# Generated by default/object.tt package Paws::SageMaker::InputConfig; use Moose; has DataInputConfig => (is => 'ro', isa => 'Str', required => 1); has Framework => (is => 'ro', isa => 'Str', required => 1); has FrameworkVersion => (is => 'ro', isa => 'Str'); has S3Uri => (is => 'ro', isa => 'Str', required => 1); 1; ### main pod documentation begin ### =head1 NAME Paws::SageMaker::InputConfig =head1 USAGE This class represents one of two things: =head3 Arguments in a call to a service Use the attributes of this class as arguments to methods. You shouldn't make instances of this class. Each attribute should be used as a named argument in the calls that expect this type of object. As an example, if Att1 is expected to be a Paws::SageMaker::InputConfig object: $service_obj->Method(Att1 => { DataInputConfig => $value, ..., S3Uri => $value }); =head3 Results returned from an API call Use accessors for each attribute. If Att1 is expected to be an Paws::SageMaker::InputConfig object: $result = $service_obj->Method(...); $result->Att1->DataInputConfig =head1 DESCRIPTION Contains information about the location of input model artifacts, the name and shape of the expected data inputs, and the framework in which the model was trained. =head1 ATTRIBUTES =head2 B DataInputConfig => Str Specifies the name and shape of the expected data inputs for your trained model with a JSON dictionary form. The data inputs are InputConfig$Framework specific. =over =item * C: You must specify the name and shape (NHWC format) of the expected data inputs using a dictionary format for your trained model. The dictionary formats required for the console and CLI are different. =over =item * Examples for one input: =over =item * If using the console, C<{"input":[1,1024,1024,3]}> =item * If using the CLI, C<{\"input\":[1,1024,1024,3]}> =back =item * Examples for two inputs: =over =item * If using the console, C<{"data1": [1,28,28,1], "data2":[1,28,28,1]}> =item * If using the CLI, C<{\"data1\": [1,28,28,1], \"data2\":[1,28,28,1]}> =back =back =item * C: You must specify the name and shape (NCHW format) of expected data inputs using a dictionary format for your trained model. Note that while Keras model artifacts should be uploaded in NHWC (channel-last) format, C should be specified in NCHW (channel-first) format. The dictionary formats required for the console and CLI are different. =over =item * Examples for one input: =over =item * If using the console, C<{"input_1":[1,3,224,224]}> =item * If using the CLI, C<{\"input_1\":[1,3,224,224]}> =back =item * Examples for two inputs: =over =item * If using the console, C<{"input_1": [1,3,224,224], "input_2":[1,3,224,224]}> =item * If using the CLI, C<{\"input_1\": [1,3,224,224], \"input_2\":[1,3,224,224]}> =back =back =item * C: You must specify the name and shape (NCHW format) of the expected data inputs in order using a dictionary format for your trained model. The dictionary formats required for the console and CLI are different. =over =item * Examples for one input: =over =item * If using the console, C<{"data":[1,3,1024,1024]}> =item * If using the CLI, C<{\"data\":[1,3,1024,1024]}> =back =item * Examples for two inputs: =over =item * If using the console, C<{"var1": [1,1,28,28], "var2":[1,1,28,28]}> =item * If using the CLI, C<{\"var1\": [1,1,28,28], \"var2\":[1,1,28,28]}> =back =back =item * C: You can either specify the name and shape (NCHW format) of expected data inputs in order using a dictionary format for your trained model or you can specify the shape only using a list format. The dictionary formats required for the console and CLI are different. The list formats for the console and CLI are the same. =over =item * Examples for one input in dictionary format: =over =item * If using the console, C<{"input0":[1,3,224,224]}> =item * If using the CLI, C<{\"input0\":[1,3,224,224]}> =back =item * Example for one input in list format: C<[[1,3,224,224]]> =item * Examples for two inputs in dictionary format: =over =item * If using the console, C<{"input0":[1,3,224,224], "input1":[1,3,224,224]}> =item * If using the CLI, C<{\"input0\":[1,3,224,224], \"input1\":[1,3,224,224]}> =back =item * Example for two inputs in list format: C<[[1,3,224,224], [1,3,224,224]]> =back =item * C: input data name and shape are not needed. =back C supports the following parameters for C OutputConfig$TargetDevice (ML Model format): =over =item * C: Input shape, for example C<{"input_1": {"shape": [1,224,224,3]}}>. In addition to static input shapes, CoreML converter supports Flexible input shapes: =over =item * Range Dimension. You can use the Range Dimension feature if you know the input shape will be within some specific interval in that dimension, for example: C<{"input_1": {"shape": ["1..10", 224, 224, 3]}}> =item * Enumerated shapes. Sometimes, the models are trained to work only on a select set of inputs. You can enumerate all supported input shapes, for example: C<{"input_1": {"shape": [[1, 224, 224, 3], [1, 160, 160, 3]]}}> =back =item * C: Default input shape. You can set a default shape during conversion for both Range Dimension and Enumerated Shapes. For example C<{"input_1": {"shape": ["1..10", 224, 224, 3], "default_shape": [1, 224, 224, 3]}}> =item * C: Input type. Allowed values: C and C. By default, the converter generates an ML Model with inputs of type Tensor (MultiArray). User can set input type to be Image. Image input type requires additional input parameters such as C and C. =item * C: If the input type is an Image, you need to provide the bias vector. =item * C: If the input type is an Image, you need to provide a scale factor. =back CoreML C parameters can be specified using OutputConfig$CompilerOptions. CoreML converter supports Tensorflow and PyTorch models. CoreML conversion examples: =over =item * Tensor type input: =over =item * C<"DataInputConfig": {"input_1": {"shape": [[1,224,224,3], [1,160,160,3]], "default_shape": [1,224,224,3]}}> =back =item * Tensor type input without input name (PyTorch): =over =item * C<"DataInputConfig": [{"shape": [[1,3,224,224], [1,3,160,160]], "default_shape": [1,3,224,224]}]> =back =item * Image type input: =over =item * C<"DataInputConfig": {"input_1": {"shape": [[1,224,224,3], [1,160,160,3]], "default_shape": [1,224,224,3], "type": "Image", "bias": [-1,-1,-1], "scale": 0.007843137255}}> =item * C<"CompilerOptions": {"class_labels": "imagenet_labels_1000.txt"}> =back =item * Image type input without input name (PyTorch): =over =item * C<"DataInputConfig": [{"shape": [[1,3,224,224], [1,3,160,160]], "default_shape": [1,3,224,224], "type": "Image", "bias": [-1,-1,-1], "scale": 0.007843137255}]> =item * C<"CompilerOptions": {"class_labels": "imagenet_labels_1000.txt"}> =back =back Depending on the model format, C requires the following parameters for C OutputConfig:TargetDevice (https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_OutputConfig.html#sagemaker-Type-OutputConfig-TargetDevice). =over =item * For TensorFlow models saved in the SavedModel format, specify the input names from C and the input model shapes for C. Specify the C in C (https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_OutputConfig.html#sagemaker-Type-OutputConfig-CompilerOptions) if the model does not use TensorFlow's default signature def key. For example: =over =item * C<"DataInputConfig": {"inputs": [1, 224, 224, 3]}> =item * C<"CompilerOptions": {"signature_def_key": "serving_custom"}> =back =item * For TensorFlow models saved as a frozen graph, specify the input tensor names and shapes in C and the output tensor names for C in C (https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_OutputConfig.html#sagemaker-Type-OutputConfig-CompilerOptions). For example: =over =item * C<"DataInputConfig": {"input_tensor:0": [1, 224, 224, 3]}> =item * C<"CompilerOptions": {"output_names": ["output_tensor:0"]}> =back =back =head2 B Framework => Str Identifies the framework in which the model was trained. For example: TENSORFLOW. =head2 FrameworkVersion => Str Specifies the framework version to use. This API field is only supported for PyTorch framework versions C<1.4>, C<1.5>, and C<1.6> for cloud instance target devices: C, C, C, C, C, C, and C. =head2 B S3Uri => Str The S3 path where the model artifacts, which result from model training, are stored. This path must point to a single gzip compressed tar archive (.tar.gz suffix). =head1 SEE ALSO This class forms part of L, describing an object used in L =head1 BUGS and CONTRIBUTIONS The source code is located here: L Please report bugs to: L =cut