Aprenda o que há de mais recente em aprendizado de máquina, IA generativa e muito mais no WiML Symposium 2023
Registre-se
Mantenha tudo organizado com as coleções
Salve e categorize o conteúdo com base nas suas preferências.
#include <ops.h>
O Initializer permite construir um objeto Input a partir de vários tipos de constantes C ++, como constantes primitivas simples e listas de inicializadores aninhados que representam uma matriz multidimensional.
Resumo
Os construtores do inicializador são todos modelos, portanto, os tipos mencionados de constantes C ++ podem ser usados para construir um inicializador . O Initializer armazena o valor com o qual foi construído em um objeto Tensor .
Construtores e Destruidores |
---|
Initializer (const T & v)
Construir a partir de um valor escalar de um tipo aritmético ou um tipo que pode ser convertido em uma string (por exemplo, |
Initializer (const Tensor & t)
|
Initializer (const T & v, const TensorShape & shape)
Construa a partir de um valor escalar e uma forma explícita. |
Initializer (const std::initializer_list< T > & v)
Construa a partir de uma lista inicializadora de escalares (um tensor unidimensional). |
Initializer (const std::initializer_list< T > & v, const TensorShape & shape)
Construa a partir de uma lista inicializadora de escalares e uma forma explícita. |
Initializer (const std::initializer_list< Initializer > & v)
Construa um tensor multidimensional a partir de uma lista de inicializadores aninhados. |
Atributos públicos
Funções públicas
Exceto em caso de indicação contrária, o conteúdo desta página é licenciado de acordo com a Licença de atribuição 4.0 do Creative Commons, e as amostras de código são licenciadas de acordo com a Licença Apache 2.0. Para mais detalhes, consulte as políticas do site do Google Developers. Java é uma marca registrada da Oracle e/ou afiliadas.
Última atualização 2020-04-20 UTC.
[null,null,["Última atualização 2020-04-20 UTC."],[],[],null,["# tensorflow::Input::Initializer Struct Reference\n\ntensorflow::Input::Initializer\n==============================\n\n`#include \u003cops.h\u003e`\n\n[Initializer](/versions/r2.1/api_docs/cc/struct/tensorflow/input/initializer#structtensorflow_1_1_input_1_1_initializer) enables constructing an [Input](/versions/r2.1/api_docs/cc/class/tensorflow/input#classtensorflow_1_1_input) object from various kinds of C++ constants such as simple primitive constants and nested initializer lists representing a multi-dimensional array.\n\nSummary\n-------\n\n[Initializer](/versions/r2.1/api_docs/cc/struct/tensorflow/input/initializer#structtensorflow_1_1_input_1_1_initializer) constructors are all templates, so the aforementioned kinds of C++ constants can be used to construct an [Initializer](/versions/r2.1/api_docs/cc/struct/tensorflow/input/initializer#structtensorflow_1_1_input_1_1_initializer). [Initializer](/versions/r2.1/api_docs/cc/struct/tensorflow/input/initializer#structtensorflow_1_1_input_1_1_initializer) stores the value it got constructed with in a [Tensor](/versions/r2.1/api_docs/cc/class/tensorflow/tensor#classtensorflow_1_1_tensor) object.\n\n| ### Constructors and Destructors ||\n|---|---|\n| [Initializer](#structtensorflow_1_1_input_1_1_initializer_1ade60a4fdcfa9a530604fbf39d3b5be12)`(const T & v)` Construct from a scalar value of an arithmetic type or a type that can be converted to a string (eg. ||\n| [Initializer](#structtensorflow_1_1_input_1_1_initializer_1a9314222b3303dcf97314a4bcbcaa94ad)`(const `[Tensor](/versions/r2.1/api_docs/cc/class/tensorflow/tensor#classtensorflow_1_1_tensor)` & t)` ||\n| [Initializer](#structtensorflow_1_1_input_1_1_initializer_1ab77d0712180868a7311936ca9a034835)`(const T & v, const TensorShape & shape)` Construct from a scalar value and an explicit shape. ||\n| [Initializer](#structtensorflow_1_1_input_1_1_initializer_1a91bd52431434dc5358ae8aa39070fe5f)`(const std::initializer_list\u003c T \u003e & v)` Construct from a initializer list of scalars (a one-dimensional tensor). ||\n| [Initializer](#structtensorflow_1_1_input_1_1_initializer_1a3f572c2835a2310e2d5c28138e69ae76)`(const std::initializer_list\u003c T \u003e & v, const TensorShape & shape)` Construct from a initializer list of scalars and an explicit shape. ||\n| [Initializer](#structtensorflow_1_1_input_1_1_initializer_1a8099f954da757c77ac7d8e1c32df88ce)`(const std::initializer_list\u003c `[Initializer](/versions/r2.1/api_docs/cc/struct/tensorflow/input/initializer#structtensorflow_1_1_input_1_1_initializer)` \u003e & v)` Construct a multi-dimensional tensor from a nested initializer list. ||\n\n| ### Public attributes ||\n|------------------------------------------------------------------------------------------|-----------------------------------------------------------------------------------------|\n| [status](#structtensorflow_1_1_input_1_1_initializer_1af0ab9526e575fd7d4b9d5f7dbabcb7e4) | [Status](/versions/r2.1/api_docs/cc/class/tensorflow/status#classtensorflow_1_1_status) |\n| [tensor](#structtensorflow_1_1_input_1_1_initializer_1a7b520438780dc80f0162a480a3cadb74) | [Tensor](/versions/r2.1/api_docs/cc/class/tensorflow/tensor#classtensorflow_1_1_tensor) |\n\n| ### Public functions ||\n|-----------------------------------------------------------------------------------------------------|---------------|\n| [AsTensorProto](#structtensorflow_1_1_input_1_1_initializer_1a6b1e360b983fec2140b756971fe7699d)`()` | `TensorProto` |\n\nPublic attributes\n-----------------\n\n### status\n\n```text\nStatus tensorflow::Input::Initializer::status\n``` \n\n### tensor\n\n```text\nTensor tensorflow::Input::Initializer::tensor\n``` \n\nPublic functions\n----------------\n\n### AsTensorProto\n\n```text\nTensorProto tensorflow::Input::Initializer::AsTensorProto()\n``` \n\n### Initializer\n\n```gdscript\n tensorflow::Input::Initializer::Initializer(\n const T & v\n)\n``` \nConstruct from a scalar value of an arithmetic type or a type that can be converted to a string (eg.\n\na string literal). \n\n### Initializer\n\n```gdscript\n tensorflow::Input::Initializer::Initializer(\n const Tensor & t\n)\n``` \n\n### Initializer\n\n```gdscript\n tensorflow::Input::Initializer::Initializer(\n const T & v,\n const TensorShape & shape\n)\n``` \nConstruct from a scalar value and an explicit shape. \n\n### Initializer\n\n```gdscript\n tensorflow::Input::Initializer::Initializer(\n const std::initializer_list\u003c T \u003e & v\n)\n``` \nConstruct from a initializer list of scalars (a one-dimensional tensor). \n\n### Initializer\n\n```gdscript\n tensorflow::Input::Initializer::Initializer(\n const std::initializer_list\u003c T \u003e & v,\n const TensorShape & shape\n)\n``` \nConstruct from a initializer list of scalars and an explicit shape. \n\n### Initializer\n\n```gdscript\n tensorflow::Input::Initializer::Initializer(\n const std::initializer_list\u003c Initializer \u003e & v\n)\n``` \nConstruct a multi-dimensional tensor from a nested initializer list.\n\nNote that C++ syntax allows nesting of arbitrarily typed initializer lists, so such invalid initializers cannot be disallowed at compile time. This function performs checks to make sure that the nested initializer list is indeed a valid multi-dimensional tensor."]]