Skip to main content

aws_sdk_sagemaker/types/
_auto_ml_job_config.rs

1// Code generated by software.amazon.smithy.rust.codegen.smithy-rs. DO NOT EDIT.
2
3/// <p>A collection of settings used for an AutoML job.</p>
4#[non_exhaustive]
5#[derive(::std::clone::Clone, ::std::cmp::PartialEq, ::std::fmt::Debug)]
6pub struct AutoMlJobConfig {
7    /// <p>How long an AutoML job is allowed to run, or how many candidates a job is allowed to generate.</p>
8    pub completion_criteria: ::std::option::Option<crate::types::AutoMlJobCompletionCriteria>,
9    /// <p>The security configuration for traffic encryption or Amazon VPC settings.</p>
10    pub security_config: ::std::option::Option<crate::types::AutoMlSecurityConfig>,
11    /// <p>The configuration for generating a candidate for an AutoML job (optional).</p>
12    pub candidate_generation_config: ::std::option::Option<crate::types::AutoMlCandidateGenerationConfig>,
13    /// <p>The configuration for splitting the input training dataset.</p>
14    /// <p>Type: AutoMLDataSplitConfig</p>
15    pub data_split_config: ::std::option::Option<crate::types::AutoMlDataSplitConfig>,
16    /// <p>The method that Autopilot uses to train the data. You can either specify the mode manually or let Autopilot choose for you based on the dataset size by selecting <code>AUTO</code>. In <code>AUTO</code> mode, Autopilot chooses <code>ENSEMBLING</code> for datasets smaller than 100 MB, and <code>HYPERPARAMETER_TUNING</code> for larger ones.</p>
17    /// <p>The <code>ENSEMBLING</code> mode uses a multi-stack ensemble model to predict classification and regression tasks directly from your dataset. This machine learning mode combines several base models to produce an optimal predictive model. It then uses a stacking ensemble method to combine predictions from contributing members. A multi-stack ensemble model can provide better performance over a single model by combining the predictive capabilities of multiple models. See <a href="https://docs.aws.amazon.com/sagemaker/latest/dg/autopilot-model-support-validation.html#autopilot-algorithm-support">Autopilot algorithm support</a> for a list of algorithms supported by <code>ENSEMBLING</code> mode.</p>
18    /// <p>The <code>HYPERPARAMETER_TUNING</code> (HPO) mode uses the best hyperparameters to train the best version of a model. HPO automatically selects an algorithm for the type of problem you want to solve. Then HPO finds the best hyperparameters according to your objective metric. See <a href="https://docs.aws.amazon.com/sagemaker/latest/dg/autopilot-model-support-validation.html#autopilot-algorithm-support">Autopilot algorithm support</a> for a list of algorithms supported by <code>HYPERPARAMETER_TUNING</code> mode.</p>
19    pub mode: ::std::option::Option<crate::types::AutoMlMode>,
20}
21impl AutoMlJobConfig {
22    /// <p>How long an AutoML job is allowed to run, or how many candidates a job is allowed to generate.</p>
23    pub fn completion_criteria(&self) -> ::std::option::Option<&crate::types::AutoMlJobCompletionCriteria> {
24        self.completion_criteria.as_ref()
25    }
26    /// <p>The security configuration for traffic encryption or Amazon VPC settings.</p>
27    pub fn security_config(&self) -> ::std::option::Option<&crate::types::AutoMlSecurityConfig> {
28        self.security_config.as_ref()
29    }
30    /// <p>The configuration for generating a candidate for an AutoML job (optional).</p>
31    pub fn candidate_generation_config(&self) -> ::std::option::Option<&crate::types::AutoMlCandidateGenerationConfig> {
32        self.candidate_generation_config.as_ref()
33    }
34    /// <p>The configuration for splitting the input training dataset.</p>
35    /// <p>Type: AutoMLDataSplitConfig</p>
36    pub fn data_split_config(&self) -> ::std::option::Option<&crate::types::AutoMlDataSplitConfig> {
37        self.data_split_config.as_ref()
38    }
39    /// <p>The method that Autopilot uses to train the data. You can either specify the mode manually or let Autopilot choose for you based on the dataset size by selecting <code>AUTO</code>. In <code>AUTO</code> mode, Autopilot chooses <code>ENSEMBLING</code> for datasets smaller than 100 MB, and <code>HYPERPARAMETER_TUNING</code> for larger ones.</p>
40    /// <p>The <code>ENSEMBLING</code> mode uses a multi-stack ensemble model to predict classification and regression tasks directly from your dataset. This machine learning mode combines several base models to produce an optimal predictive model. It then uses a stacking ensemble method to combine predictions from contributing members. A multi-stack ensemble model can provide better performance over a single model by combining the predictive capabilities of multiple models. See <a href="https://docs.aws.amazon.com/sagemaker/latest/dg/autopilot-model-support-validation.html#autopilot-algorithm-support">Autopilot algorithm support</a> for a list of algorithms supported by <code>ENSEMBLING</code> mode.</p>
41    /// <p>The <code>HYPERPARAMETER_TUNING</code> (HPO) mode uses the best hyperparameters to train the best version of a model. HPO automatically selects an algorithm for the type of problem you want to solve. Then HPO finds the best hyperparameters according to your objective metric. See <a href="https://docs.aws.amazon.com/sagemaker/latest/dg/autopilot-model-support-validation.html#autopilot-algorithm-support">Autopilot algorithm support</a> for a list of algorithms supported by <code>HYPERPARAMETER_TUNING</code> mode.</p>
42    pub fn mode(&self) -> ::std::option::Option<&crate::types::AutoMlMode> {
43        self.mode.as_ref()
44    }
45}
46impl AutoMlJobConfig {
47    /// Creates a new builder-style object to manufacture [`AutoMlJobConfig`](crate::types::AutoMlJobConfig).
48    pub fn builder() -> crate::types::builders::AutoMlJobConfigBuilder {
49        crate::types::builders::AutoMlJobConfigBuilder::default()
50    }
51}
52
53/// A builder for [`AutoMlJobConfig`](crate::types::AutoMlJobConfig).
54#[derive(::std::clone::Clone, ::std::cmp::PartialEq, ::std::default::Default, ::std::fmt::Debug)]
55#[non_exhaustive]
56pub struct AutoMlJobConfigBuilder {
57    pub(crate) completion_criteria: ::std::option::Option<crate::types::AutoMlJobCompletionCriteria>,
58    pub(crate) security_config: ::std::option::Option<crate::types::AutoMlSecurityConfig>,
59    pub(crate) candidate_generation_config: ::std::option::Option<crate::types::AutoMlCandidateGenerationConfig>,
60    pub(crate) data_split_config: ::std::option::Option<crate::types::AutoMlDataSplitConfig>,
61    pub(crate) mode: ::std::option::Option<crate::types::AutoMlMode>,
62}
63impl AutoMlJobConfigBuilder {
64    /// <p>How long an AutoML job is allowed to run, or how many candidates a job is allowed to generate.</p>
65    pub fn completion_criteria(mut self, input: crate::types::AutoMlJobCompletionCriteria) -> Self {
66        self.completion_criteria = ::std::option::Option::Some(input);
67        self
68    }
69    /// <p>How long an AutoML job is allowed to run, or how many candidates a job is allowed to generate.</p>
70    pub fn set_completion_criteria(mut self, input: ::std::option::Option<crate::types::AutoMlJobCompletionCriteria>) -> Self {
71        self.completion_criteria = input;
72        self
73    }
74    /// <p>How long an AutoML job is allowed to run, or how many candidates a job is allowed to generate.</p>
75    pub fn get_completion_criteria(&self) -> &::std::option::Option<crate::types::AutoMlJobCompletionCriteria> {
76        &self.completion_criteria
77    }
78    /// <p>The security configuration for traffic encryption or Amazon VPC settings.</p>
79    pub fn security_config(mut self, input: crate::types::AutoMlSecurityConfig) -> Self {
80        self.security_config = ::std::option::Option::Some(input);
81        self
82    }
83    /// <p>The security configuration for traffic encryption or Amazon VPC settings.</p>
84    pub fn set_security_config(mut self, input: ::std::option::Option<crate::types::AutoMlSecurityConfig>) -> Self {
85        self.security_config = input;
86        self
87    }
88    /// <p>The security configuration for traffic encryption or Amazon VPC settings.</p>
89    pub fn get_security_config(&self) -> &::std::option::Option<crate::types::AutoMlSecurityConfig> {
90        &self.security_config
91    }
92    /// <p>The configuration for generating a candidate for an AutoML job (optional).</p>
93    pub fn candidate_generation_config(mut self, input: crate::types::AutoMlCandidateGenerationConfig) -> Self {
94        self.candidate_generation_config = ::std::option::Option::Some(input);
95        self
96    }
97    /// <p>The configuration for generating a candidate for an AutoML job (optional).</p>
98    pub fn set_candidate_generation_config(mut self, input: ::std::option::Option<crate::types::AutoMlCandidateGenerationConfig>) -> Self {
99        self.candidate_generation_config = input;
100        self
101    }
102    /// <p>The configuration for generating a candidate for an AutoML job (optional).</p>
103    pub fn get_candidate_generation_config(&self) -> &::std::option::Option<crate::types::AutoMlCandidateGenerationConfig> {
104        &self.candidate_generation_config
105    }
106    /// <p>The configuration for splitting the input training dataset.</p>
107    /// <p>Type: AutoMLDataSplitConfig</p>
108    pub fn data_split_config(mut self, input: crate::types::AutoMlDataSplitConfig) -> Self {
109        self.data_split_config = ::std::option::Option::Some(input);
110        self
111    }
112    /// <p>The configuration for splitting the input training dataset.</p>
113    /// <p>Type: AutoMLDataSplitConfig</p>
114    pub fn set_data_split_config(mut self, input: ::std::option::Option<crate::types::AutoMlDataSplitConfig>) -> Self {
115        self.data_split_config = input;
116        self
117    }
118    /// <p>The configuration for splitting the input training dataset.</p>
119    /// <p>Type: AutoMLDataSplitConfig</p>
120    pub fn get_data_split_config(&self) -> &::std::option::Option<crate::types::AutoMlDataSplitConfig> {
121        &self.data_split_config
122    }
123    /// <p>The method that Autopilot uses to train the data. You can either specify the mode manually or let Autopilot choose for you based on the dataset size by selecting <code>AUTO</code>. In <code>AUTO</code> mode, Autopilot chooses <code>ENSEMBLING</code> for datasets smaller than 100 MB, and <code>HYPERPARAMETER_TUNING</code> for larger ones.</p>
124    /// <p>The <code>ENSEMBLING</code> mode uses a multi-stack ensemble model to predict classification and regression tasks directly from your dataset. This machine learning mode combines several base models to produce an optimal predictive model. It then uses a stacking ensemble method to combine predictions from contributing members. A multi-stack ensemble model can provide better performance over a single model by combining the predictive capabilities of multiple models. See <a href="https://docs.aws.amazon.com/sagemaker/latest/dg/autopilot-model-support-validation.html#autopilot-algorithm-support">Autopilot algorithm support</a> for a list of algorithms supported by <code>ENSEMBLING</code> mode.</p>
125    /// <p>The <code>HYPERPARAMETER_TUNING</code> (HPO) mode uses the best hyperparameters to train the best version of a model. HPO automatically selects an algorithm for the type of problem you want to solve. Then HPO finds the best hyperparameters according to your objective metric. See <a href="https://docs.aws.amazon.com/sagemaker/latest/dg/autopilot-model-support-validation.html#autopilot-algorithm-support">Autopilot algorithm support</a> for a list of algorithms supported by <code>HYPERPARAMETER_TUNING</code> mode.</p>
126    pub fn mode(mut self, input: crate::types::AutoMlMode) -> Self {
127        self.mode = ::std::option::Option::Some(input);
128        self
129    }
130    /// <p>The method that Autopilot uses to train the data. You can either specify the mode manually or let Autopilot choose for you based on the dataset size by selecting <code>AUTO</code>. In <code>AUTO</code> mode, Autopilot chooses <code>ENSEMBLING</code> for datasets smaller than 100 MB, and <code>HYPERPARAMETER_TUNING</code> for larger ones.</p>
131    /// <p>The <code>ENSEMBLING</code> mode uses a multi-stack ensemble model to predict classification and regression tasks directly from your dataset. This machine learning mode combines several base models to produce an optimal predictive model. It then uses a stacking ensemble method to combine predictions from contributing members. A multi-stack ensemble model can provide better performance over a single model by combining the predictive capabilities of multiple models. See <a href="https://docs.aws.amazon.com/sagemaker/latest/dg/autopilot-model-support-validation.html#autopilot-algorithm-support">Autopilot algorithm support</a> for a list of algorithms supported by <code>ENSEMBLING</code> mode.</p>
132    /// <p>The <code>HYPERPARAMETER_TUNING</code> (HPO) mode uses the best hyperparameters to train the best version of a model. HPO automatically selects an algorithm for the type of problem you want to solve. Then HPO finds the best hyperparameters according to your objective metric. See <a href="https://docs.aws.amazon.com/sagemaker/latest/dg/autopilot-model-support-validation.html#autopilot-algorithm-support">Autopilot algorithm support</a> for a list of algorithms supported by <code>HYPERPARAMETER_TUNING</code> mode.</p>
133    pub fn set_mode(mut self, input: ::std::option::Option<crate::types::AutoMlMode>) -> Self {
134        self.mode = input;
135        self
136    }
137    /// <p>The method that Autopilot uses to train the data. You can either specify the mode manually or let Autopilot choose for you based on the dataset size by selecting <code>AUTO</code>. In <code>AUTO</code> mode, Autopilot chooses <code>ENSEMBLING</code> for datasets smaller than 100 MB, and <code>HYPERPARAMETER_TUNING</code> for larger ones.</p>
138    /// <p>The <code>ENSEMBLING</code> mode uses a multi-stack ensemble model to predict classification and regression tasks directly from your dataset. This machine learning mode combines several base models to produce an optimal predictive model. It then uses a stacking ensemble method to combine predictions from contributing members. A multi-stack ensemble model can provide better performance over a single model by combining the predictive capabilities of multiple models. See <a href="https://docs.aws.amazon.com/sagemaker/latest/dg/autopilot-model-support-validation.html#autopilot-algorithm-support">Autopilot algorithm support</a> for a list of algorithms supported by <code>ENSEMBLING</code> mode.</p>
139    /// <p>The <code>HYPERPARAMETER_TUNING</code> (HPO) mode uses the best hyperparameters to train the best version of a model. HPO automatically selects an algorithm for the type of problem you want to solve. Then HPO finds the best hyperparameters according to your objective metric. See <a href="https://docs.aws.amazon.com/sagemaker/latest/dg/autopilot-model-support-validation.html#autopilot-algorithm-support">Autopilot algorithm support</a> for a list of algorithms supported by <code>HYPERPARAMETER_TUNING</code> mode.</p>
140    pub fn get_mode(&self) -> &::std::option::Option<crate::types::AutoMlMode> {
141        &self.mode
142    }
143    /// Consumes the builder and constructs a [`AutoMlJobConfig`](crate::types::AutoMlJobConfig).
144    pub fn build(self) -> crate::types::AutoMlJobConfig {
145        crate::types::AutoMlJobConfig {
146            completion_criteria: self.completion_criteria,
147            security_config: self.security_config,
148            candidate_generation_config: self.candidate_generation_config,
149            data_split_config: self.data_split_config,
150            mode: self.mode,
151        }
152    }
153}