machine-learning

machine-learning copied to clipboard

When a user uploads a dataset, on the web-interface, it's possible a segment is malformed. We will need to adjust the backend validator, to return the segment that is incorrect,...

jeff1evesque

We will provide each user the ability to inventory each collection of datasets they have permission to access. Then, we will nest the corresponding documents under each collection. The user...

jeff1evesque

Stress test 'programmatic-api'

2

We will standardize some python scripts, which will manually stress test the programmatic api. Since stress testing will occur less frequent then both the automatic, and manual unit tests, it...

jeff1evesque

We need to create unit tests for the web-interface, as an anonymous user.

jeff1evesque

Create unit tests for web-interface as authenticated

2

We need to create unit tests for the web-interface, as an authenticated user.

jeff1evesque

Create AdaBoost Wrapper (Regressor)

3

Create a python function for the ada boost regressor methods which supports SVC, KNN, RF, and normal decision trees.

protojas

We need to unit test the adaboost (classifier) ensemble method, and develop any backend dependency, to allow the programmatic-api, to cooperate with the desired unit test.

jeff1evesque

Investigate 'bootstrap aggregation' implementation

22

We need to investigate the requirements to streamline the following bootstrap aggregation methods: - [bagging classifier](http://scikit-learn.org/stable/modules/generated/sklearn.ensemble.BaggingClassifier.html) - [bagging regressor](http://scikit-learn.org/stable/modules/generated/sklearn.ensemble.BaggingRegressor.html)

jeff1evesque

Investigate 'boosting' implementation

2

We need to investigate the requirements to streamline boosting implementation

jeff1evesque

We need to restructure our `model_type` from `application.yaml`: ```yaml model_type: - svm - svr - bagc - bagr ``` To the following: ```yaml model_type: classifiers: - svm - bagc regressors:...

jeff1evesque