Christopher Choquette Choo

Building from Docker on clean machine fails

8

In trying to build this project using the docker images, with the following commands: make check_gss OS_ID=ubuntu1804 RM_CONTAINER=false and make check_gss OS_ID=ubuntu1804 on a clean matching with ubuntu 20 OS...

Unittests show that the high level BoltOnModel has issues on some fit tests for beta1. alpha0 and beta0 should be stable for the new PR.

#### Description Uses hyperparam to enable automatic param distribution for foreshadow object.

#### Description Update the documentation, doctests, and tutorials to be user friendly. Make the users guide and the developers guide (use the project and customize/make components). Estimate: 3 days

Are we removing this test case? _Originally posted by @JingJZ160 in https://github.com/georgianpartners/foreshadow/pull/143_

#### Description In sklearn BaseSearchCV.fit, which calls _validation._fit_and_score() with a cloned estimator for each parallel call, the cloned estimators have a new instance of column_sharer created for each subobject rather...

Switch patching to use patchy

1

#### Description https://pypi.org/project/patchy/ _fit from sklearn.pipeline.Pipeline. Estimate: 1 day

Hyperparemeter Tuning

1

#### Description Hyperparameter tuning will involve the following: An object that will enable 'combinations' to be tried. This object will be serialized much like any other and when deserialized, will...

Automated hyperparameter Tuning for FS.

1

#### Description see #133 which is manual hyperparameter by the user. This issue directly extends from #133 as FeatureEngineerer and Intents both may change the pipeline (# steps, # columns)....

#### Description Make sure that docs are working (are displayed, compiled properly, etc.). Deployment for readthedoc tool/server was failing. Solution: check the blog post published by @adithyabsk Estimate: 1 day