Known issues

The following are known issues in the Splunk Machine Learning Toolkit version 2.2.0.

Date filed	Issue number	Description
2017-09-07	MLA-1927	Python for Scientific Computing does not support REST handling
2017-06-08	MLA-1856	MTLK< 2.2 models cannot be used on MLTK >= 2.2
2017-04-28	MLA-1845	KMeans error when nulls are present Workaround: Add a fillnull to the search before KMeans is used.
2017-04-26	MLA-1842	Partial fit SGDClassifier gets no handle_categorical attribute
2017-04-18	MLA-1811	Fields with mixed numeric and categorical values may be misrepresented during fit and apply, causing a column mismatch error
2017-02-28	MLA-1670	"Outliers Over Time" plot doesn't display any data if the _time field does not contain a Unix timestamp.
2017-02-28	MLA-1666	The Data and Outliers visualization in the Detect Numeric Outliers assistant is not always displayed.
2017-02-24	MLA-1648	Outliers Chart is blank if timeseries data are not sorted
2016-12-13	MLA-1485	In the Forecast Time Series assistant, forecasting a very large number of points may cause the browser to crash.
2016-12-12	MLA-1473	DBSCAN algorithm sometimes outputs floating point cluster labels.
2016-11-30	MLA-1450	ElasticNet fails if only one field is specified for "Fields to use for predicting" and "estimate the intercept" is disabled. Workaround: Use Linear Regression for this use case.
2016-11-18	MLA-1421	MLTK Custom Visualizations may be out of date after updating the app if using Splunk Enterprise 6.4. Workaround: Clear your browser cache or force the Splunk platform to reset its web cache with the following URL: <yoursplunk>/en-US/_bump
2016-11-16	MLA-1392	Edit Job Settings search bar menu item does not work on all dashboards.
2016-08-24	MLA-1145	Machine Learning Toolkit does not load correctly if its folder is renamed.
2016-07-19	MLA-1014	Fields with integer values and missing values are converted into floating point values