Reference documentation and code samples for the Google Cloud Ai Platform V1 Client class StratifiedSplit.
Assigns input data to the training, validation, and test sets so that the
distribution of values found in the categorical column (as specified by thekeyfield) is mirrored within each split. The fraction values determine
the relative sizes of the splits.
For example, if the specified column has three values, with 50% of the rows
having value "A", 25% value "B", and 25% value "C", and the split fractions
are specified as 80/10/10, then the training set will constitute 80% of the
training data, with about 50% of the training set rows having the value "A"
for the specified column, about 25% having the value "B", and about 25%
having the value "C".
Only the top 500 occurring values are used; any values not in the top
500 values are randomly assigned to a split. If less than three rows contain
a specific value, those rows are randomly assigned.
Supported only for tabular Datasets.
Generated from protobuf messagegoogle.cloud.aiplatform.v1.StratifiedSplit
Namespace
Google \ Cloud \ AIPlatform \ V1
Methods
__construct
Constructor.
Parameters
Name
Description
data
array
Optional. Data for populating the Message object.
↳ training_fraction
float
The fraction of the input data that is to be used to train the Model.
↳ validation_fraction
float
The fraction of the input data that is to be used to validate the Model.
↳ test_fraction
float
The fraction of the input data that is to be used to evaluate the Model.
↳ key
string
Required. The key is a name of one of the Dataset's data columns. The key provided must be for a categorical column.
getTrainingFraction
The fraction of the input data that is to be used to train the Model.
Returns
Type
Description
float
setTrainingFraction
The fraction of the input data that is to be used to train the Model.
Parameter
Name
Description
var
float
Returns
Type
Description
$this
getValidationFraction
The fraction of the input data that is to be used to validate the Model.
Returns
Type
Description
float
setValidationFraction
The fraction of the input data that is to be used to validate the Model.
Parameter
Name
Description
var
float
Returns
Type
Description
$this
getTestFraction
The fraction of the input data that is to be used to evaluate the Model.
Returns
Type
Description
float
setTestFraction
The fraction of the input data that is to be used to evaluate the Model.
Parameter
Name
Description
var
float
Returns
Type
Description
$this
getKey
Required. The key is a name of one of the Dataset's data columns.
The key provided must be for a categorical column.
Returns
Type
Description
string
setKey
Required. The key is a name of one of the Dataset's data columns.
The key provided must be for a categorical column.
[[["Easy to understand","easyToUnderstand","thumb-up"],["Solved my problem","solvedMyProblem","thumb-up"],["Other","otherUp","thumb-up"]],[["Hard to understand","hardToUnderstand","thumb-down"],["Incorrect information or sample code","incorrectInformationOrSampleCode","thumb-down"],["Missing the information/samples I need","missingTheInformationSamplesINeed","thumb-down"],["Other","otherDown","thumb-down"]],["Last updated 2025-09-04 UTC."],[],[],null,["# Google Cloud Ai Platform V1 Client - Class StratifiedSplit (1.35.0)\n\nVersion latestkeyboard_arrow_down\n\n- [1.35.0 (latest)](/php/docs/reference/cloud-ai-platform/latest/V1.StratifiedSplit)\n- [1.34.0](/php/docs/reference/cloud-ai-platform/1.34.0/V1.StratifiedSplit)\n- [1.33.0](/php/docs/reference/cloud-ai-platform/1.33.0/V1.StratifiedSplit)\n- [1.32.1](/php/docs/reference/cloud-ai-platform/1.32.1/V1.StratifiedSplit)\n- [1.31.0](/php/docs/reference/cloud-ai-platform/1.31.0/V1.StratifiedSplit)\n- [1.30.0](/php/docs/reference/cloud-ai-platform/1.30.0/V1.StratifiedSplit)\n- [1.26.0](/php/docs/reference/cloud-ai-platform/1.26.0/V1.StratifiedSplit)\n- [1.23.0](/php/docs/reference/cloud-ai-platform/1.23.0/V1.StratifiedSplit)\n- [1.22.0](/php/docs/reference/cloud-ai-platform/1.22.0/V1.StratifiedSplit)\n- [1.21.0](/php/docs/reference/cloud-ai-platform/1.21.0/V1.StratifiedSplit)\n- [1.20.0](/php/docs/reference/cloud-ai-platform/1.20.0/V1.StratifiedSplit)\n- [1.19.0](/php/docs/reference/cloud-ai-platform/1.19.0/V1.StratifiedSplit)\n- [1.18.0](/php/docs/reference/cloud-ai-platform/1.18.0/V1.StratifiedSplit)\n- [1.17.0](/php/docs/reference/cloud-ai-platform/1.17.0/V1.StratifiedSplit)\n- [1.16.0](/php/docs/reference/cloud-ai-platform/1.16.0/V1.StratifiedSplit)\n- [1.15.0](/php/docs/reference/cloud-ai-platform/1.15.0/V1.StratifiedSplit)\n- [1.14.0](/php/docs/reference/cloud-ai-platform/1.14.0/V1.StratifiedSplit)\n- [1.13.1](/php/docs/reference/cloud-ai-platform/1.13.1/V1.StratifiedSplit)\n- [1.12.0](/php/docs/reference/cloud-ai-platform/1.12.0/V1.StratifiedSplit)\n- [1.11.0](/php/docs/reference/cloud-ai-platform/1.11.0/V1.StratifiedSplit)\n- [1.10.0](/php/docs/reference/cloud-ai-platform/1.10.0/V1.StratifiedSplit)\n- [1.9.0](/php/docs/reference/cloud-ai-platform/1.9.0/V1.StratifiedSplit)\n- [1.8.0](/php/docs/reference/cloud-ai-platform/1.8.0/V1.StratifiedSplit)\n- [1.7.0](/php/docs/reference/cloud-ai-platform/1.7.0/V1.StratifiedSplit)\n- [1.6.0](/php/docs/reference/cloud-ai-platform/1.6.0/V1.StratifiedSplit)\n- [1.5.0](/php/docs/reference/cloud-ai-platform/1.5.0/V1.StratifiedSplit)\n- [1.4.0](/php/docs/reference/cloud-ai-platform/1.4.0/V1.StratifiedSplit)\n- [1.3.0](/php/docs/reference/cloud-ai-platform/1.3.0/V1.StratifiedSplit)\n- [1.2.0](/php/docs/reference/cloud-ai-platform/1.2.0/V1.StratifiedSplit)\n- [1.1.0](/php/docs/reference/cloud-ai-platform/1.1.0/V1.StratifiedSplit)\n- [1.0.0](/php/docs/reference/cloud-ai-platform/1.0.0/V1.StratifiedSplit)\n- [0.39.0](/php/docs/reference/cloud-ai-platform/0.39.0/V1.StratifiedSplit)\n- [0.38.0](/php/docs/reference/cloud-ai-platform/0.38.0/V1.StratifiedSplit)\n- [0.37.1](/php/docs/reference/cloud-ai-platform/0.37.1/V1.StratifiedSplit)\n- [0.32.0](/php/docs/reference/cloud-ai-platform/0.32.0/V1.StratifiedSplit)\n- [0.31.0](/php/docs/reference/cloud-ai-platform/0.31.0/V1.StratifiedSplit)\n- [0.30.0](/php/docs/reference/cloud-ai-platform/0.30.0/V1.StratifiedSplit)\n- [0.29.0](/php/docs/reference/cloud-ai-platform/0.29.0/V1.StratifiedSplit)\n- [0.28.0](/php/docs/reference/cloud-ai-platform/0.28.0/V1.StratifiedSplit)\n- [0.27.0](/php/docs/reference/cloud-ai-platform/0.27.0/V1.StratifiedSplit)\n- [0.26.2](/php/docs/reference/cloud-ai-platform/0.26.2/V1.StratifiedSplit)\n- [0.25.0](/php/docs/reference/cloud-ai-platform/0.25.0/V1.StratifiedSplit)\n- [0.24.0](/php/docs/reference/cloud-ai-platform/0.24.0/V1.StratifiedSplit)\n- [0.23.0](/php/docs/reference/cloud-ai-platform/0.23.0/V1.StratifiedSplit)\n- [0.22.0](/php/docs/reference/cloud-ai-platform/0.22.0/V1.StratifiedSplit)\n- [0.21.0](/php/docs/reference/cloud-ai-platform/0.21.0/V1.StratifiedSplit)\n- [0.20.0](/php/docs/reference/cloud-ai-platform/0.20.0/V1.StratifiedSplit)\n- [0.19.0](/php/docs/reference/cloud-ai-platform/0.19.0/V1.StratifiedSplit)\n- [0.18.0](/php/docs/reference/cloud-ai-platform/0.18.0/V1.StratifiedSplit)\n- [0.17.0](/php/docs/reference/cloud-ai-platform/0.17.0/V1.StratifiedSplit)\n- [0.16.0](/php/docs/reference/cloud-ai-platform/0.16.0/V1.StratifiedSplit)\n- [0.15.0](/php/docs/reference/cloud-ai-platform/0.15.0/V1.StratifiedSplit)\n- [0.13.0](/php/docs/reference/cloud-ai-platform/0.13.0/V1.StratifiedSplit)\n- [0.12.0](/php/docs/reference/cloud-ai-platform/0.12.0/V1.StratifiedSplit)\n- [0.11.1](/php/docs/reference/cloud-ai-platform/0.11.1/V1.StratifiedSplit)\n- [0.10.0](/php/docs/reference/cloud-ai-platform/0.10.0/V1.StratifiedSplit) \nReference documentation and code samples for the Google Cloud Ai Platform V1 Client class StratifiedSplit.\n\nAssigns input data to the training, validation, and test sets so that the\ndistribution of values found in the categorical column (as specified by the\n`key` field) is mirrored within each split. The fraction values determine\nthe relative sizes of the splits.\n\nFor example, if the specified column has three values, with 50% of the rows\nhaving value \"A\", 25% value \"B\", and 25% value \"C\", and the split fractions\nare specified as 80/10/10, then the training set will constitute 80% of the\ntraining data, with about 50% of the training set rows having the value \"A\"\nfor the specified column, about 25% having the value \"B\", and about 25%\nhaving the value \"C\".\nOnly the top 500 occurring values are used; any values not in the top\n500 values are randomly assigned to a split. If less than three rows contain\na specific value, those rows are randomly assigned.\nSupported only for tabular Datasets.\n\nGenerated from protobuf message `google.cloud.aiplatform.v1.StratifiedSplit`\n\nNamespace\n---------\n\nGoogle \\\\ Cloud \\\\ AIPlatform \\\\ V1\n\nMethods\n-------\n\n### __construct\n\nConstructor.\n\n### getTrainingFraction\n\nThe fraction of the input data that is to be used to train the Model.\n\n### setTrainingFraction\n\nThe fraction of the input data that is to be used to train the Model.\n\n### getValidationFraction\n\nThe fraction of the input data that is to be used to validate the Model.\n\n### setValidationFraction\n\nThe fraction of the input data that is to be used to validate the Model.\n\n### getTestFraction\n\nThe fraction of the input data that is to be used to evaluate the Model.\n\n### setTestFraction\n\nThe fraction of the input data that is to be used to evaluate the Model.\n\n### getKey\n\nRequired. The key is a name of one of the Dataset's data columns.\n\nThe key provided must be for a categorical column.\n\n### setKey\n\nRequired. The key is a name of one of the Dataset's data columns.\n\nThe key provided must be for a categorical column."]]