Release 0.8.6
Need Help?
The SuperCowPowers team is happy to give any assistance needed when setting up AWS and SageWorks. So please contact us at sageworks@supercowpowers.com or on chat us up on Discord
The SageWorks framework continues to flex to support different real world use cases when operating a set of production machine learning pipelines. We've also fixed various corner cases mostly around 'half constructed' AWS artifacts (models/endpoints).
Additional Functionality
- View Support (phase 1)
- CloudWatch (phase 1)
- Exception Log Forwarding Exception Handling
- Lambda Layers SageWorks Lambda Layers
- Better Docs for Deploying Plugins Deploying Plugins
Issues Addressed
-
Model to Endpoint under AWS Throttle
A corner case where the
to_endpoint()
method would fail when not 'knowing' the model input. This happened when AWS was throttling responses and theget_input()
of the Endpoint returnedunknown
which caused aNoneType
error when using the 'unknown' model. -
Empty Model Package Groups
There are cases where customers might construct a Model Package Group (MPG) container and not put any Model Packages in that Group. SageWorks has assumed that all MPGs would have at least one model package. The current 'support' for empty MPGs treats it as an error condition but the API tries to accommodate the condition and will properly display the model group. The group will indicate that it's 'empty' and provides an alert health icons.
Questions?
The SuperCowPowers team is happy to answer any questions you may have about AWS and SageWorks. Please contact us at sageworks@supercowpowers.com or on chat us up on Discord