package tensorflow_docs.api_report.schema

Get desktop application:
View/edit binary Protocol Buffers messages

• optional google.protobuf.Timestamp timestamp = 1

• optional string date = 2

  Will be filled in via Timestamp().ToJsonString().

• repeated ApiSymbolMetric symbol_metric = 3

  Repeated field to lint each API symbol in the package as traversed by the api_generator.

This message tracks the symbol and the statistics of the lints that are run on the docstring of the symbol.

Used in: ApiReport

• optional string symbol_name = 1

  The full name of the symbol being linted.

• optional ObjectType object_type = 2

  The type of object the symbol is.

• optional ParameterLint parameter_lint = 3

  Statistics after linting the parameters of the docstring of the symbol.

• optional DescriptionLint desc_lint = 4

  Statistics after linting the description of the docstring of the symbol.

• optional UsageExampleLint usage_example_lint = 5

  Statistics after linting the usage examples in the docstring of the symbol.

• optional ReturnLint return_lint = 6

  Statistics after linting the return blocks in the docstring of the symbol.

• optional RaisesLint raises_lint = 7

  Statistics after linting the raises blocks in the docstring of the symbol.

• optional string package_group = 8

  The package group a symbol belongs too. For example: tf.keras.layers.Conv2D -> tf.keras tf.Variable -> tf tf.Variable.assign (method) -> tf tf.data.tensor_slices -> tf.data

This message is used to track the description of a docstring. It tracks the one-line brief and the multi-line description.

Used in: ApiSymbolMetric

• optional float len_brief = 1

  Length of the one line brief description of a docstring. If 0, it means that the one line brief doesn't exist.

• optional float len_long_desc = 3

  Length of the multi-line description of a docstring. If 0, it means that the multi-line description doesn't exist.

This Enum is used to describe the type of object. `api_generator` traverses and extracts docstrings from the objects in the pip package of a library.

Used in: ApiSymbolMetric

• CLASS = 0

• METHOD = 1

• FUNCTION = 2

• MODULE = 3

• TYPE_ALIAS = 4

This message is used to track the parameters in a docstring and lint it. This includes `Args` and `Attributes`.

Used in: ApiSymbolMetric

• optional float num_empty_param_desc_args = 1

  The number of empty parameter descriptions that are present in the args section of a docstring.

• optional float num_args_in_code = 2

  The total number of parameters in the signature of an method/function.

• optional float num_empty_param_desc_attr = 3

  The number of empty parameter descriptions that are present in the attributes section of a docstring.

• optional float total_attr_param = 4

  The total number of parameters in the attributes section of a docstring.

• optional float num_args_in_doc = 5

  The total number of parameters in the args section of a docstring.

This message is used to track if `Raises` is defined in a docstring if `raise` keyword is used in the code.

Used in: ApiSymbolMetric

• optional float num_raises_defined = 1

  Number of raises that are defined in the docstring. If two raises are of the same name (`ValueError`) but the error raised is different then both of them should be documented.

• optional float total_raises_in_code = 2

  Total number of unique raises that are present in the code.

This message is used to track if `Returns` is defined in a docstring if `return` keyword is used in the code.

Used in: ApiSymbolMetric

• optional bool returns_defined = 1

  Whether a `Returns` block exists in the docstring.

This message is used to track the examples in a docstring that demonstrate how to use that API.

Used in: ApiSymbolMetric

• optional float num_doctest = 1

  Number of testable examples in a docstring. go/testable-docstrings describes how to write a testable example.

• optional float num_untested_examples = 2

  Number of untested examples in a docstring. These examples are usually wrapped in triple backticks (```).