The following are known issues in the Splunk Machine Learning Toolkit version 5.3.1. If no known issues are listed, no issues have been reported.
|Date filed||Issue number||Description|
|2022-08-24||MLA-4320||Experiments' Search bar does not work with map command when used with $$ such as date_hour=$hour$|
Run your SPL in the regular Search bar. If you want to see SODA outlier visualizations, add
|2022-07-14||MLA-4301||exclude_dist parameter breaks backward compatibility if the model is created with an earlier version of MLTK and partial_fit is used|
The user can update their modelfile name and run Template:Partial fit to create a new modelfile with the new version of MLTK 5.3.1 and have no issues afterwards.
|2022-07-03||MLA-4298, MLA-4160||'apply' search processor too aggressive in throwing exceptions|
|2022-04-11||MLA-4277||Smart Outlier Detection - All time search runs even if time picker is used on define page|
|2022-02-14||MLA-4256||Save button under Settings is disabled|
To update the local value of any parameter add its new value into the $SPLUNK_HOME/etc/apps/Splunk_ML_Toolkit/local/mlspl.conf file.
For example: \[DensityFunction]
max_groups = 4096
|2022-01-24||MLA-4247||Infodesk - MLTK Experiment creation doesn't return after preprocessing step|
|2022-01-10||MLA-4234||Deleting an alert after creating it does not work|
Go to the list of alerts and delete your alert from there.
|2021-11-25||MLA-4178||Experiment Alert search not created correctly from Assistant|
Manually adjust the created search in savedsearches.conf or through the UI. Use the SPL created by the assistants a start.
|2021-07-09||MLA-4026||What-If Analysis does not properly display on some Unix platforms|
|2020-08-03||MLA-3950||Apply command does not run if used from a sub search|
appendcolscommand to append the column that will be returned from the subsearch to the savedsearch's columns, instead of assigning its value to a non-existing column in the savedsearch. Example:
| inputlookup track_day.csv | fit LogisticRegression fit_intercept=true "vehicleType" from "batteryVoltage" "engineCoolantTemperature" "engineSpeed" "lateralGForce" "longitudeGForce" "speed" "verticalGForce" into "example_vehicle_type"
makeresults count=50000save as
|savedsearch MLA3950 | appendcols \[| inputlookup track_day.csv | apply "example_vehicle_type" | table vehicleType]This will append the
vehicleTypecolumn from the
|2020-07-09||MLA-3944||MLTK forecast RMSE and R-square calculations are incorrect because it is using ALL data points instead of just data points in holdback period|
|2019-12-17||MLA-3701||Search bars in Assistants will not display more than approximately 500k rows|
This is mostly a visual issue - the actual number of results the Assistant will use when fitting a model is not affected.
|2019-12-11||MLA-3686||MLTK Assistants do not support quote characters in field names|
Use the "rename" or "eval" command to rename the field prior to using it.
|2019-10-09||MLA-3559||Loading settings from experiment history while in the Learn stage will lead to warnings, even when inputs are correct|
|2019-08-30||MLA-3423||Add a warning message when kfold_cv is set to number greater than half of the datasize|
|2019-07-12||MLA-3237||Anova table does not allow some special characters in formula so in field names also|
Rename your field names before running "score anova" if they include one or more of those special characters:
|2019-03-28||MLA-2965||"Edit Job Settings" does not work in Assistants|
|2018-09-24||MLA-2547||Experiments can't be renamed until they have been filled and saved|
Load the Experiment, fill all necessary fields, and save it. You'll be able to rename the Experiment in the Save dialog.
|2018-06-12||MLA-2258||TFIDF returns a vague error message when the "token_pattern" parameter is invalid|
|2018-04-18||MLA-2147||search.log does not show tracebacks on Windows on Splunk 7.1|
The user can still look at the logs on windows in mlspl.log which captures the traceback. However, many users may not have access to mlspl.log (as it is indexed in the _internal index which normal users don't have by default). If this is the case, they will need to ask their Splunk administrators for access to this index.
|2017-08-25||MLA-1916||Uncaught StatsModels error with ARIMA order=0-2-0|
|2017-02-28||MLA-1670||"Outliers Over Time" plot doesn't display any data if the _time field does not contain a Unix timestamp.|
|2016-12-13||MLA-1485||In the Forecast Time Series assistant, forecasting a very large number of points may cause the browser to crash.|
This documentation applies to the following versions of Splunk® Machine Learning Toolkit: 5.3.1
Feedback submitted, thanks!