honeymoose
/
OpenSearch
mirror of https://github.com/honeymoose/OpenSearch.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

[DOCS] Adds delta and offset parameters to Evaluate DFA API docs (#63317) (#63329)

This commit is contained in:
István Zoltán Szabó 2020-10-06 16:49:08 +02:00 committed by GitHub
parent 7a59ae8fa2
commit a3a373b67f
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 33 additions and 17 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

50
docs/reference/ml/df-analytics/apis/evaluate-dfanalytics.asciidoc
View File

@ -26,7 +26,8 @@ If the {es} {security-features} are enabled, you must have the following privile
* cluster: `monitor_ml` * cluster: `monitor_ml`
For more information, see <<security-privileges>> and {ml-docs-setup-privileges}. For more information, see <<security-privileges>> and
{ml-docs-setup-privileges}.
[[ml-evaluate-dfanalytics-desc]] [[ml-evaluate-dfanalytics-desc]]
@ -68,8 +69,8 @@ source index. See <<query-dsl>>.
[[oldetection-resources]] [[oldetection-resources]]
=== {oldetection-cap} evaluation objects === {oldetection-cap} evaluation objects
{oldetection-cap} evaluates the results of an {oldetection} analysis which outputs {oldetection-cap} evaluates the results of an {oldetection} analysis which
the probability that each document is an outlier. outputs the probability that each document is an outlier.
`actual_field`:: `actual_field`::
(Required, string) The field of the `index` which contains the `ground truth`. (Required, string) The field of the `index` which contains the `ground truth`.
@ -120,24 +121,39 @@ which outputs a prediction of values.
in other words the results of the {regression} analysis. in other words the results of the {regression} analysis.
`metrics`:: `metrics`::
(Optional, object) Specifies the metrics that are used for the evaluation. (Optional, object) Specifies the metrics that are used for the evaluation. For
more information on `mse`, `msle`, and `huber`, consult
https://github.com/elastic/examples/tree/master/Machine%20Learning/Regression%20Loss%20Functions[the Jupyter notebook on regression loss functions].
Available metrics: Available metrics:
`mse`::: `mse`:::
(Optional, object) Average squared difference between the predicted values and the actual (`ground truth`) value. (Optional, object) Average squared difference between the predicted values
For more information, read {wikipedia}/Mean_squared_error[this wiki article]. and the actual (`ground truth`) value. For more information, read
{wikipedia}/Mean_squared_error[this wiki article].
`msle`::: `msle`:::
(Optional, object) Average squared difference between the logarithm of the predicted values and the logarithm of the actual (Optional, object) Average squared difference between the logarithm of the
(`ground truth`) value. predicted values and the logarithm of the actual (`ground truth`) value.
`offset`::::
(Optional, double) Defines the transition point at which you switch from
minimizing quadratic error to minimizing quadratic log error. Defaults to
`1`.
`huber`::: `huber`:::
(Optional, object) Pseudo Huber loss function. (Optional, object) Pseudo Huber loss function. For more information, read
For more information, read {wikipedia}/Huber_loss#Pseudo-Huber_loss_function[this wiki article]. {wikipedia}/Huber_loss#Pseudo-Huber_loss_function[this wiki article].
`delta`::::
(Optional, double) Approximates 1/2 (prediction - actual)^2^ for values
much less than delta and approximates a straight line with slope delta for
values much larger than delta. Defaults to `1`. Delta needs to be greater
than `0`.
`r_squared`::: `r_squared`:::
(Optional, object) Proportion of the variance in the dependent variable that is predictable from the independent variables. (Optional, object) Proportion of the variance in the dependent variable that
For more information, read {wikipedia}/Coefficient_of_determination[this wiki article]. is predictable from the independent variables. For more information, read
{wikipedia}/Coefficient_of_determination[this wiki article].
@ -171,16 +187,16 @@ belongs.
`auc_roc`::: `auc_roc`:::
(Optional, object) The AUC ROC (area under the curve of the receiver (Optional, object) The AUC ROC (area under the curve of the receiver
operating characteristic) score and optionally the curve. operating characteristic) score and optionally the curve.
It is calculated for a specific class (provided as "class_name") It is calculated for a specific class (provided as "class_name") treated as
treated as positive. positive.
`class_name`:::: `class_name`::::
(Required, string) Name of the only class that will be treated as (Required, string) Name of the only class that will be treated as
positive during AUC ROC calculation. Other classes will be treated as positive during AUC ROC calculation. Other classes will be treated as
negative ("one-vs-all" strategy). Documents which do not have `class_name` negative ("one-vs-all" strategy). Documents which do not have `class_name`
in the list of their top classes will not be taken into account for evaluation. in the list of their top classes will not be taken into account for
The number of documents taken into account is returned in the evaluation result evaluation. The number of documents taken into account is returned in the
(`auc_roc.doc_count` field). evaluation result (`auc_roc.doc_count` field).
`include_curve`:::: `include_curve`::::
(Optional, boolean) Whether or not the curve should be returned in (Optional, boolean) Whether or not the curve should be returned in