Brian Wylie

Add a page to the dashboard that shows license info. The main page will have a link to it when you click the top details. - Company - AWS Account...

We're going to go through these SaaS guidelines and check off as many of these boxes as we can. https://docs.aws.amazon.com/marketplace/latest/userguide/saas-guidelines.html

We have a new views class that manages Athena views, it has the display, training, etc views we just need to have the current data_source, feature_set class use the display...

https://sageworks-dashboard.com/models So when we switch the inference runs, there's something flakey going on when updating the inference metrics beneath.

The new Quantile Regression + Residuals model current just uses features for inference, we can also take in an optional target column and add a 'abs distance from q_50' bit...

Right now we use both the enumerated type and the string type randomly. We should figure out exactly what we want to do. I think we'll mostly use the enumerated...

- sageworks/core/artifacts/model_core.py", line 616 Endpoint Core ``` # Compute the standard performance metrics for this model model_type = self.model_type() if model_type == ModelType.REGRESSOR.value: prediction_df = self.residuals(target_column, prediction_df) metrics = self.regression_metrics(target_column,...

Let's take a deep dive on this paper and see if we could do an implementation in SageWorks: https://jmlr.csail.mit.edu/papers/volume22/20-1462/20-1462.pdf

Let's add Quartile Regression Models. This will allow us to provide some metrics around uncertainly for regression predictions.

Onboarding for DataSources is doing a LOT of meta data refreshes.. let's find out why ``` ds.set_display_columns(ds.column_names()) 2024-09-04 13:40:19 (data_source_abstract.py:98) IMPORTANT Setting Display Columns...['id', 'name', 'height', 'weight', 'salary', 'age', 'iq_score',...