deepmd.tf2.common

Attributes

_PACKAGE_ROOT
_DPMODEL_TO_TF2
_AUTO_WRAPPED_CLASSES
_REGISTRATIONS_READY
_REGISTRATIONS_IN_PROGRESS
_REGISTRATION_MODULES
T

Classes

TF2List

Built-in mutable sequence.

Functions

to_tensorflow_array(→ Any)	Convert an object to an ndtensorflow Array.
to_tf_tensor(→ tensorflow.Tensor | None)	Unwrap a TensorFlow-backed Array to a TensorFlow tensor.
wrap_tensor(→ Any | None)	Wrap a TensorFlow tensor as an ndtensorflow Array.
wrap_value(→ Any)	Recursively wrap TensorFlow tensors as ndtensorflow Arrays.
unwrap_value(→ Any)	Recursively unwrap ndtensorflow Arrays for TensorFlow SavedModel returns.
register_dpmodel_mapping(→ None)	Register how to convert a dpmodel object to its tf2 wrapper.
_looks_like_dpmodel_class(→ bool)
_looks_like_dpmodel_object(→ bool)
_looks_like_tf2_object(→ bool)
_ensure_registrations(→ None)
try_convert_module(→ Any | None)	Convert a registered dpmodel object to its tf2 wrapper.
_auto_wrap_native_op(→ Any)
_try_convert_list(→ list[Any])
convert_tf2_value(→ Any)
tf2_setattr(→ Any)
tf2_module(→ type[T])	Wrap a dpmodel subclass as a TensorFlow tf.Module.

Module Contents

deepmd.tf2.common.to_tensorflow_array(array: Any | None) → Any

Convert an object to an ndtensorflow Array.

Parameters:

array

The object to convert.

Returns:

ndtensorflow.Array

The TensorFlow-backed array.

deepmd.tf2.common.to_tf_tensor(array: Any | None) → tf.Tensor | None

Unwrap a TensorFlow-backed Array to a TensorFlow tensor.

deepmd.tf2.common.wrap_tensor(tensor: Any | None) → Any | None

Wrap a TensorFlow tensor as an ndtensorflow Array.

deepmd.tf2.common.wrap_value(value: Any) → Any

Recursively wrap TensorFlow tensors as ndtensorflow Arrays.

deepmd.tf2.common.unwrap_value(value: Any) → Any

Recursively unwrap ndtensorflow Arrays for TensorFlow SavedModel returns.

deepmd.tf2.common._PACKAGE_ROOT

deepmd.tf2.common._DPMODEL_TO_TF2: dict[type[Any], collections.abc.Callable[[Any], Any]]

deepmd.tf2.common._AUTO_WRAPPED_CLASSES: dict[type[deepmd.dpmodel.common.NativeOP], type[Any]]

deepmd.tf2.common._REGISTRATIONS_READY = False

deepmd.tf2.common._REGISTRATIONS_IN_PROGRESS = False

deepmd.tf2.common._REGISTRATION_MODULES

class deepmd.tf2.common.TF2List

Bases: list

Built-in mutable sequence.

If no argument is given, the constructor creates a new empty list. The argument must be an iterable if specified.

append(item: Any) → None

Append object to the end of the list.

extend(items: list[Any]) → None

Extend list by appending elements from the iterable.

insert(index: int, item: Any) → None

Insert object before index.

__setitem__(index: Any, item: Any) → None

Set self[key] to value.

deepmd.tf2.common.register_dpmodel_mapping(dpmodel_cls: type[Any], converter: collections.abc.Callable[[Any], Any]) → None

Register how to convert a dpmodel object to its tf2 wrapper.

deepmd.tf2.common._looks_like_dpmodel_class(cls: type[Any]) → bool

deepmd.tf2.common._looks_like_dpmodel_object(value: Any) → bool

deepmd.tf2.common._looks_like_tf2_object(value: Any) → bool

deepmd.tf2.common._ensure_registrations() → None

deepmd.tf2.common.try_convert_module(value: Any) → Any | None

Convert a registered dpmodel object to its tf2 wrapper.

deepmd.tf2.common._auto_wrap_native_op(value: deepmd.dpmodel.common.NativeOP) → Any

deepmd.tf2.common._try_convert_list(value: list[Any], *, keep_converting: bool = False) → list[Any]

deepmd.tf2.common.convert_tf2_value(value: Any) → Any

deepmd.tf2.common.tf2_setattr(obj: Any, name: str, value: Any) → Any

deepmd.tf2.common.T

deepmd.tf2.common.tf2_module(module: type[T]) → type[T]

Wrap a dpmodel subclass as a TensorFlow tf.Module.