Developers

Get creative with machine learning data and the OpenML API.


Download OpenML data to use it in novel ways, or upload new data to increase their impact (and yours).

Latest source code

Get the latest source code for all OpenML tools and plugins by downloading it directly from GitHub. We will soon add links to the developer versions of individual plugins.

Download OpenML

Clone or fork via GitHub

Clone the entire project or fork your own version of OpenML tools to make them your own by visiting us on GitHub. Or, join a team working on an OpenML plugin.

OpenML on GitHub

Database snapshots

OpenML consists of two databases, a public database (the experiment database) and a private database (containing all user data, etc). The Snapshot of the public database contains all experiment runs, evaluations and links to datasets, implementations and result files. The Snapshot of the private database contains the table structure and important records, like usergroups. Both in SQL format (gzipped).

Download public database Download private database

Feature requests

Feature request, as well as issues, can be posted in the community discussions, or by directly opening an issue in our GitHub project.

An overview of the most common use cases with working examples and links to the full documentation of the services involved.

Search datasets

Will be added soon

Search implementations

Will be added soon

Search tasks

Will be added soon

Search results

Will be added soon

Download a dataset

  1. User asks for a dataset using the openml.data.description service and a dataset id. The dataset id is typically part of a task, or returned when searching for datasets.
  2. OpenML returns a description of the dataset as an XML file. Try it now
  3. The dataset description contains the URL where the dataset can be downloaded. The user calls that URL to download the dataset.
  4. The dataset is returned by the server hosting the dataset. This can be OpenML, but also any other data repository. Try it now
Services:

Download an implementation

  1. User asks for an implementation using the openml.implementation.get service and a implementation id. The implementation id is typically returned when searching for implementations.
  2. OpenML returns a description of the implementation as an XML file. Try it now
  3. The implementation description contains the URL where the implementation can be downloaded, either as source, binary or both, as well as additional information on history, dependencies and licence. The user calls the right URL to download it.
  4. The implementation is returned by the server hosting it. This can be OpenML, but also any other code repository. Try it now
Services:

Download a task

  1. User asks for a task using the openml.tasks.search service and a task id. The task id is typically returned when searching for tasks.
  2. OpenML returns a description of the task as an XML file. Try it now
  3. The task description contains the dataset id(s) of the datasets involved in this task. The user asks for the dataset using the openml.data.description service and the dataset id.
  4. OpenML returns a description of the dataset as an XML file. Try it now
  5. The dataset description contains the URL where the dataset can be downloaded. The user calls that URL to download the dataset.
  6. The dataset is returned by the server hosting it. This can be OpenML, but also any other data repository. Try it now
  7. (Optional) The task description may also contain links to other resources, such as the train-test splits to be used in cross-validation. The user calls that URL to download the train-test splits.
  8. (Optional) The train-test splits are returned by OpenML. Try it now
Services:

Upload a dataset

  1. User authenticates herself by calling openml.authenticate with her username and (hashed) password. OpenML will return an authentication session token.
  2. The user uploads the dataset together a dataset description and her session token to openml.data.upload. The dataset description is an XML file that contains at least the dataset name and a textual description. For now, the only truly supported dataset format is ARFF.
  3. OpenML stores the uploaded dataset and returns the registered dataset id.
Services:

Upload an implementation

  1. User authenticates herself by calling openml.authenticate with her username and (hashed) password. OpenML will return an authentication session token.
  2. The user uploads her session token, implementation description, the implementation binary and/or the implementation source to openml.implementation.upload. The implementation description is an XML file that contains at least the implementation name and a textual description. The implementation binary and source will typically be a ZIP file. An implementation can be a single algorithm or a composed workflow.
  3. OpenML stores the uploaded implementation and returns the registered implementation id.
Services:

Upload a run

  1. User authenticates herself by calling openml.authenticate with her username and (hashed) password. OpenML will return an authentication session token.
  2. The user uploads a run description and any run result files together with her session token to openml.run.upload. The run description is an XML file that contains the task id of the task it addresses and (optionally) a list of parameter settings if these differ from the default settings in the used implementation. The run result files contain the results of the run as detailed in the corresponding task description.
  3. OpenML stores the uploaded run and its results and returns a task-specific response. This can include, for instance, evaluations computed by the server based on uploaded predictions.
Services:

Details of all OpenML services, with their expected arguments, file formats, responses and error codes.

Using REST services

REST is the simplest request format to use - it's a simple HTTP GET or POST action.

The REST Endpoint URL is

http://www.openml.org/api/

For instance, to request the openml.data.description service, invoke like this:

http://www.openml.org/api/?f=openml.data.description&data_id=1

Responses are always in XML format, also when an error is returned. Error messages will look like this:

<oml:error xmlns:oml="http://openml.org/error">
  <oml:code>100</oml:code>
  <oml:message>Please invoke legal function</oml:message>
  <oml:additional_information>Additional information, not always available. </oml:additional_information>
</oml:error>

The error codes and messages for each service are listed below.

openml.authenticate

Returns a session_hash, which can be used for writing to the API.

Arguments
POST username (Required)
The username to be authenticated with
POST password (Required)
An md5 hash of the password, corresponding to the username
Example Response
<oml:authenticate xmlns:oml="http://openml.org/openml">
  <oml:session_hash>CYWJBLVYIPQ42IGB1NHSTGP181Y4TQIWE45GGQ4P</oml:session_hash>
  <oml:valid_until>2020-01-01 00:00:00</oml:valid_until>
</oml:authenticate>
Error codes
250: Please provide username
Please provide the username as a POST variable
251: Please provide password
Please provide the password (hashed as a MD5) as a POST variable
252: Authentication failed
The username and password did not match any record in the database. Please note that the password should be hashed using md5

openml.data.description

Returns dataset descriptions in XML

Arguments
GET data_id (Required)
The dataset id
Response
Dataset description file (XML)

Contains a URL where the dataset can be downloaded, as well as additional information on history, format and licence.

XSD Schema
Example Response
<oml:data_set_description xmlns:oml="http://openml.org/openml">
  <oml:id>1</oml:id>
  <oml:name>anneal</oml:name>
  <oml:version>1.0</oml:version>
  <oml:description>...</oml:description>
  <oml:format>ARFF</oml:format>
  <oml:upload_date>2013-02-26 16:20:57</oml:upload_date>
  <oml:licence>public domain</oml:licence>
  <oml:url>http://expdb.cs.kuleuven.be/expdb/data/uci/nominal/anneal.arff</oml:url>
  <oml:md5_checksum>08dc9d6bf8e5196de0d56bfc89631931</oml:md5_checksum>
<oml:data_set_description>
Error codes
110: Please provide data_id
Please provide data_id
111: Unknown dataset
Data set description with data_id was not found in the database

openml.data.features

Returns the features (attributes) of a given dataset

Arguments
GET data_id (Required)
The dataset id
Response
Dataset features (XML)

Contains a list of the dataset features (attributes), including index (ordering) and data type.

XSD Schema
Example Response
<oml:data_set_features xmlns:oml="http://openml.org/openml">
  <oml:data_set>
    <oml:id>61</oml:id>
    <oml:name>iris</oml:name>
  </oml:data_set>
  <oml:feature>
    <oml:name>sepallength</oml:name>
    <oml:data_type>numeric</oml:data_type>
    <oml:index>0</oml:index>
  </oml:feature>
  <oml:feature>
    <oml:name>sepalwidth</oml:name>
    <oml:data_type>numeric</oml:data_type>
    <oml:index>1</oml:index>
  </oml:feature>
  <oml:feature>
    <oml:name>petallength</oml:name>
    <oml:data_type>numeric</oml:data_type>
    <oml:index>2</oml:index>
  </oml:feature>
  <oml:feature>
    <oml:name>petalwidth</oml:name>
    <oml:data_type>numeric</oml:data_type>
    <oml:index>3</oml:index>
  </oml:feature>
  <oml:feature>
    <oml:name>class</oml:name>
    <oml:data_type>nominal</oml:data_type>
    <oml:index>4</oml:index>
  </oml:feature>
</oml:data_set_features>
Error codes
270: Please provide data_id
Please provide data_id
271: Unknown dataset
Data set description with data_id was not found in the database
272: No features found
The registered dataset did not contain any features

openml.data.licences

Returns a list of known data licences.

Arguments
None
Response
Dataset licences (XML)

List of known dataset licences.

XSD Schema
Example Response
<oml:data_licences xmlns:oml="http://openml.org/openml">
  <oml:licences>
    <oml:licence>public domain</oml:licence>
  </oml:licences>
</oml:data_licences>
Error codes
None

openml.data.upload

Uploads and registers new datasets.

Arguments
POST description (Required)
An XML file containing the data set description
POST dataset (Required)
The dataset file to be stored on the server
POST session_hash (Required)
The session hash, provided by the server on authentication (1 hour valid)
Required file
Dataset description (XML)

Description of the uploaded dataset. Should contain at least name and a textual description, but can also contain versioning, creator, format and licensing information.

XSD Schema
Response
Uploaded dataset (XML)

The id of the stored dataset.

XSD Schema
Example Response
<oml:upload_data_set xmlns:oml="http://openml.org/openml">
  <oml:id>719294</oml:id>
</oml:upload_data_set>
Error codes
130: Problem with file uploading
There was a problem with the file upload
131: Problem validating uploaded description file
The XML description format does not meet the standards
132: Failed to move the files
Internal server error, please contact api administrators
133: Failed to make checksum of datafile
Internal server error, please contact api administrators
134: Failed to insert record in database
Internal server error, please contact api administrators
135: Please provide description xml
Please provide description xml
136: Error slot open
Error slot open, will be filled by not yet defined error
137: Please provide session_hash
In order to share content, please authenticate (openml.authenticate) and provide session_hash
138: Authentication failed
The session_hash was not valid. Please try to login again, or contact api administrators
139: Combination name / version already exists
The combination of name and version of this dataset already exists. Leave version out for auto increment
140: Both dataset file and dataset url provided. Please provide only one
The system is confused since both a dataset file (post) and a dataset url (xml) are provided. Please remove one
141: Neither dataset file or dataset url are provided
Please provide either a dataset file as POST variable, xor a dataset url in the description XML

openml.tasks.types

Returns a list of known machine learning task types.

Arguments
None
Response
Task types (XML)

A list of available task types with name and description.

XSD Schema
Example Response
<oml:task_types xmlns:oml="http://openml.org/openml">
  <oml:task_type>
    <oml:id>1</oml:id>
    <oml:name>Supervised Classification</oml:name>
    <oml:description>Given a dataset with a classification target and a set of train/test splits, e.g. generated by a cross-validation procedure, train a model and return the predictions of that model.</oml:description>
    <oml:creator>Joaquin Vanschoren</oml:creator>
  </oml:task_type>
  <oml:task_type>
    <oml:id>2</oml:id>
    <oml:name>Supervised Regression</oml:name>
    <oml:description>Given a dataset with a numeric target and a set of train/test splits, e.g. generated by a cross-validation procedure, train a model and return the predictions of that model.</oml:description>
    <oml:creator>Joaquin Vanschoren</oml:creator>
  </oml:task_type>
</oml:task_types>
Error codes
None

Returns a definition (template) of a certain task type.

Arguments
GET task_type_id (Required)
The task type id
Response
Task types (XML)

A list of available task types with name and description.

XSD Schema
Example Response
<oml:task_type xmlns:oml="http://openml.org/openml">
  <oml:id>1</oml:id>
  <oml:name>Supervised Classification</oml:name>
  <oml:description>Given a dataset with a classification target and a set of train/test splits, e.g. generated by a cross-validation procedure, train a model and return the predictions of that model.</oml:description>
  <oml:creator>Joaquin Vanschoren</oml:creator>
  <oml:contributor>Jan van Rijn</oml:contributor>
  <oml:contributor>Bo Gao</oml:contributor>
  <oml:contributor>Simon Fischer</oml:contributor>
  <oml:contributor>Venkatesh Umaashankar</oml:contributor>
  <oml:contributor>Luis Torgo</oml:contributor>
  <oml:contributor>Bernd Bischl</oml:contributor>
  <oml:contributor>Michael Berthold</oml:contributor>
  <oml:contributor>Bernd Wiswedel</oml:contributor>
  <oml:contributor>Patrick Winter</oml:contributor>
  <oml:date>21-01-2013</oml:date>
  <oml:input name="source_data">
    <oml:description>The source data used to evaluate the model</oml:description>
    <oml:data_set>
      <oml:data_set_id>input:1</oml:data_set_id>
      <oml:target_feature>input:2</oml:target_feature>
    </oml:data_set>
  </oml:input>
  <oml:input name="estimation_procedure">
    <oml:description>The evaluation procedure used to evaluate the model</oml:description>
    <oml:estimation_procedure>
      <oml:type>input:3</oml:type>
      <oml:data_splits_url>input:4</oml:data_splits_url>
      <oml:parameter name="number_folds">input:6</oml:parameter>
      <oml:parameter name="number_repeats">input:5</oml:parameter>
      <oml:parameter name="stratified_sampling">true</oml:parameter>
    </oml:estimation_procedure>
  </oml:input>
  <oml:input name="evaluation_measures">
    <oml:description>Optional. A list of evaluation measures to optimize for</oml:description>
    <oml:evaluation_measures>input:8</oml:evaluation_measures>
  </oml:input>
  <oml:output name="predictions">
    <oml:description>The predictions returned by your implementation.</oml:description>
    <oml:predictions>
      <oml:format>ARFF</oml:format>
      <oml:feature name="confidence.classname" type="numeric"/>
      <oml:feature name="fold" type="integer"/>
      <oml:feature name="prediction" type="string"/>
      <oml:feature name="repeat" type="integer"/>
      <oml:feature name="row_id" type="integer"/>
    </oml:predictions>
  </oml:output>
</oml:task_type>
Error codes
240: Please provide task_type_id
Please provide task_type_id
241: Unknown task type
The task type with this id was not found in the database

Returns tasks in XML.

Arguments
GET task_id (Required)
The task id
Response
Task description (XML)

A task description defines exactly the input data needed to solve the task, and what output data should be returned.

XSD Schema
Example Response
<oml:task xmlns:oml="http://openml.org/openml">
  <oml:task_id>1</oml:task_id>
  <oml:task_type>Supervised Classification</oml:task_type>
  <oml:input name="source_data">
    <oml:data_set>
      <oml:data_set_id>61</oml:data_set_id>
      <oml:target_feature>class</oml:target_feature>
    </oml:data_set>
  </oml:input>
  <oml:input name="estimation_procedure">
    <oml:estimation_procedure>
      <oml:type>cross_validation</oml:type>
      <oml:data_splits_url>http://expdb.cs.kuleuven.be/expdb/data/splits/iris_splits_CV_10_2.arff</oml:data_splits_url>
      <oml:parameter name="number_folds">10</oml:parameter>
      <oml:parameter name="number_repeats">2</oml:parameter>
      <oml:parameter name="stratified_sampling">true</oml:parameter>
    </oml:estimation_procedure>
  </oml:input>
  <oml:input name="evaluation_measures">
    <oml:evaluation_measures>
      <oml:evaluation_measure>predictive_accuracy</oml:evaluation_measure>
    </oml:evaluation_measures>
  </oml:input>
  <oml:output name="predictions">
    <oml:predictions>
      <oml:format>ARFF</oml:format>
      <oml:feature name="confidence.classname" type="numeric"/>
      <oml:feature name="fold" type="integer"/>
      <oml:feature name="prediction" type="string"/>
      <oml:feature name="repeat" type="integer"/>
      <oml:feature name="row_id" type="integer"/>
    </oml:predictions>
  </oml:output>
</oml:task>
Error codes
150: Please provide task_id
Please provide task_id
151: Unknown task
The task with this id was not found in the database

openml.implementation.licences

Returns a list of known implementation licences.

Arguments
None
Example Response
<oml:implementation_licences xmlns:oml="http://openml.org/openml">
  <oml:licences/>
</oml:implementation_licences>
Error codes
None

openml.implementation.upload

Uploads and registers new implementations.

Arguments
POST description (Required)
An XML file containing the implementation meta data
POST source
The source code of the implementation. If multiple files, please zip them. Either source or binary is required.
POST binary
The binary of the implementation. If multiple files, please zip them. Either source or binary is required.
POST session_hash (Required)
The session hash, provided by the server on authentication (1 hour valid)
Required file
Implementation description (XML)

Description of the implementation. Should at least contain a name and textual description, but can also contain versioning, creator, format and licensing information.

XSD Schema
Response
Uploaded implementation (XML)

The id of the registered implementation.

XSD Schema
Example Response
<oml:upload_implementation xmlns:oml="http://openml.org/openml">
  <oml:id>knime.janvanrijn.solveTaskA_1.5.10</oml:id>
</oml:upload_implementation>
Error codes
160: Error in file uploading
There was a problem with the file upload
161: Please provide description xml
Please provide description xml
162: Please provide source or binary file
Please provide source or binary file. It is also allowed to upload both
163: Problem validating uploaded description file
The XML description format does not meet the standards
164: Implementation already stored in database
Please change name or version number
165: Failed to move the files
Internal server error, please contact api administrators
166: Failed to add implementation to database
Internal server error, please contact api administrators
167: Illegal files uploaded
An non required file was uploaded.
168: The provided md5 hash equals not the server generated md5 hash of the file
The provided md5 hash equals not the server generated md5 hash of the file
169: Please provide session_hash
In order to share content, please authenticate (openml.authenticate) and provide session_hash
170: Authentication failed
The session_hash was not valid. Please try to login again, or contact api administrators

openml.implementation.get

Returns the description file of an implementation.

Arguments
GET implementation_id (Required)
The implementation id (typically: name_version)
Response
Implementation description (XML)

An implementation has a URL where it can be downloaded, either as source, binary or both, as well as additional information on history, dependencies and licence.

XSD Schema
Example Response
<oml:error xmlns:oml="http://openml.org/openml">
  <oml:code>181</oml:code>
  <oml:message>Unknown implementation</oml:message>
</oml:error>
Error codes
180: Please provide implementation_id
Please provide implementation_id
181: Unknown implementation
The implementation with this ID was not found in the database

openml.evaluation.measures

Returns a list of supported evaluation measures.

Arguments
None
Example Response
<oml:evaluation_measures xmlns:oml="http://openml.org/openml">
  <oml:measures>
    <oml:measure>area_under_roc_curve</oml:measure>
    <oml:measure>average_cost</oml:measure>
    <oml:measure>build_cpu_time</oml:measure>
    <oml:measure>build_memory</oml:measure>
    <oml:measure>class_complexity</oml:measure>
    <oml:measure>class_complexity_gain</oml:measure>
    <oml:measure>confusion_matrix</oml:measure>
    <oml:measure>correlation_coefficient</oml:measure>
    <oml:measure>f_measure</oml:measure>
    <oml:measure>kappa</oml:measure>
    <oml:measure>kb_relative_information_score</oml:measure>
    <oml:measure>kohavi_wolpert_bias_squared</oml:measure>
    <oml:measure>kohavi_wolpert_error</oml:measure>
    <oml:measure>kohavi_wolpert_sigma_squared</oml:measure>
    <oml:measure>kohavi_wolpert_variance</oml:measure>
    <oml:measure>kononenko_bratko_information_score</oml:measure>
    <oml:measure>matthews_correlation_coefficient</oml:measure>
    <oml:measure>mean_absolute_error</oml:measure>
    <oml:measure>mean_area_under_roc_curve</oml:measure>
    <oml:measure>mean_class_complexity</oml:measure>
    <oml:measure>mean_class_complexity_gain</oml:measure>
    <oml:measure>mean_f_measure</oml:measure>
    <oml:measure>mean_kononenko_bratko_information_score</oml:measure>
    <oml:measure>mean_precision</oml:measure>
    <oml:measure>mean_prior_absolute_error</oml:measure>
    <oml:measure>mean_prior_class_complexity</oml:measure>
    <oml:measure>mean_recall</oml:measure>
    <oml:measure>mean_weighted_f_measure</oml:measure>
    <oml:measure>mean_weighted_precision</oml:measure>
    <oml:measure>mean_weighted_recall</oml:measure>
    <oml:measure>precision</oml:measure>
    <oml:measure>predictive_accuracy</oml:measure>
    <oml:measure>prior_class_complexity</oml:measure>
    <oml:measure>prior_entropy</oml:measure>
    <oml:measure>recall</oml:measure>
    <oml:measure>relative_absolute_error</oml:measure>
    <oml:measure>root_mean_prior_squared_error</oml:measure>
    <oml:measure>root_mean_squared_error</oml:measure>
    <oml:measure>root_relative_squared_error</oml:measure>
    <oml:measure>run_cpu_time</oml:measure>
    <oml:measure>run_memory</oml:measure>
    <oml:measure>run_virtual_memory</oml:measure>
    <oml:measure>single_point_area_under_roc_curve</oml:measure>
    <oml:measure>total_cost</oml:measure>
    <oml:measure>unclassified_instance_count</oml:measure>
    <oml:measure>webb_bias</oml:measure>
    <oml:measure>webb_error</oml:measure>
    <oml:measure>webb_variance</oml:measure>
  </oml:measures>
</oml:evaluation_measures>
Error codes
None

openml.evaluation.methods

Returns a list of supported evaluation methods.

Arguments
None
Example Response
<oml:evaluation_methods xmlns:oml="http://openml.org/openml">
  <oml:methods>
    <oml:method>
      <oml:type>cross_validation</oml:type>
      <oml:folds>undefined</oml:folds>
      <oml:repeats>undefined</oml:repeats>
    </oml:method>
  </oml:methods>
</oml:evaluation_methods>
Error codes
None

openml.run.upload

Uploads new runs.

Arguments
POST description (Required)
An XML file describing the run
POST <output_files> (Required)
All output files that should be generated by the run, as described in the task xml. For supervised classification tasks, this is typically a file containing predictions
POST session_hash (Required)
The session hash, provided by the server on authentication (1 hour valid)
Required file
Run description (XML)

Description of the run. Should contain at least task_id and implementation_id, and optionally any parameter setting that are specific for this run. Only set the field error_message in case of an error that prevented the algorithm from successfully executing. In this case, any predictions generated will not be evaluated, and the run is stored as unfinished.

XSD Schema
Response
Run response (XML)

The response file returned depending on the task type, containing at least the run id of the stored run. For some task types, it may contain additional information. For instance, for supervised classification, it will contain evaluation measures based on the uploaded predictions, computed on the server.

XSD Schema
Example Response
<oml:upload_run xmlns:oml="http://openml.org/openml">
  <oml:run_id>718193</oml:run_id>
</oml:upload_run>
Error codes
200: Please provide session_hash
In order to share content, please authenticate (openml.authenticate) and provide session_hash
201: Authentication failed
The session_hash was not valid. Please try to login again, or contact api administrators
202: Please provide run xml
Please provide run xml
203: Could not validate run xml by xsd
Please double check that the xml is valid.
204: Unknown task
The task with this id was not found in the database
205: Unknown implementation
The implementation with this id was not found in the database
206: Invalid number of files
The number of uploaded files did not match the number of files expected for this task type
207: File upload failed
One of the files uploaded has a problem
208: Error inserting setup record
Internal server error, please contact api administrators
209: Unable to store cvrun
Internal server error, please contact api administrators
210: Unable to store run
Internal server error, please contact api administrators
211: Dataset not in databse
One of the datasets of this task was not included in database, please contact api administrators
212: Unable to store file
Internal server error, please contact api administrators
213: Parameter in run xml unknown
One of the parameters provided in the run xml is not registered as parameter for the implementation nor its components
214: Unable to store input setting
Internal server error, please contact API support team
215: Unable to evaluate predictions
Internal server error, please contact API support team
216: Error thrown by Java Application
The Java application has thrown an error. Additional information field is provided

openml.run.get

Returns all relevant information to the run, e.g., setup, input data and output data.

Arguments
GET run_id (Required)
The run id
Response
Run response (XML)

The response file returns an XML file obaying the following XSD file:

XSD Schema
Example Response
<?xml version="1.0" encoding="UTF-8"?>
<oml:get_run xmlns:oml="http://openml.org/openml">
  <oml:run_id>718354</oml:run_id>
  <oml:task_id>2</oml:task_id>
  <oml:user_id>1</oml:user_id>
  <oml:implementation>J48(1.0)</oml:implementation>
  <oml:setup_id>635061</oml:setup_id>
  <oml:input_data>
    <oml:dataset>
      <oml:did>61</oml:did>
      <oml:name>iris</oml:name>
      <oml:url>http://expdb.cs.kuleuven.be/expdb/data/uci/nominal/iris.arff</oml:url>
    </oml:dataset>
  </oml:input_data>
  <oml:output_data>
    <oml:dataset>
      <oml:did>719987</oml:did>
      <oml:name>run-718354-predictions_task2.arff</oml:name>
      <oml:url>http://www.openml.org/files/download/138/predictions_task2.arff</oml:url>
    </oml:dataset>
    <oml:evaluation>
      <oml:name>mean_absolute_error</oml:name>
      <oml:value>0.0354851786666666</oml:value>
    </oml:evaluation>
    <oml:evaluation>
      <oml:name>mean_prior_absolute_error</oml:name>
      <oml:value>0.444444444444454</oml:value>
    </oml:evaluation>
    <oml:evaluation>
      <oml:name>root_mean_squared_error</oml:name>
      <oml:value>0.157833805003545</oml:value>
    </oml:evaluation>
    <oml:evaluation>
      <oml:name>root_mean_prior_squared_error</oml:name>
      <oml:value>0.471404520791037</oml:value>
    </oml:evaluation>
    <oml:evaluation>
      <oml:name>relative_absolute_error</oml:name>
      <oml:value>0.0798416519999981</oml:value>
    </oml:evaluation>
    <oml:evaluation>
      <oml:name>root_relative_squared_error</oml:name>
      <oml:value>0.334816061455441</oml:value>
    </oml:evaluation>
    <oml:evaluation>
      <oml:name>predictive_accuracy</oml:name>
      <oml:value>0.955333333333333</oml:value>
    </oml:evaluation>
    <oml:evaluation>
      <oml:name>kappa</oml:name>
      <oml:value>0.933</oml:value>
    </oml:evaluation>
    <oml:evaluation>
      <oml:name>prior_entropy</oml:name>
      <oml:value>1.58496250072116</oml:value>
    </oml:evaluation>
    <oml:evaluation>
      <oml:name>kb_relative_information_score</oml:name>
      <oml:value>1396.29996987614</oml:value>
    </oml:evaluation>
  </oml:output_data>
</oml:get_run>
Error codes
220: Please provide run_id
In order to view run details, please provide run_id
221: Run not found
The run id was not found