SdkInternalList<T> tags
The key-value pairs to use to create tags. If you specify a key without specifying a value, Amazon ML creates a tag with the specified key and a value of null.
String resourceId
The ID of the ML object to tag. For example, exampleModelId.
String resourceType
The type of the ML object to tag.
String batchPredictionId
The ID assigned to the BatchPrediction at creation. This value should be identical to the value of
the BatchPredictionID in the request.
String mLModelId
The ID of the MLModel that generated predictions for the BatchPrediction request.
String batchPredictionDataSourceId
The ID of the DataSource that points to the group of observations to predict.
String inputDataLocationS3
The location of the data file or directory in Amazon Simple Storage Service (Amazon S3).
String createdByIamUser
The AWS user account that invoked the BatchPrediction. The account type can be either an AWS root
account or an AWS Identity and Access Management (IAM) user account.
Date createdAt
The time that the BatchPrediction was created. The time is expressed in epoch time.
Date lastUpdatedAt
The time of the most recent edit to the BatchPrediction. The time is expressed in epoch time.
String name
A user-supplied name or description of the BatchPrediction.
String status
The status of the BatchPrediction. This element can have one of the following values:
PENDING - Amazon Machine Learning (Amazon ML) submitted a request to generate predictions for a
batch of observations.
INPROGRESS - The process is underway.
FAILED - The request to perform a batch prediction did not run to completion. It is not usable.
COMPLETED - The batch prediction process completed successfully.
DELETED - The BatchPrediction is marked as deleted. It is not usable.
String outputUri
The location of an Amazon S3 bucket or directory to receive the operation results. The following substrings are
not allowed in the s3 key portion of the outputURI field: ':', '//', '/./', '/../'.
String message
A description of the most recent details about processing the batch prediction request.
Long computeTime
Date finishedAt
Date startedAt
Long totalRecordCount
Long invalidRecordCount
String batchPredictionId
A user-supplied ID that uniquely identifies the BatchPrediction.
String batchPredictionName
A user-supplied name or description of the BatchPrediction. BatchPredictionName can
only use the UTF-8 character set.
String mLModelId
The ID of the MLModel that will generate predictions for the group of observations.
String batchPredictionDataSourceId
The ID of the DataSource that points to the group of observations to predict.
String outputUri
The location of an Amazon Simple Storage Service (Amazon S3) bucket or directory to store the batch prediction
results. The following substrings are not allowed in the s3 key portion of the
outputURI field: ':', '//', '/./', '/../'.
Amazon ML needs permissions to store and retrieve the logs on your behalf. For information about how to set permissions, see the Amazon Machine Learning Developer Guide.
String batchPredictionId
A user-supplied ID that uniquely identifies the BatchPrediction. This value is identical to the
value of the BatchPredictionId in the request.
String dataSourceId
A user-supplied ID that uniquely identifies the DataSource. Typically, an Amazon Resource Number
(ARN) becomes the ID for a DataSource.
String dataSourceName
A user-supplied name or description of the DataSource.
RDSDataSpec rDSData
The data specification of an Amazon RDS DataSource:
DatabaseInformation -
DatabaseName - The name of the Amazon RDS database.
InstanceIdentifier - A unique identifier for the Amazon RDS database instance.
DatabaseCredentials - AWS Identity and Access Management (IAM) credentials that are used to connect to the Amazon RDS database.
ResourceRole - A role (DataPipelineDefaultResourceRole) assumed by an EC2 instance to carry out the copy task from Amazon RDS to Amazon Simple Storage Service (Amazon S3). For more information, see Role templates for data pipelines.
ServiceRole - A role (DataPipelineDefaultRole) assumed by the AWS Data Pipeline service to monitor the progress of the copy task from Amazon RDS to Amazon S3. For more information, see Role templates for data pipelines.
SecurityInfo - The security information to use to access an RDS DB instance. You need to set up appropriate
ingress rules for the security entity IDs provided to allow access to the Amazon RDS instance. Specify a [
SubnetId, SecurityGroupIds] pair for a VPC-based RDS DB instance.
SelectSqlQuery - A query that is used to retrieve the observation data for the Datasource.
S3StagingLocation - The Amazon S3 location for staging Amazon RDS data. The data retrieved from Amazon RDS using
SelectSqlQuery is stored in this location.
DataSchemaUri - The Amazon S3 location of the DataSchema.
DataSchema - A JSON string representing the schema. This is not required if DataSchemaUri is
specified.
DataRearrangement - A JSON string that represents the splitting and rearrangement requirements for the
Datasource.
Sample - "{\"splitting\":{\"percentBegin\":10,\"percentEnd\":60}}"
String roleARN
The role that Amazon ML assumes on behalf of the user to create and activate a data pipeline in the user's
account and copy data using the SelectSqlQuery query from Amazon RDS to Amazon S3.
Boolean computeStatistics
The compute statistics for a DataSource. The statistics are generated from the observation data
referenced by a DataSource. Amazon ML uses the statistics internally during MLModel
training. This parameter must be set to true if the
DataSource needs to be used for MLModel training.
String dataSourceId
A user-supplied ID that uniquely identifies the datasource. This value should be identical to the value of the
DataSourceID in the request.
String dataSourceId
A user-supplied ID that uniquely identifies the DataSource.
String dataSourceName
A user-supplied name or description of the DataSource.
RedshiftDataSpec dataSpec
The data specification of an Amazon Redshift DataSource:
DatabaseInformation -
DatabaseName - The name of the Amazon Redshift database.
ClusterIdentifier - The unique ID for the Amazon Redshift cluster.
DatabaseCredentials - The AWS Identity and Access Management (IAM) credentials that are used to connect to the Amazon Redshift database.
SelectSqlQuery - The query that is used to retrieve the observation data for the Datasource.
S3StagingLocation - The Amazon Simple Storage Service (Amazon S3) location for staging Amazon Redshift data. The
data retrieved from Amazon Redshift using the SelectSqlQuery query is stored in this location.
DataSchemaUri - The Amazon S3 location of the DataSchema.
DataSchema - A JSON string representing the schema. This is not required if DataSchemaUri is
specified.
DataRearrangement - A JSON string that represents the splitting and rearrangement requirements for the
DataSource.
Sample - "{\"splitting\":{\"percentBegin\":10,\"percentEnd\":60}}"
String roleARN
A fully specified role Amazon Resource Name (ARN). Amazon ML assumes the role on behalf of the user to create the following:
A security group to allow Amazon ML to execute the SelectSqlQuery query on an Amazon Redshift
cluster
An Amazon S3 bucket policy to grant Amazon ML read/write permissions on the S3StagingLocation
Boolean computeStatistics
The compute statistics for a DataSource. The statistics are generated from the observation data
referenced by a DataSource. Amazon ML uses the statistics internally during MLModel
training. This parameter must be set to true if the DataSource needs to be used for
MLModel training.
String dataSourceId
A user-supplied ID that uniquely identifies the datasource. This value should be identical to the value of the
DataSourceID in the request.
String dataSourceId
A user-supplied identifier that uniquely identifies the DataSource.
String dataSourceName
A user-supplied name or description of the DataSource.
S3DataSpec dataSpec
The data specification of a DataSource:
DataLocationS3 - The Amazon S3 location of the observation data.
DataSchemaLocationS3 - The Amazon S3 location of the DataSchema.
DataSchema - A JSON string representing the schema. This is not required if DataSchemaUri is
specified.
DataRearrangement - A JSON string that represents the splitting and rearrangement requirements for the
Datasource.
Sample - "{\"splitting\":{\"percentBegin\":10,\"percentEnd\":60}}"
Boolean computeStatistics
The compute statistics for a DataSource. The statistics are generated from the observation data
referenced by a DataSource. Amazon ML uses the statistics internally during MLModel
training. This parameter must be set to true if the
DataSource needs to be used for MLModel training.
String dataSourceId
A user-supplied ID that uniquely identifies the DataSource. This value should be identical to the
value of the DataSourceID in the request.
String evaluationId
A user-supplied ID that uniquely identifies the Evaluation.
String evaluationName
A user-supplied name or description of the Evaluation.
String mLModelId
The ID of the MLModel to evaluate.
The schema used in creating the MLModel must match the schema of the DataSource used in
the Evaluation.
String evaluationDataSourceId
The ID of the DataSource for the evaluation. The schema of the DataSource must match
the schema used to create the MLModel.
String evaluationId
The user-supplied ID that uniquely identifies the Evaluation. This value should be identical to the
value of the EvaluationId in the request.
String mLModelId
A user-supplied ID that uniquely identifies the MLModel.
String mLModelName
A user-supplied name or description of the MLModel.
String mLModelType
The category of supervised learning that this MLModel will address. Choose from the following types:
Choose REGRESSION if the MLModel will be used to predict a numeric value.
Choose BINARY if the MLModel result has two possible values.
Choose MULTICLASS if the MLModel result has a limited number of values.
For more information, see the Amazon Machine Learning Developer Guide.
SdkInternalMap<K,V> parameters
A list of the training parameters in the MLModel. The list is implemented as a map of key-value
pairs.
The following is the current set of training parameters:
sgd.maxMLModelSizeInBytes - The maximum allowed size of the model. Depending on the input data, the
size of the model might affect its performance.
The value is an integer that ranges from 100000 to 2147483648. The default value is
33554432.
sgd.maxPasses - The number of times that the training process traverses the observations to build
the MLModel. The value is an integer that ranges from 1 to 10000. The
default value is 10.
sgd.shuffleType - Whether Amazon ML shuffles the training data. Shuffling the data improves a
model's ability to find the optimal solution for a variety of data types. The valid values are auto
and none. The default value is none. We strongly recommend that you shuffle your data.
sgd.l1RegularizationAmount - The coefficient regularization L1 norm. It controls overfitting the
data by penalizing large coefficients. This tends to drive coefficients to zero, resulting in a sparse feature
set. If you use this parameter, start by specifying a small value, such as 1.0E-08.
The value is a double that ranges from 0 to MAX_DOUBLE. The default is to not use L1
normalization. This parameter can't be used when L2 is specified. Use this parameter sparingly.
sgd.l2RegularizationAmount - The coefficient regularization L2 norm. It controls overfitting the
data by penalizing large coefficients. This tends to drive coefficients to small, nonzero values. If you use this
parameter, start by specifying a small value, such as 1.0E-08.
The value is a double that ranges from 0 to MAX_DOUBLE. The default is to not use L2
normalization. This parameter can't be used when L1 is specified. Use this parameter sparingly.
String trainingDataSourceId
The DataSource that points to the training data.
String recipe
The data recipe for creating the MLModel. You must specify either the recipe or its URI. If you
don't specify a recipe or its URI, Amazon ML creates a default.
String recipeUri
The Amazon Simple Storage Service (Amazon S3) location and file name that contains the MLModel
recipe. You must specify either the recipe or its URI. If you don't specify a recipe or its URI, Amazon ML
creates a default.
String mLModelId
A user-supplied ID that uniquely identifies the MLModel. This value should be identical to the value
of the MLModelId in the request.
String mLModelId
The ID assigned to the MLModel during creation.
String mLModelId
A user-supplied ID that uniquely identifies the MLModel. This value should be identical to the value
of the MLModelId in the request.
RealtimeEndpointInfo realtimeEndpointInfo
The endpoint information of the MLModel
String dataSourceId
The ID that is assigned to the DataSource during creation.
String dataLocationS3
The location and name of the data in Amazon Simple Storage Service (Amazon S3) that is used by a
DataSource.
String dataRearrangement
A JSON string that represents the splitting and rearrangement requirement used when this DataSource
was created.
String createdByIamUser
The AWS user account from which the DataSource was created. The account type can be either an AWS
root account or an AWS Identity and Access Management (IAM) user account.
Date createdAt
The time that the DataSource was created. The time is expressed in epoch time.
Date lastUpdatedAt
The time of the most recent edit to the BatchPrediction. The time is expressed in epoch time.
Long dataSizeInBytes
The total number of observations contained in the data files that the DataSource references.
Long numberOfFiles
The number of data files referenced by the DataSource.
String name
A user-supplied name or description of the DataSource.
String status
The current status of the DataSource. This element can have one of the following values:
PENDING - Amazon Machine Learning (Amazon ML) submitted a request to create a DataSource.
INPROGRESS - The creation process is underway.
FAILED - The request to create a DataSource did not run to completion. It is not usable.
COMPLETED - The creation process completed successfully.
DELETED - The DataSource is marked as deleted. It is not usable.
String message
A description of the most recent details about creating the DataSource.
RedshiftMetadata redshiftMetadata
RDSMetadata rDSMetadata
String roleARN
Boolean computeStatistics
The parameter is true if statistics need to be generated from the observation data.
Long computeTime
Date finishedAt
Date startedAt
String batchPredictionId
A user-supplied ID that uniquely identifies the BatchPrediction.
String batchPredictionId
A user-supplied ID that uniquely identifies the BatchPrediction. This value should be identical to
the value of the BatchPredictionID in the request.
String dataSourceId
A user-supplied ID that uniquely identifies the DataSource.
String dataSourceId
A user-supplied ID that uniquely identifies the DataSource. This value should be identical to the
value of the DataSourceID in the request.
String evaluationId
A user-supplied ID that uniquely identifies the Evaluation to delete.
String evaluationId
A user-supplied ID that uniquely identifies the Evaluation. This value should be identical to the
value of the EvaluationId in the request.
String mLModelId
A user-supplied ID that uniquely identifies the MLModel.
String mLModelId
A user-supplied ID that uniquely identifies the MLModel. This value should be identical to the value
of the MLModelID in the request.
String mLModelId
The ID assigned to the MLModel during creation.
String mLModelId
A user-supplied ID that uniquely identifies the MLModel. This value should be identical to the value
of the MLModelId in the request.
RealtimeEndpointInfo realtimeEndpointInfo
The endpoint information of the MLModel
SdkInternalList<T> tagKeys
One or more tags to delete.
String resourceId
The ID of the tagged ML object. For example, exampleModelId.
String resourceType
The type of the tagged ML object.
String filterVariable
Use one of the following variables to filter a list of BatchPrediction:
CreatedAt - Sets the search criteria to the BatchPrediction creation date.
Status - Sets the search criteria to the BatchPrediction status.
Name - Sets the search criteria to the contents of the BatchPrediction
Name.
IAMUser - Sets the search criteria to the user account that invoked the BatchPrediction
creation.
MLModelId - Sets the search criteria to the MLModel used in the
BatchPrediction.
DataSourceId - Sets the search criteria to the DataSource used in the
BatchPrediction.
DataURI - Sets the search criteria to the data file(s) used in the BatchPrediction. The
URL can identify either a file or an Amazon Simple Storage Solution (Amazon S3) bucket or directory.
String eQ
The equal to operator. The BatchPrediction results will have FilterVariable values that
exactly match the value specified with EQ.
String gT
The greater than operator. The BatchPrediction results will have FilterVariable values
that are greater than the value specified with GT.
String lT
The less than operator. The BatchPrediction results will have FilterVariable values
that are less than the value specified with LT.
String gE
The greater than or equal to operator. The BatchPrediction results will have
FilterVariable values that are greater than or equal to the value specified with GE.
String lE
The less than or equal to operator. The BatchPrediction results will have
FilterVariable values that are less than or equal to the value specified with LE.
String nE
The not equal to operator. The BatchPrediction results will have FilterVariable values
not equal to the value specified with NE.
String prefix
A string that is found at the beginning of a variable, such as Name or Id.
For example, a Batch Prediction operation could have the Name
2014-09-09-HolidayGiftMailer. To search for this BatchPrediction, select
Name for the FilterVariable and any of the following strings for the
Prefix:
2014-09
2014-09-09
2014-09-09-Holiday
String sortOrder
A two-value parameter that determines the sequence of the resulting list of MLModels.
asc - Arranges the list in ascending order (A-Z, 0-9).
dsc - Arranges the list in descending order (Z-A, 9-0).
Results are sorted by FilterVariable.
String nextToken
An ID of the page in the paginated results.
Integer limit
The number of pages of information to include in the result. The range of acceptable values is 1
through 100. The default value is 100.
SdkInternalList<T> results
A list of BatchPrediction objects that meet the search criteria.
String nextToken
The ID of the next page in the paginated results that indicates at least one more page follows.
String filterVariable
Use one of the following variables to filter a list of DataSource:
CreatedAt - Sets the search criteria to DataSource creation dates.
Status - Sets the search criteria to DataSource statuses.
Name - Sets the search criteria to the contents of DataSource Name.
DataUri - Sets the search criteria to the URI of data files used to create the
DataSource. The URI can identify either a file or an Amazon Simple Storage Service (Amazon S3)
bucket or directory.
IAMUser - Sets the search criteria to the user account that invoked the DataSource
creation.
String eQ
The equal to operator. The DataSource results will have FilterVariable values that
exactly match the value specified with EQ.
String gT
The greater than operator. The DataSource results will have FilterVariable values that
are greater than the value specified with GT.
String lT
The less than operator. The DataSource results will have FilterVariable values that are
less than the value specified with LT.
String gE
The greater than or equal to operator. The DataSource results will have FilterVariable
values that are greater than or equal to the value specified with GE.
String lE
The less than or equal to operator. The DataSource results will have FilterVariable
values that are less than or equal to the value specified with LE.
String nE
The not equal to operator. The DataSource results will have FilterVariable values not
equal to the value specified with NE.
String prefix
A string that is found at the beginning of a variable, such as Name or Id.
For example, a DataSource could have the Name 2014-09-09-HolidayGiftMailer
. To search for this DataSource, select Name for the FilterVariable and
any of the following strings for the Prefix:
2014-09
2014-09-09
2014-09-09-Holiday
String sortOrder
A two-value parameter that determines the sequence of the resulting list of DataSource.
asc - Arranges the list in ascending order (A-Z, 0-9).
dsc - Arranges the list in descending order (Z-A, 9-0).
Results are sorted by FilterVariable.
String nextToken
The ID of the page in the paginated results.
Integer limit
The maximum number of DataSource to include in the result.
SdkInternalList<T> results
A list of DataSource that meet the search criteria.
String nextToken
An ID of the next page in the paginated results that indicates at least one more page follows.
String filterVariable
Use one of the following variable to filter a list of Evaluation objects:
CreatedAt - Sets the search criteria to the Evaluation creation date.
Status - Sets the search criteria to the Evaluation status.
Name - Sets the search criteria to the contents of Evaluation
Name.
IAMUser - Sets the search criteria to the user account that invoked an Evaluation.
MLModelId - Sets the search criteria to the MLModel that was evaluated.
DataSourceId - Sets the search criteria to the DataSource used in
Evaluation.
DataUri - Sets the search criteria to the data file(s) used in Evaluation. The URL can
identify either a file or an Amazon Simple Storage Solution (Amazon S3) bucket or directory.
String eQ
The equal to operator. The Evaluation results will have FilterVariable values that
exactly match the value specified with EQ.
String gT
The greater than operator. The Evaluation results will have FilterVariable values that
are greater than the value specified with GT.
String lT
The less than operator. The Evaluation results will have FilterVariable values that are
less than the value specified with LT.
String gE
The greater than or equal to operator. The Evaluation results will have FilterVariable
values that are greater than or equal to the value specified with GE.
String lE
The less than or equal to operator. The Evaluation results will have FilterVariable
values that are less than or equal to the value specified with LE.
String nE
The not equal to operator. The Evaluation results will have FilterVariable values not
equal to the value specified with NE.
String prefix
A string that is found at the beginning of a variable, such as Name or Id.
For example, an Evaluation could have the Name
2014-09-09-HolidayGiftMailer. To search for this Evaluation, select Name
for the FilterVariable and any of the following strings for the Prefix:
2014-09
2014-09-09
2014-09-09-Holiday
String sortOrder
A two-value parameter that determines the sequence of the resulting list of Evaluation.
asc - Arranges the list in ascending order (A-Z, 0-9).
dsc - Arranges the list in descending order (Z-A, 9-0).
Results are sorted by FilterVariable.
String nextToken
The ID of the page in the paginated results.
Integer limit
The maximum number of Evaluation to include in the result.
SdkInternalList<T> results
A list of Evaluation that meet the search criteria.
String nextToken
The ID of the next page in the paginated results that indicates at least one more page follows.
String filterVariable
Use one of the following variables to filter a list of MLModel:
CreatedAt - Sets the search criteria to MLModel creation date.
Status - Sets the search criteria to MLModel status.
Name - Sets the search criteria to the contents of MLModel Name.
IAMUser - Sets the search criteria to the user account that invoked the MLModel
creation.
TrainingDataSourceId - Sets the search criteria to the DataSource used to train one or
more MLModel.
RealtimeEndpointStatus - Sets the search criteria to the MLModel real-time endpoint
status.
MLModelType - Sets the search criteria to MLModel type: binary, regression, or
multi-class.
Algorithm - Sets the search criteria to the algorithm that the MLModel uses.
TrainingDataURI - Sets the search criteria to the data file(s) used in training a
MLModel. The URL can identify either a file or an Amazon Simple Storage Service (Amazon S3) bucket
or directory.
String eQ
The equal to operator. The MLModel results will have FilterVariable values that exactly
match the value specified with EQ.
String gT
The greater than operator. The MLModel results will have FilterVariable values that are
greater than the value specified with GT.
String lT
The less than operator. The MLModel results will have FilterVariable values that are
less than the value specified with LT.
String gE
The greater than or equal to operator. The MLModel results will have FilterVariable
values that are greater than or equal to the value specified with GE.
String lE
The less than or equal to operator. The MLModel results will have FilterVariable values
that are less than or equal to the value specified with LE.
String nE
The not equal to operator. The MLModel results will have FilterVariable values not
equal to the value specified with NE.
String prefix
A string that is found at the beginning of a variable, such as Name or Id.
For example, an MLModel could have the Name 2014-09-09-HolidayGiftMailer.
To search for this MLModel, select Name for the FilterVariable and any of
the following strings for the Prefix:
2014-09
2014-09-09
2014-09-09-Holiday
String sortOrder
A two-value parameter that determines the sequence of the resulting list of MLModel.
asc - Arranges the list in ascending order (A-Z, 0-9).
dsc - Arranges the list in descending order (Z-A, 9-0).
Results are sorted by FilterVariable.
String nextToken
The ID of the page in the paginated results.
Integer limit
The number of pages of information to include in the result. The range of acceptable values is 1
through 100. The default value is 100.
SdkInternalList<T> results
A list of MLModel that meet the search criteria.
String nextToken
The ID of the next page in the paginated results that indicates at least one more page follows.
String resourceId
The ID of the tagged ML object.
String resourceType
The type of the tagged ML object.
SdkInternalList<T> tags
A list of tags associated with the ML object.
String evaluationId
The ID that is assigned to the Evaluation at creation.
String mLModelId
The ID of the MLModel that is the focus of the evaluation.
String evaluationDataSourceId
The ID of the DataSource that is used to evaluate the MLModel.
String inputDataLocationS3
The location and name of the data in Amazon Simple Storage Server (Amazon S3) that is used in the evaluation.
String createdByIamUser
The AWS user account that invoked the evaluation. The account type can be either an AWS root account or an AWS Identity and Access Management (IAM) user account.
Date createdAt
The time that the Evaluation was created. The time is expressed in epoch time.
Date lastUpdatedAt
The time of the most recent edit to the Evaluation. The time is expressed in epoch time.
String name
A user-supplied name or description of the Evaluation.
String status
The status of the evaluation. This element can have one of the following values:
PENDING - Amazon Machine Learning (Amazon ML) submitted a request to evaluate an
MLModel.
INPROGRESS - The evaluation is underway.
FAILED - The request to evaluate an MLModel did not run to completion. It is not
usable.
COMPLETED - The evaluation process completed successfully.
DELETED - The Evaluation is marked as deleted. It is not usable.
PerformanceMetrics performanceMetrics
Measurements of how well the MLModel performed, using observations referenced by the
DataSource. One of the following metrics is returned, based on the type of the MLModel:
BinaryAUC: A binary MLModel uses the Area Under the Curve (AUC) technique to measure performance.
RegressionRMSE: A regression MLModel uses the Root Mean Square Error (RMSE) technique to measure
performance. RMSE measures the difference between predicted and actual values for a single variable.
MulticlassAvgFScore: A multiclass MLModel uses the F1 score technique to measure performance.
For more information about performance metrics, please see the Amazon Machine Learning Developer Guide.
String message
A description of the most recent details about evaluating the MLModel.
Long computeTime
Date finishedAt
Date startedAt
String batchPredictionId
An ID assigned to the BatchPrediction at creation.
String batchPredictionId
An ID assigned to the BatchPrediction at creation. This value should be identical to the value of
the BatchPredictionID in the request.
String mLModelId
The ID of the MLModel that generated predictions for the BatchPrediction request.
String batchPredictionDataSourceId
The ID of the DataSource that was used to create the BatchPrediction.
String inputDataLocationS3
The location of the data file or directory in Amazon Simple Storage Service (Amazon S3).
String createdByIamUser
The AWS user account that invoked the BatchPrediction. The account type can be either an AWS root
account or an AWS Identity and Access Management (IAM) user account.
Date createdAt
The time when the BatchPrediction was created. The time is expressed in epoch time.
Date lastUpdatedAt
The time of the most recent edit to BatchPrediction. The time is expressed in epoch time.
String name
A user-supplied name or description of the BatchPrediction.
String status
The status of the BatchPrediction, which can be one of the following values:
PENDING - Amazon Machine Learning (Amazon ML) submitted a request to generate batch predictions.
INPROGRESS - The batch predictions are in progress.
FAILED - The request to perform a batch prediction did not run to completion. It is not usable.
COMPLETED - The batch prediction process completed successfully.
DELETED - The BatchPrediction is marked as deleted. It is not usable.
String outputUri
The location of an Amazon S3 bucket or directory to receive the operation results.
String logUri
A link to the file that contains logs of the CreateBatchPrediction operation.
String message
A description of the most recent details about processing the batch prediction request.
Long computeTime
The approximate CPU time in milliseconds that Amazon Machine Learning spent processing the
BatchPrediction, normalized and scaled on computation resources. ComputeTime is only
available if the BatchPrediction is in the COMPLETED state.
Date finishedAt
The epoch time when Amazon Machine Learning marked the BatchPrediction as COMPLETED or
FAILED. FinishedAt is only available when the BatchPrediction is in the
COMPLETED or FAILED state.
Date startedAt
The epoch time when Amazon Machine Learning marked the BatchPrediction as INPROGRESS.
StartedAt isn't available if the BatchPrediction is in the PENDING state.
Long totalRecordCount
The number of total records that Amazon Machine Learning saw while processing the BatchPrediction.
Long invalidRecordCount
The number of invalid records that Amazon Machine Learning saw while processing the BatchPrediction.
String dataSourceId
The ID assigned to the DataSource at creation. This value should be identical to the value of the
DataSourceId in the request.
String dataLocationS3
The location of the data file or directory in Amazon Simple Storage Service (Amazon S3).
String dataRearrangement
A JSON string that represents the splitting and rearrangement requirement used when this DataSource
was created.
String createdByIamUser
The AWS user account from which the DataSource was created. The account type can be either an AWS
root account or an AWS Identity and Access Management (IAM) user account.
Date createdAt
The time that the DataSource was created. The time is expressed in epoch time.
Date lastUpdatedAt
The time of the most recent edit to the DataSource. The time is expressed in epoch time.
Long dataSizeInBytes
The total size of observations in the data files.
Long numberOfFiles
The number of data files referenced by the DataSource.
String name
A user-supplied name or description of the DataSource.
String status
The current status of the DataSource. This element can have one of the following values:
PENDING - Amazon ML submitted a request to create a DataSource.
INPROGRESS - The creation process is underway.
FAILED - The request to create a DataSource did not run to completion. It is not
usable.
COMPLETED - The creation process completed successfully.
DELETED - The DataSource is marked as deleted. It is not usable.
String logUri
A link to the file containing logs of CreateDataSourceFrom* operations.
String message
The user-supplied description of the most recent details about creating the DataSource.
RedshiftMetadata redshiftMetadata
RDSMetadata rDSMetadata
String roleARN
Boolean computeStatistics
The parameter is true if statistics need to be generated from the observation data.
Long computeTime
The approximate CPU time in milliseconds that Amazon Machine Learning spent processing the
DataSource, normalized and scaled on computation resources. ComputeTime is only
available if the DataSource is in the COMPLETED state and the
ComputeStatistics is set to true.
Date finishedAt
The epoch time when Amazon Machine Learning marked the DataSource as COMPLETED or
FAILED. FinishedAt is only available when the DataSource is in the
COMPLETED or FAILED state.
Date startedAt
The epoch time when Amazon Machine Learning marked the DataSource as INPROGRESS.
StartedAt isn't available if the DataSource is in the PENDING state.
String dataSourceSchema
The schema used by all of the data files of this DataSource.
Note: This parameter is provided as part of the verbose format.
String evaluationId
The ID of the Evaluation to retrieve. The evaluation of each MLModel is recorded and
cataloged. The ID provides the means to access the information.
String evaluationId
The evaluation ID which is same as the EvaluationId in the request.
String mLModelId
The ID of the MLModel that was the focus of the evaluation.
String evaluationDataSourceId
The DataSource used for this evaluation.
String inputDataLocationS3
The location of the data file or directory in Amazon Simple Storage Service (Amazon S3).
String createdByIamUser
The AWS user account that invoked the evaluation. The account type can be either an AWS root account or an AWS Identity and Access Management (IAM) user account.
Date createdAt
The time that the Evaluation was created. The time is expressed in epoch time.
Date lastUpdatedAt
The time of the most recent edit to the Evaluation. The time is expressed in epoch time.
String name
A user-supplied name or description of the Evaluation.
String status
The status of the evaluation. This element can have one of the following values:
PENDING - Amazon Machine Language (Amazon ML) submitted a request to evaluate an
MLModel.
INPROGRESS - The evaluation is underway.
FAILED - The request to evaluate an MLModel did not run to completion. It is not
usable.
COMPLETED - The evaluation process completed successfully.
DELETED - The Evaluation is marked as deleted. It is not usable.
PerformanceMetrics performanceMetrics
Measurements of how well the MLModel performed using observations referenced by the
DataSource. One of the following metric is returned based on the type of the MLModel:
BinaryAUC: A binary MLModel uses the Area Under the Curve (AUC) technique to measure performance.
RegressionRMSE: A regression MLModel uses the Root Mean Square Error (RMSE) technique to measure
performance. RMSE measures the difference between predicted and actual values for a single variable.
MulticlassAvgFScore: A multiclass MLModel uses the F1 score technique to measure performance.
For more information about performance metrics, please see the Amazon Machine Learning Developer Guide.
String logUri
A link to the file that contains logs of the CreateEvaluation operation.
String message
A description of the most recent details about evaluating the MLModel.
Long computeTime
The approximate CPU time in milliseconds that Amazon Machine Learning spent processing the
Evaluation, normalized and scaled on computation resources. ComputeTime is only
available if the Evaluation is in the COMPLETED state.
Date finishedAt
The epoch time when Amazon Machine Learning marked the Evaluation as COMPLETED or
FAILED. FinishedAt is only available when the Evaluation is in the
COMPLETED or FAILED state.
Date startedAt
The epoch time when Amazon Machine Learning marked the Evaluation as INPROGRESS.
StartedAt isn't available if the Evaluation is in the PENDING state.
String mLModelId
The MLModel ID, which is same as the MLModelId in the request.
String trainingDataSourceId
The ID of the training DataSource.
String createdByIamUser
The AWS user account from which the MLModel was created. The account type can be either an AWS root
account or an AWS Identity and Access Management (IAM) user account.
Date createdAt
The time that the MLModel was created. The time is expressed in epoch time.
Date lastUpdatedAt
The time of the most recent edit to the MLModel. The time is expressed in epoch time.
String name
A user-supplied name or description of the MLModel.
String status
The current status of the MLModel. This element can have one of the following values:
PENDING - Amazon Machine Learning (Amazon ML) submitted a request to describe a MLModel
.
INPROGRESS - The request is processing.
FAILED - The request did not run to completion. The ML model isn't usable.
COMPLETED - The request completed successfully.
DELETED - The MLModel is marked as deleted. It isn't usable.
Long sizeInBytes
RealtimeEndpointInfo endpointInfo
The current endpoint of the MLModel
SdkInternalMap<K,V> trainingParameters
A list of the training parameters in the MLModel. The list is implemented as a map of key-value
pairs.
The following is the current set of training parameters:
sgd.maxMLModelSizeInBytes - The maximum allowed size of the model. Depending on the input data, the
size of the model might affect its performance.
The value is an integer that ranges from 100000 to 2147483648. The default value is
33554432.
sgd.maxPasses - The number of times that the training process traverses the observations to build
the MLModel. The value is an integer that ranges from 1 to 10000. The
default value is 10.
sgd.shuffleType - Whether Amazon ML shuffles the training data. Shuffling data improves a model's
ability to find the optimal solution for a variety of data types. The valid values are auto and
none. The default value is none. We strongly recommend that you shuffle your data.
sgd.l1RegularizationAmount - The coefficient regularization L1 norm. It controls overfitting the
data by penalizing large coefficients. This tends to drive coefficients to zero, resulting in a sparse feature
set. If you use this parameter, start by specifying a small value, such as 1.0E-08.
The value is a double that ranges from 0 to MAX_DOUBLE. The default is to not use L1
normalization. This parameter can't be used when L2 is specified. Use this parameter sparingly.
sgd.l2RegularizationAmount - The coefficient regularization L2 norm. It controls overfitting the
data by penalizing large coefficients. This tends to drive coefficients to small, nonzero values. If you use this
parameter, start by specifying a small value, such as 1.0E-08.
The value is a double that ranges from 0 to MAX_DOUBLE. The default is to not use L2
normalization. This parameter can't be used when L1 is specified. Use this parameter sparingly.
String inputDataLocationS3
The location of the data file or directory in Amazon Simple Storage Service (Amazon S3).
String mLModelType
Identifies the MLModel category. The following are the available types:
REGRESSION -- Produces a numeric result. For example, "What price should a house be listed at?"
BINARY -- Produces one of two possible results. For example, "Is this an e-commerce website?"
MULTICLASS -- Produces one of several possible results. For example, "Is this a HIGH, LOW or MEDIUM risk trade?"
Float scoreThreshold
The scoring threshold is used in binary classification MLModel models. It marks the boundary between
a positive prediction and a negative prediction.
Output values greater than or equal to the threshold receive a positive result from the MLModel, such as
true. Output values less than the threshold receive a negative response from the MLModel, such as
false.
Date scoreThresholdLastUpdatedAt
The time of the most recent edit to the ScoreThreshold. The time is expressed in epoch time.
String logUri
A link to the file that contains logs of the CreateMLModel operation.
String message
A description of the most recent details about accessing the MLModel.
Long computeTime
The approximate CPU time in milliseconds that Amazon Machine Learning spent processing the MLModel,
normalized and scaled on computation resources. ComputeTime is only available if the
MLModel is in the COMPLETED state.
Date finishedAt
The epoch time when Amazon Machine Learning marked the MLModel as COMPLETED or
FAILED. FinishedAt is only available when the MLModel is in the
COMPLETED or FAILED state.
Date startedAt
The epoch time when Amazon Machine Learning marked the MLModel as INPROGRESS.
StartedAt isn't available if the MLModel is in the PENDING state.
String recipe
The recipe to use when training the MLModel. The Recipe provides detailed information
about the observation data to use during training, and manipulations to perform on the observation data during
training.
Note: This parameter is provided as part of the verbose format.
String schema
The schema used by all of the data files referenced by the DataSource.
Note: This parameter is provided as part of the verbose format.
Integer code
Integer code
Integer code
Integer code
String mLModelId
The ID assigned to the MLModel at creation.
String trainingDataSourceId
The ID of the training DataSource. The CreateMLModel operation uses the
TrainingDataSourceId.
String createdByIamUser
The AWS user account from which the MLModel was created. The account type can be either an AWS root
account or an AWS Identity and Access Management (IAM) user account.
Date createdAt
The time that the MLModel was created. The time is expressed in epoch time.
Date lastUpdatedAt
The time of the most recent edit to the MLModel. The time is expressed in epoch time.
String name
A user-supplied name or description of the MLModel.
String status
The current status of an MLModel. This element can have one of the following values:
PENDING - Amazon Machine Learning (Amazon ML) submitted a request to create an MLModel.
INPROGRESS - The creation process is underway.
FAILED - The request to create an MLModel didn't run to completion. The model isn't
usable.
COMPLETED - The creation process completed successfully.
DELETED - The MLModel is marked as deleted. It isn't usable.
Long sizeInBytes
RealtimeEndpointInfo endpointInfo
The current endpoint of the MLModel.
SdkInternalMap<K,V> trainingParameters
A list of the training parameters in the MLModel. The list is implemented as a map of key-value
pairs.
The following is the current set of training parameters:
sgd.maxMLModelSizeInBytes - The maximum allowed size of the model. Depending on the input data, the
size of the model might affect its performance.
The value is an integer that ranges from 100000 to 2147483648. The default value is
33554432.
sgd.maxPasses - The number of times that the training process traverses the observations to build
the MLModel. The value is an integer that ranges from 1 to 10000. The
default value is 10.
sgd.shuffleType - Whether Amazon ML shuffles the training data. Shuffling the data improves a
model's ability to find the optimal solution for a variety of data types. The valid values are auto
and none. The default value is none.
sgd.l1RegularizationAmount - The coefficient regularization L1 norm, which controls overfitting the
data by penalizing large coefficients. This parameter tends to drive coefficients to zero, resulting in sparse
feature set. If you use this parameter, start by specifying a small value, such as 1.0E-08.
The value is a double that ranges from 0 to MAX_DOUBLE. The default is to not use L1
normalization. This parameter can't be used when L2 is specified. Use this parameter sparingly.
sgd.l2RegularizationAmount - The coefficient regularization L2 norm, which controls overfitting the
data by penalizing large coefficients. This tends to drive coefficients to small, nonzero values. If you use this
parameter, start by specifying a small value, such as 1.0E-08.
The value is a double that ranges from 0 to MAX_DOUBLE. The default is to not use L2
normalization. This parameter can't be used when L1 is specified. Use this parameter sparingly.
String inputDataLocationS3
The location of the data file or directory in Amazon Simple Storage Service (Amazon S3).
String algorithm
The algorithm used to train the MLModel. The following algorithm is supported:
SGD -- Stochastic gradient descent. The goal of SGD is to minimize the gradient of the
loss function.
String mLModelType
Identifies the MLModel category. The following are the available types:
REGRESSION - Produces a numeric result. For example, "What price should a house be listed at?"
BINARY - Produces one of two possible results. For example, "Is this a child-friendly web site?".
MULTICLASS - Produces one of several possible results. For example,
"Is this a HIGH-, LOW-, or MEDIUM-risk trade?".
Float scoreThreshold
Date scoreThresholdLastUpdatedAt
The time of the most recent edit to the ScoreThreshold. The time is expressed in epoch time.
String message
A description of the most recent details about accessing the MLModel.
Long computeTime
Date finishedAt
Date startedAt
SdkInternalMap<K,V> properties
String predictedLabel
The prediction label for either a BINARY or MULTICLASS MLModel.
Float predictedValue
The prediction value for REGRESSION MLModel.
SdkInternalMap<K,V> predictedScores
SdkInternalMap<K,V> details
String mLModelId
A unique identifier of the MLModel.
SdkInternalMap<K,V> record
String predictEndpoint
Prediction prediction
RDSDatabase databaseInformation
Describes the DatabaseName and InstanceIdentifier of an Amazon RDS database.
String selectSqlQuery
The query that is used to retrieve the observation data for the DataSource.
RDSDatabaseCredentials databaseCredentials
The AWS Identity and Access Management (IAM) credentials that are used connect to the Amazon RDS database.
String s3StagingLocation
The Amazon S3 location for staging Amazon RDS data. The data retrieved from Amazon RDS using
SelectSqlQuery is stored in this location.
String dataRearrangement
A JSON string that represents the splitting and rearrangement processing to be applied to a
DataSource. If the DataRearrangement parameter is not provided, all of the input data
is used to create the Datasource.
There are multiple parameters that control what data is used to create a datasource:
percentBegin
Use percentBegin to indicate the beginning of the range of the data used to create the Datasource.
If you do not include percentBegin and percentEnd, Amazon ML includes all of the data
when creating the datasource.
percentEnd
Use percentEnd to indicate the end of the range of the data used to create the Datasource. If you do
not include percentBegin and percentEnd, Amazon ML includes all of the data when
creating the datasource.
complement
The complement parameter instructs Amazon ML to use the data that is not included in the range of
percentBegin to percentEnd to create a datasource. The complement
parameter is useful if you need to create complementary datasources for training and evaluation. To create a
complementary datasource, use the same values for percentBegin and percentEnd, along
with the complement parameter.
For example, the following two datasources do not share any data, and can be used to train and evaluate a model. The first datasource has 25 percent of the data, and the second one has 75 percent of the data.
Datasource for evaluation: {"splitting":{"percentBegin":0, "percentEnd":25}}
Datasource for training: {"splitting":{"percentBegin":0, "percentEnd":25, "complement":"true"}}
strategy
To change how Amazon ML splits the data for a datasource, use the strategy parameter.
The default value for the strategy parameter is sequential, meaning that Amazon ML
takes all of the data records between the percentBegin and percentEnd parameters for
the datasource, in the order that the records appear in the input data.
The following two DataRearrangement lines are examples of sequentially ordered training and
evaluation datasources:
Datasource for evaluation:
{"splitting":{"percentBegin":70, "percentEnd":100, "strategy":"sequential"}}
Datasource for training:
{"splitting":{"percentBegin":70, "percentEnd":100, "strategy":"sequential", "complement":"true"}}
To randomly split the input data into the proportions indicated by the percentBegin and percentEnd parameters,
set the strategy parameter to random and provide a string that is used as the seed
value for the random data splitting (for example, you can use the S3 path to your data as the random seed
string). If you choose the random split strategy, Amazon ML assigns each row of data a pseudo-random number
between 0 and 100, and then selects the rows that have an assigned number between percentBegin and
percentEnd. Pseudo-random numbers are assigned using both the input seed string value and the byte
offset as a seed, so changing the data results in a different split. Any existing ordering is preserved. The
random splitting strategy ensures that variables in the training and evaluation data are distributed similarly.
It is useful in the cases where the input data may have an implicit sort order, which would otherwise result in
training and evaluation datasources containing non-similar data records.
The following two DataRearrangement lines are examples of non-sequentially ordered training and
evaluation datasources:
Datasource for evaluation:
{"splitting":{"percentBegin":70, "percentEnd":100, "strategy":"random", "randomSeed"="s3://my_s3_path/bucket/file.csv"}}
Datasource for training:
{"splitting":{"percentBegin":70, "percentEnd":100, "strategy":"random", "randomSeed"="s3://my_s3_path/bucket/file.csv", "complement":"true"}}
String dataSchema
A JSON string that represents the schema for an Amazon RDS DataSource. The DataSchema
defines the structure of the observation data in the data file(s) referenced in the DataSource.
A DataSchema is not required if you specify a DataSchemaUri
Define your DataSchema as a series of key-value pairs. attributes and
excludedVariableNames have an array of key-value pairs for their value. Use the following format to
define your DataSchema.
{ "version": "1.0",
"recordAnnotationFieldName": "F1",
"recordWeightFieldName": "F2",
"targetFieldName": "F3",
"dataFormat": "CSV",
"dataFileContainsHeader": true,
"attributes": [
{ "fieldName": "F1", "fieldType": "TEXT" }, { "fieldName": "F2", "fieldType": "NUMERIC" }, { "fieldName": "F3", "fieldType": "CATEGORICAL" }, { "fieldName": "F4", "fieldType": "NUMERIC" }, { "fieldName": "F5", "fieldType": "CATEGORICAL" }, { "fieldName": "F6", "fieldType": "TEXT" }, { "fieldName": "F7", "fieldType": "WEIGHTED_INT_SEQUENCE" }, { "fieldName": "F8", "fieldType": "WEIGHTED_STRING_SEQUENCE" } ],
"excludedVariableNames": [ "F6" ] }
String dataSchemaUri
The Amazon S3 location of the DataSchema.
String resourceRole
The role (DataPipelineDefaultResourceRole) assumed by an Amazon Elastic Compute Cloud (Amazon EC2) instance to carry out the copy operation from Amazon RDS to an Amazon S3 task. For more information, see Role templates for data pipelines.
String serviceRole
The role (DataPipelineDefaultRole) assumed by AWS Data Pipeline service to monitor the progress of the copy task from Amazon RDS to Amazon S3. For more information, see Role templates for data pipelines.
String subnetId
The subnet ID to be used to access a VPC-based RDS DB instance. This attribute is used by Data Pipeline to carry out the copy task from Amazon RDS to Amazon S3.
SdkInternalList<T> securityGroupIds
The security group IDs to be used to access a VPC-based RDS DB instance. Ensure that there are appropriate ingress rules set up to allow access to the RDS DB instance. This attribute is used by Data Pipeline to carry out the copy operation from Amazon RDS to an Amazon S3 task.
RDSDatabase database
The database details required to connect to an Amazon RDS.
String databaseUserName
String selectSqlQuery
The SQL query that is supplied during CreateDataSourceFromRDS. Returns only if Verbose is
true in GetDataSourceInput.
String resourceRole
The role (DataPipelineDefaultResourceRole) assumed by an Amazon EC2 instance to carry out the copy task from Amazon RDS to Amazon S3. For more information, see Role templates for data pipelines.
String serviceRole
The role (DataPipelineDefaultRole) assumed by the Data Pipeline service to monitor the progress of the copy task from Amazon RDS to Amazon S3. For more information, see Role templates for data pipelines.
String dataPipelineId
The ID of the Data Pipeline instance that is used to carry to copy data from Amazon RDS to Amazon S3. You can use the ID to find details about the instance in the Data Pipeline console.
Integer peakRequestsPerSecond
The maximum processing rate for the real-time endpoint for MLModel, measured in incoming requests
per second.
Date createdAt
The time that the request to create the real-time endpoint for the MLModel was received. The time is
expressed in epoch time.
String endpointUrl
The URI that specifies where to send real-time prediction requests for the MLModel.
Note: The application must wait until the real-time endpoint is ready before using this URI.
String endpointStatus
The current status of the real-time endpoint for the MLModel. This element can have one of the
following values:
NONE - Endpoint does not exist or was previously deleted.
READY - Endpoint is ready to be used for real-time predictions.
UPDATING - Updating/creating the endpoint.
RedshiftDatabase databaseInformation
Describes the DatabaseName and ClusterIdentifier for an Amazon Redshift
DataSource.
String selectSqlQuery
Describes the SQL Query to execute on an Amazon Redshift database for an Amazon Redshift DataSource.
RedshiftDatabaseCredentials databaseCredentials
Describes AWS Identity and Access Management (IAM) credentials that are used connect to the Amazon Redshift database.
String s3StagingLocation
Describes an Amazon S3 location to store the result set of the SelectSqlQuery query.
String dataRearrangement
A JSON string that represents the splitting and rearrangement processing to be applied to a
DataSource. If the DataRearrangement parameter is not provided, all of the input data
is used to create the Datasource.
There are multiple parameters that control what data is used to create a datasource:
percentBegin
Use percentBegin to indicate the beginning of the range of the data used to create the Datasource.
If you do not include percentBegin and percentEnd, Amazon ML includes all of the data
when creating the datasource.
percentEnd
Use percentEnd to indicate the end of the range of the data used to create the Datasource. If you do
not include percentBegin and percentEnd, Amazon ML includes all of the data when
creating the datasource.
complement
The complement parameter instructs Amazon ML to use the data that is not included in the range of
percentBegin to percentEnd to create a datasource. The complement
parameter is useful if you need to create complementary datasources for training and evaluation. To create a
complementary datasource, use the same values for percentBegin and percentEnd, along
with the complement parameter.
For example, the following two datasources do not share any data, and can be used to train and evaluate a model. The first datasource has 25 percent of the data, and the second one has 75 percent of the data.
Datasource for evaluation: {"splitting":{"percentBegin":0, "percentEnd":25}}
Datasource for training: {"splitting":{"percentBegin":0, "percentEnd":25, "complement":"true"}}
strategy
To change how Amazon ML splits the data for a datasource, use the strategy parameter.
The default value for the strategy parameter is sequential, meaning that Amazon ML
takes all of the data records between the percentBegin and percentEnd parameters for
the datasource, in the order that the records appear in the input data.
The following two DataRearrangement lines are examples of sequentially ordered training and
evaluation datasources:
Datasource for evaluation:
{"splitting":{"percentBegin":70, "percentEnd":100, "strategy":"sequential"}}
Datasource for training:
{"splitting":{"percentBegin":70, "percentEnd":100, "strategy":"sequential", "complement":"true"}}
To randomly split the input data into the proportions indicated by the percentBegin and percentEnd parameters,
set the strategy parameter to random and provide a string that is used as the seed
value for the random data splitting (for example, you can use the S3 path to your data as the random seed
string). If you choose the random split strategy, Amazon ML assigns each row of data a pseudo-random number
between 0 and 100, and then selects the rows that have an assigned number between percentBegin and
percentEnd. Pseudo-random numbers are assigned using both the input seed string value and the byte
offset as a seed, so changing the data results in a different split. Any existing ordering is preserved. The
random splitting strategy ensures that variables in the training and evaluation data are distributed similarly.
It is useful in the cases where the input data may have an implicit sort order, which would otherwise result in
training and evaluation datasources containing non-similar data records.
The following two DataRearrangement lines are examples of non-sequentially ordered training and
evaluation datasources:
Datasource for evaluation:
{"splitting":{"percentBegin":70, "percentEnd":100, "strategy":"random", "randomSeed"="s3://my_s3_path/bucket/file.csv"}}
Datasource for training:
{"splitting":{"percentBegin":70, "percentEnd":100, "strategy":"random", "randomSeed"="s3://my_s3_path/bucket/file.csv", "complement":"true"}}
String dataSchema
A JSON string that represents the schema for an Amazon Redshift DataSource. The
DataSchema defines the structure of the observation data in the data file(s) referenced in the
DataSource.
A DataSchema is not required if you specify a DataSchemaUri.
Define your DataSchema as a series of key-value pairs. attributes and
excludedVariableNames have an array of key-value pairs for their value. Use the following format to
define your DataSchema.
{ "version": "1.0",
"recordAnnotationFieldName": "F1",
"recordWeightFieldName": "F2",
"targetFieldName": "F3",
"dataFormat": "CSV",
"dataFileContainsHeader": true,
"attributes": [
{ "fieldName": "F1", "fieldType": "TEXT" }, { "fieldName": "F2", "fieldType": "NUMERIC" }, { "fieldName": "F3", "fieldType": "CATEGORICAL" }, { "fieldName": "F4", "fieldType": "NUMERIC" }, { "fieldName": "F5", "fieldType": "CATEGORICAL" }, { "fieldName": "F6", "fieldType": "TEXT" }, { "fieldName": "F7", "fieldType": "WEIGHTED_INT_SEQUENCE" }, { "fieldName": "F8", "fieldType": "WEIGHTED_STRING_SEQUENCE" } ],
"excludedVariableNames": [ "F6" ] }
String dataSchemaUri
Describes the schema location for an Amazon Redshift DataSource.
RedshiftDatabase redshiftDatabase
String databaseUserName
String selectSqlQuery
The SQL query that is specified during CreateDataSourceFromRedshift. Returns only if Verbose
is true in GetDataSourceInput.
Integer code
String dataLocationS3
The location of the data file(s) used by a DataSource. The URI specifies a data file or an Amazon
Simple Storage Service (Amazon S3) directory or bucket containing data files.
String dataRearrangement
A JSON string that represents the splitting and rearrangement processing to be applied to a
DataSource. If the DataRearrangement parameter is not provided, all of the input data
is used to create the Datasource.
There are multiple parameters that control what data is used to create a datasource:
percentBegin
Use percentBegin to indicate the beginning of the range of the data used to create the Datasource.
If you do not include percentBegin and percentEnd, Amazon ML includes all of the data
when creating the datasource.
percentEnd
Use percentEnd to indicate the end of the range of the data used to create the Datasource. If you do
not include percentBegin and percentEnd, Amazon ML includes all of the data when
creating the datasource.
complement
The complement parameter instructs Amazon ML to use the data that is not included in the range of
percentBegin to percentEnd to create a datasource. The complement
parameter is useful if you need to create complementary datasources for training and evaluation. To create a
complementary datasource, use the same values for percentBegin and percentEnd, along
with the complement parameter.
For example, the following two datasources do not share any data, and can be used to train and evaluate a model. The first datasource has 25 percent of the data, and the second one has 75 percent of the data.
Datasource for evaluation: {"splitting":{"percentBegin":0, "percentEnd":25}}
Datasource for training: {"splitting":{"percentBegin":0, "percentEnd":25, "complement":"true"}}
strategy
To change how Amazon ML splits the data for a datasource, use the strategy parameter.
The default value for the strategy parameter is sequential, meaning that Amazon ML
takes all of the data records between the percentBegin and percentEnd parameters for
the datasource, in the order that the records appear in the input data.
The following two DataRearrangement lines are examples of sequentially ordered training and
evaluation datasources:
Datasource for evaluation:
{"splitting":{"percentBegin":70, "percentEnd":100, "strategy":"sequential"}}
Datasource for training:
{"splitting":{"percentBegin":70, "percentEnd":100, "strategy":"sequential", "complement":"true"}}
To randomly split the input data into the proportions indicated by the percentBegin and percentEnd parameters,
set the strategy parameter to random and provide a string that is used as the seed
value for the random data splitting (for example, you can use the S3 path to your data as the random seed
string). If you choose the random split strategy, Amazon ML assigns each row of data a pseudo-random number
between 0 and 100, and then selects the rows that have an assigned number between percentBegin and
percentEnd. Pseudo-random numbers are assigned using both the input seed string value and the byte
offset as a seed, so changing the data results in a different split. Any existing ordering is preserved. The
random splitting strategy ensures that variables in the training and evaluation data are distributed similarly.
It is useful in the cases where the input data may have an implicit sort order, which would otherwise result in
training and evaluation datasources containing non-similar data records.
The following two DataRearrangement lines are examples of non-sequentially ordered training and
evaluation datasources:
Datasource for evaluation:
{"splitting":{"percentBegin":70, "percentEnd":100, "strategy":"random", "randomSeed"="s3://my_s3_path/bucket/file.csv"}}
Datasource for training:
{"splitting":{"percentBegin":70, "percentEnd":100, "strategy":"random", "randomSeed"="s3://my_s3_path/bucket/file.csv", "complement":"true"}}
String dataSchema
A JSON string that represents the schema for an Amazon S3 DataSource. The DataSchema
defines the structure of the observation data in the data file(s) referenced in the DataSource.
You must provide either the DataSchema or the DataSchemaLocationS3.
Define your DataSchema as a series of key-value pairs. attributes and
excludedVariableNames have an array of key-value pairs for their value. Use the following format to
define your DataSchema.
{ "version": "1.0",
"recordAnnotationFieldName": "F1",
"recordWeightFieldName": "F2",
"targetFieldName": "F3",
"dataFormat": "CSV",
"dataFileContainsHeader": true,
"attributes": [
{ "fieldName": "F1", "fieldType": "TEXT" }, { "fieldName": "F2", "fieldType": "NUMERIC" }, { "fieldName": "F3", "fieldType": "CATEGORICAL" }, { "fieldName": "F4", "fieldType": "NUMERIC" }, { "fieldName": "F5", "fieldType": "CATEGORICAL" }, { "fieldName": "F6", "fieldType": "TEXT" }, { "fieldName": "F7", "fieldType": "WEIGHTED_INT_SEQUENCE" }, { "fieldName": "F8", "fieldType": "WEIGHTED_STRING_SEQUENCE" } ],
"excludedVariableNames": [ "F6" ] }
String dataSchemaLocationS3
Describes the schema location in Amazon S3. You must provide either the DataSchema or the
DataSchemaLocationS3.
String key
A unique identifier for the tag. Valid characters include Unicode letters, digits, white space, _, ., /, =, +, -, %, and @.
String value
An optional string, typically used to describe or define the tag. Valid characters include Unicode letters, digits, white space, _, ., /, =, +, -, %, and @.
String batchPredictionId
The ID assigned to the BatchPrediction during creation. This value should be identical to the value
of the BatchPredictionId in the request.
String dataSourceId
The ID assigned to the DataSource during creation. This value should be identical to the value of
the DataSourceID in the request.
String evaluationId
The ID assigned to the Evaluation during creation. This value should be identical to the value of
the Evaluation in the request.
String mLModelId
The ID assigned to the MLModel during creation.
String mLModelName
A user-supplied name or description of the MLModel.
Float scoreThreshold
The ScoreThreshold used in binary classification MLModel that marks the boundary
between a positive prediction and a negative prediction.
Output values greater than or equal to the ScoreThreshold receive a positive result from the
MLModel, such as true. Output values less than the ScoreThreshold receive
a negative response from the MLModel, such as false.
String mLModelId
The ID assigned to the MLModel during creation. This value should be identical to the value of the
MLModelID in the request.
Copyright © 2023. All rights reserved.