.
| @@ -0,0 +1,23 @@ | |||
| apiVersion: v1 | |||
| categories: | |||
| - langchain | |||
There was a problem hiding this comment.
@Eyal-Danieli, which categories do we support?
Should it be on general for now?
There was a problem hiding this comment.
@guy1992l do we have an option to deploy the agent code into the user cluster?
There was a problem hiding this comment.
In general, we have a closed mapping of the categories that you can see in the UI (https://www.mlrun.org/hub/) but your question is in place as we should add it to the contribution guidelines (I'll open a ticket). As for the new categories within this PR, we already added them (will be available once this PR is merged).
| @@ -0,0 +1,3 @@ | |||
| pytest | |||
| langchain | |||
There was a problem hiding this comment.
Worth pointing out to a specific version, to make sure the function do not change between package releases
| @@ -0,0 +1,899 @@ | |||
| { | |||
There was a problem hiding this comment.
I think we can generate the stream path by using the project name of the user or this env MLRUN_ACTIVE_PROJECT in addition, model monitoring always uses the project container so we can just set these values as default.
In this example, the user does not create any serving function, so we can just set the value under the hood if the user does not provide one.
I think we can remove this line - This is a temporary workaround until custom endpoint creation support is added to MLRun.
Once MLRun will add this feature, we will update the notebook
Reply via ReviewNB
| @@ -0,0 +1,899 @@ | |||
| { | |||
There was a problem hiding this comment.
- We should use the same
requirements.txtfile. If it should be a different file I think it would be better to create anapplication_requirements.txt - Maybe worth to write another step that invoke the chain after the application deployed
Reply via ReviewNB
Created the module
langchain_mlrun- MLRun's integration with LangChain, allowing to auto-trace and monitor LangChain and LangGraph based code.