<br />

<button value="stable" default="">See Stable</button> <button value="nightly">See Nightly</button>

<br />

Public API for tf._api.v2.experimental namespace

## Modules

[`extension_type`](https://www.tensorflow.org/api_docs/python/tf/compat/v1/experimental/extension_type) module: Public API for tf._api.v2.experimental.extension_type namespace

## Classes

[`class BatchableExtensionType`](https://www.tensorflow.org/api_docs/python/tf/experimental/BatchableExtensionType): An ExtensionType that can be batched and unbatched.

[`class DynamicRaggedShape`](https://www.tensorflow.org/api_docs/python/tf/experimental/DynamicRaggedShape): The shape of a ragged or dense tensor.

[`class ExtensionType`](https://www.tensorflow.org/api_docs/python/tf/experimental/ExtensionType): Base class for TensorFlow `ExtensionType` classes.

[`class ExtensionTypeBatchEncoder`](https://www.tensorflow.org/api_docs/python/tf/experimental/ExtensionTypeBatchEncoder): Class used to encode and decode extension type values for batching.

[`class ExtensionTypeSpec`](https://www.tensorflow.org/api_docs/python/tf/experimental/ExtensionTypeSpec): Base class for tf.ExtensionType TypeSpec.

[`class Optional`](https://www.tensorflow.org/api_docs/python/tf/experimental/Optional): Represents a value that may or may not be present.

[`class RowPartition`](https://www.tensorflow.org/api_docs/python/tf/experimental/RowPartition): Partitioning of a sequence of values into contiguous subsequences ("rows").

[`class StructuredTensor`](https://www.tensorflow.org/api_docs/python/tf/experimental/StructuredTensor): A multidimensional collection of structures with the same schema.

## Functions

[`async_clear_error(...)`](https://www.tensorflow.org/api_docs/python/tf/experimental/async_clear_error): Clear pending operations and error statuses in async execution.

[`async_scope(...)`](https://www.tensorflow.org/api_docs/python/tf/experimental/async_scope): Context manager for grouping async operations.

[`dispatch_for_api(...)`](https://www.tensorflow.org/api_docs/python/tf/experimental/dispatch_for_api): Decorator that overrides the default implementation for a TensorFlow API.

[`dispatch_for_binary_elementwise_apis(...)`](https://www.tensorflow.org/api_docs/python/tf/experimental/dispatch_for_binary_elementwise_apis): Decorator to override default implementation for binary elementwise APIs.

[`dispatch_for_binary_elementwise_assert_apis(...)`](https://www.tensorflow.org/api_docs/python/tf/experimental/dispatch_for_binary_elementwise_assert_apis): Decorator to override default implementation for binary elementwise assert APIs.

[`dispatch_for_unary_elementwise_apis(...)`](https://www.tensorflow.org/api_docs/python/tf/experimental/dispatch_for_unary_elementwise_apis): Decorator to override default implementation for unary elementwise APIs.

[`enable_strict_mode(...)`](https://www.tensorflow.org/api_docs/python/tf/experimental/enable_strict_mode): If called, enables strict mode for all behaviors.

[`function_executor_type(...)`](https://www.tensorflow.org/api_docs/python/tf/experimental/function_executor_type): Context manager for setting the executor of eager defined functions.

[`output_all_intermediates(...)`](https://www.tensorflow.org/api_docs/python/tf/compat/v1/experimental/output_all_intermediates): Whether to output all intermediates from functional control flow ops.

[`register_filesystem_plugin(...)`](https://www.tensorflow.org/api_docs/python/tf/experimental/register_filesystem_plugin): Loads a TensorFlow FileSystem plugin.

[`unregister_dispatch_for(...)`](https://www.tensorflow.org/api_docs/python/tf/experimental/unregister_dispatch_for): Unregisters a function that was registered with `@dispatch_for_*`.

<br />

<br />

<br />

| ## Other Members ||
|---|---|
| float4_e2m1fn | Instance of [`tf.dtypes.DType`](https://www.tensorflow.org/api_docs/python/tf/dtypes/DType) <br /> 4-bit float with 2 exponent bits and 1 mantissa bits, with extended finite range. This type has no representation for both inf and NaN. |
| float8_e4m3b11fnuz | Instance of [`tf.dtypes.DType`](https://www.tensorflow.org/api_docs/python/tf/dtypes/DType) <br /> 8-bit float with 4 exponent bits and 3 mantissa bits, with extended finite range and 11 bits of bias. This type has no representation for inf, and only one NaN value: 0x80. |
| float8_e4m3fn | Instance of [`tf.dtypes.DType`](https://www.tensorflow.org/api_docs/python/tf/dtypes/DType) <br /> 8-bit float with 4 exponent bits and 3 mantissa bits, with extended finite range. This type has no representation for inf, and only two NaN values: 0xFF for negative NaN, and 0x7F for positive NaN. |
| float8_e4m3fnuz | Instance of [`tf.dtypes.DType`](https://www.tensorflow.org/api_docs/python/tf/dtypes/DType) <br /> 8-bit float with 4 exponent bits and 3 mantissa bits, with extended finite range. This type has no representation for inf, and only one NaN value: 0x80. |
| float8_e5m2 | Instance of [`tf.dtypes.DType`](https://www.tensorflow.org/api_docs/python/tf/dtypes/DType) <br /> 8-bit float with 5 exponent bits and 2 mantissa bits. |
| float8_e5m2fnuz | Instance of [`tf.dtypes.DType`](https://www.tensorflow.org/api_docs/python/tf/dtypes/DType) <br /> 8-bit float with 5 exponent bits and 2 mantissa bits, with extended finite range. This type has no representation for inf, and only one NaN value: 0x80. |
| int2 | Instance of [`tf.dtypes.DType`](https://www.tensorflow.org/api_docs/python/tf/dtypes/DType) <br /> Signed 2-bit integer. |
| int4 | Instance of [`tf.dtypes.DType`](https://www.tensorflow.org/api_docs/python/tf/dtypes/DType) <br /> Signed 4-bit integer. |
| uint2 | Instance of [`tf.dtypes.DType`](https://www.tensorflow.org/api_docs/python/tf/dtypes/DType) <br /> Unsigned 2-bit integer. |
| uint4 | Instance of [`tf.dtypes.DType`](https://www.tensorflow.org/api_docs/python/tf/dtypes/DType) <br /> Unsigned 4-bit integer. |

<br />