basedpyright 1.36.2 → 1.37.0-961d59bacff1958014fe4e9ed2ef3116fe2266e2

package/dist/typeshed-fallback/stubs/tensorflow/tensorflow/core/framework/summary_pb2.pyi

@@ -0,0 +1,368 @@
+"""
+@generated by mypy-protobuf.  Do not edit manually!
+isort:skip_file
+"""
+
+import builtins
+import collections.abc
+import sys
+import typing
+
+import google.protobuf.descriptor
+import google.protobuf.internal.containers
+import google.protobuf.internal.enum_type_wrapper
+import google.protobuf.message
+import tensorflow.core.framework.tensor_pb2
+import tensorflow.tsl.protobuf.histogram_pb2
+
+if sys.version_info >= (3, 10):
+    import typing as typing_extensions
+else:
+    import typing_extensions
+from tensorflow.tsl.protobuf.histogram_pb2 import HistogramProto as HistogramProto
+
+DESCRIPTOR: google.protobuf.descriptor.FileDescriptor
+
+class _DataClass:
+    ValueType = typing.NewType("ValueType", builtins.int)
+    V: typing_extensions.TypeAlias = ValueType
+
+class _DataClassEnumTypeWrapper(google.protobuf.internal.enum_type_wrapper._EnumTypeWrapper[_DataClass.ValueType], builtins.type):
+    DESCRIPTOR: google.protobuf.descriptor.EnumDescriptor
+    DATA_CLASS_UNKNOWN: _DataClass.ValueType  # 0
+    """Unknown data class, used (implicitly) for legacy data. Will not be
+    processed by data ingestion pipelines.
+    """
+    DATA_CLASS_SCALAR: _DataClass.ValueType  # 1
+    """Scalar time series. Each `Value` for the corresponding tag must have
+    `tensor` set to a rank-0 tensor of type `DT_FLOAT` (float32).
+    """
+    DATA_CLASS_TENSOR: _DataClass.ValueType  # 2
+    """Tensor time series. Each `Value` for the corresponding tag must have
+    `tensor` set. The tensor value is arbitrary, but should be small to
+    accommodate direct storage in database backends: an upper bound of a few
+    kilobytes is a reasonable rule of thumb.
+    """
+    DATA_CLASS_BLOB_SEQUENCE: _DataClass.ValueType  # 3
+    """Blob sequence time series. Each `Value` for the corresponding tag must
+    have `tensor` set to a rank-1 tensor of bytestring dtype.
+    """
+
+class DataClass(_DataClass, metaclass=_DataClassEnumTypeWrapper): ...
+
+DATA_CLASS_UNKNOWN: DataClass.ValueType  # 0
+"""Unknown data class, used (implicitly) for legacy data. Will not be
+processed by data ingestion pipelines.
+"""
+DATA_CLASS_SCALAR: DataClass.ValueType  # 1
+"""Scalar time series. Each `Value` for the corresponding tag must have
+`tensor` set to a rank-0 tensor of type `DT_FLOAT` (float32).
+"""
+DATA_CLASS_TENSOR: DataClass.ValueType  # 2
+"""Tensor time series. Each `Value` for the corresponding tag must have
+`tensor` set. The tensor value is arbitrary, but should be small to
+accommodate direct storage in database backends: an upper bound of a few
+kilobytes is a reasonable rule of thumb.
+"""
+DATA_CLASS_BLOB_SEQUENCE: DataClass.ValueType  # 3
+"""Blob sequence time series. Each `Value` for the corresponding tag must
+have `tensor` set to a rank-1 tensor of bytestring dtype.
+"""
+global___DataClass = DataClass
+
+@typing.final
+class SummaryDescription(google.protobuf.message.Message):
+    """Metadata associated with a series of Summary data"""
+
+    DESCRIPTOR: google.protobuf.descriptor.Descriptor
+
+    TYPE_HINT_FIELD_NUMBER: builtins.int
+    type_hint: builtins.str
+    """Hint on how plugins should process the data in this series.
+    Supported values include "scalar", "histogram", "image", "audio"
+    """
+    def __init__(self, *, type_hint: builtins.str | None = ...) -> None: ...
+    def ClearField(self, field_name: typing.Literal["type_hint", b"type_hint"]) -> None: ...
+
+global___SummaryDescription = SummaryDescription
+
+@typing.final
+class SummaryMetadata(google.protobuf.message.Message):
+    """A SummaryMetadata encapsulates information on which plugins are able to make
+    use of a certain summary value.
+    """
+
+    DESCRIPTOR: google.protobuf.descriptor.Descriptor
+
+    @typing.final
+    class PluginData(google.protobuf.message.Message):
+        DESCRIPTOR: google.protobuf.descriptor.Descriptor
+
+        PLUGIN_NAME_FIELD_NUMBER: builtins.int
+        CONTENT_FIELD_NUMBER: builtins.int
+        plugin_name: builtins.str
+        """The name of the plugin this data pertains to."""
+        content: builtins.bytes
+        """The content to store for the plugin. The best practice is for this to be
+        a binary serialized protocol buffer.
+        """
+        def __init__(self, *, plugin_name: builtins.str | None = ..., content: builtins.bytes | None = ...) -> None: ...
+        def ClearField(self, field_name: typing.Literal["content", b"content", "plugin_name", b"plugin_name"]) -> None: ...
+
+    PLUGIN_DATA_FIELD_NUMBER: builtins.int
+    DISPLAY_NAME_FIELD_NUMBER: builtins.int
+    SUMMARY_DESCRIPTION_FIELD_NUMBER: builtins.int
+    DATA_CLASS_FIELD_NUMBER: builtins.int
+    display_name: builtins.str
+    """Display name for viewing in TensorBoard."""
+    summary_description: builtins.str
+    """Longform readable description of the summary sequence. Markdown supported."""
+    data_class: global___DataClass.ValueType
+    """Class of data stored in this time series. Required for compatibility with
+    TensorBoard's generic data facilities (`DataProvider`, et al.). This value
+    imposes constraints on the dtype and shape of the corresponding tensor
+    values. See `DataClass` docs for details.
+    """
+    @property
+    def plugin_data(self) -> global___SummaryMetadata.PluginData:
+        """Data that associates a summary with a certain plugin."""
+
+    def __init__(
+        self,
+        *,
+        plugin_data: global___SummaryMetadata.PluginData | None = ...,
+        display_name: builtins.str | None = ...,
+        summary_description: builtins.str | None = ...,
+        data_class: global___DataClass.ValueType | None = ...,
+    ) -> None: ...
+    def HasField(self, field_name: typing.Literal["plugin_data", b"plugin_data"]) -> builtins.bool: ...
+    def ClearField(
+        self,
+        field_name: typing.Literal[
+            "data_class",
+            b"data_class",
+            "display_name",
+            b"display_name",
+            "plugin_data",
+            b"plugin_data",
+            "summary_description",
+            b"summary_description",
+        ],
+    ) -> None: ...
+
+global___SummaryMetadata = SummaryMetadata
+
+@typing.final
+class Summary(google.protobuf.message.Message):
+    """A Summary is a set of named values to be displayed by the
+    visualizer.
+
+    Summaries are produced regularly during training, as controlled by
+    the "summary_interval_secs" attribute of the training operation.
+    Summaries are also produced at the end of an evaluation.
+    """
+
+    DESCRIPTOR: google.protobuf.descriptor.Descriptor
+
+    @typing.final
+    class Image(google.protobuf.message.Message):
+        DESCRIPTOR: google.protobuf.descriptor.Descriptor
+
+        HEIGHT_FIELD_NUMBER: builtins.int
+        WIDTH_FIELD_NUMBER: builtins.int
+        COLORSPACE_FIELD_NUMBER: builtins.int
+        ENCODED_IMAGE_STRING_FIELD_NUMBER: builtins.int
+        height: builtins.int
+        """Dimensions of the image."""
+        width: builtins.int
+        colorspace: builtins.int
+        """Valid colorspace values are
+          1 - grayscale
+          2 - grayscale + alpha
+          3 - RGB
+          4 - RGBA
+          5 - DIGITAL_YUV
+          6 - BGRA
+        """
+        encoded_image_string: builtins.bytes
+        """Image data in encoded format.  All image formats supported by
+        image_codec::CoderUtil can be stored here.
+        """
+        def __init__(
+            self,
+            *,
+            height: builtins.int | None = ...,
+            width: builtins.int | None = ...,
+            colorspace: builtins.int | None = ...,
+            encoded_image_string: builtins.bytes | None = ...,
+        ) -> None: ...
+        def ClearField(
+            self,
+            field_name: typing.Literal[
+                "colorspace",
+                b"colorspace",
+                "encoded_image_string",
+                b"encoded_image_string",
+                "height",
+                b"height",
+                "width",
+                b"width",
+            ],
+        ) -> None: ...
+
+    @typing.final
+    class Audio(google.protobuf.message.Message):
+        DESCRIPTOR: google.protobuf.descriptor.Descriptor
+
+        SAMPLE_RATE_FIELD_NUMBER: builtins.int
+        NUM_CHANNELS_FIELD_NUMBER: builtins.int
+        LENGTH_FRAMES_FIELD_NUMBER: builtins.int
+        ENCODED_AUDIO_STRING_FIELD_NUMBER: builtins.int
+        CONTENT_TYPE_FIELD_NUMBER: builtins.int
+        sample_rate: builtins.float
+        """Sample rate of the audio in Hz."""
+        num_channels: builtins.int
+        """Number of channels of audio."""
+        length_frames: builtins.int
+        """Length of the audio in frames (samples per channel)."""
+        encoded_audio_string: builtins.bytes
+        """Encoded audio data and its associated RFC 2045 content type (e.g.
+        "audio/wav").
+        """
+        content_type: builtins.str
+        def __init__(
+            self,
+            *,
+            sample_rate: builtins.float | None = ...,
+            num_channels: builtins.int | None = ...,
+            length_frames: builtins.int | None = ...,
+            encoded_audio_string: builtins.bytes | None = ...,
+            content_type: builtins.str | None = ...,
+        ) -> None: ...
+        def ClearField(
+            self,
+            field_name: typing.Literal[
+                "content_type",
+                b"content_type",
+                "encoded_audio_string",
+                b"encoded_audio_string",
+                "length_frames",
+                b"length_frames",
+                "num_channels",
+                b"num_channels",
+                "sample_rate",
+                b"sample_rate",
+            ],
+        ) -> None: ...
+
+    @typing.final
+    class Value(google.protobuf.message.Message):
+        DESCRIPTOR: google.protobuf.descriptor.Descriptor
+
+        NODE_NAME_FIELD_NUMBER: builtins.int
+        TAG_FIELD_NUMBER: builtins.int
+        METADATA_FIELD_NUMBER: builtins.int
+        SIMPLE_VALUE_FIELD_NUMBER: builtins.int
+        OBSOLETE_OLD_STYLE_HISTOGRAM_FIELD_NUMBER: builtins.int
+        IMAGE_FIELD_NUMBER: builtins.int
+        HISTO_FIELD_NUMBER: builtins.int
+        AUDIO_FIELD_NUMBER: builtins.int
+        TENSOR_FIELD_NUMBER: builtins.int
+        node_name: builtins.str
+        """This field is deprecated and will not be set."""
+        tag: builtins.str
+        """Tag name for the data. Used by TensorBoard plugins to organize data. Tags
+        are often organized by scope (which contains slashes to convey
+        hierarchy). For example: foo/bar/0
+        """
+        simple_value: builtins.float
+        obsolete_old_style_histogram: builtins.bytes
+        @property
+        def metadata(self) -> global___SummaryMetadata:
+            """Contains metadata on the summary value such as which plugins may use it.
+            Take note that many summary values may lack a metadata field. This is
+            because the FileWriter only keeps a metadata object on the first summary
+            value with a certain tag for each tag. TensorBoard then remembers which
+            tags are associated with which plugins. This saves space.
+            """
+
+        @property
+        def image(self) -> global___Summary.Image: ...
+        @property
+        def histo(self) -> tensorflow.tsl.protobuf.histogram_pb2.HistogramProto: ...
+        @property
+        def audio(self) -> global___Summary.Audio: ...
+        @property
+        def tensor(self) -> tensorflow.core.framework.tensor_pb2.TensorProto: ...
+        def __init__(
+            self,
+            *,
+            node_name: builtins.str | None = ...,
+            tag: builtins.str | None = ...,
+            metadata: global___SummaryMetadata | None = ...,
+            simple_value: builtins.float | None = ...,
+            obsolete_old_style_histogram: builtins.bytes | None = ...,
+            image: global___Summary.Image | None = ...,
+            histo: tensorflow.tsl.protobuf.histogram_pb2.HistogramProto | None = ...,
+            audio: global___Summary.Audio | None = ...,
+            tensor: tensorflow.core.framework.tensor_pb2.TensorProto | None = ...,
+        ) -> None: ...
+        def HasField(
+            self,
+            field_name: typing.Literal[
+                "audio",
+                b"audio",
+                "histo",
+                b"histo",
+                "image",
+                b"image",
+                "metadata",
+                b"metadata",
+                "obsolete_old_style_histogram",
+                b"obsolete_old_style_histogram",
+                "simple_value",
+                b"simple_value",
+                "tensor",
+                b"tensor",
+                "value",
+                b"value",
+            ],
+        ) -> builtins.bool: ...
+        def ClearField(
+            self,
+            field_name: typing.Literal[
+                "audio",
+                b"audio",
+                "histo",
+                b"histo",
+                "image",
+                b"image",
+                "metadata",
+                b"metadata",
+                "node_name",
+                b"node_name",
+                "obsolete_old_style_histogram",
+                b"obsolete_old_style_histogram",
+                "simple_value",
+                b"simple_value",
+                "tag",
+                b"tag",
+                "tensor",
+                b"tensor",
+                "value",
+                b"value",
+            ],
+        ) -> None: ...
+        def WhichOneof(
+            self, oneof_group: typing.Literal["value", b"value"]
+        ) -> typing.Literal["simple_value", "obsolete_old_style_histogram", "image", "histo", "audio", "tensor"] | None: ...
+
+    VALUE_FIELD_NUMBER: builtins.int
+    @property
+    def value(self) -> google.protobuf.internal.containers.RepeatedCompositeFieldContainer[global___Summary.Value]:
+        """Set of values for the summary."""
+
+    def __init__(self, *, value: collections.abc.Iterable[global___Summary.Value] | None = ...) -> None: ...
+    def ClearField(self, field_name: typing.Literal["value", b"value"]) -> None: ...
+
+global___Summary = Summary

package/dist/typeshed-fallback/stubs/tensorflow/tensorflow/core/framework/tensor_description_pb2.pyi

@@ -0,0 +1,49 @@
+"""
+@generated by mypy-protobuf.  Do not edit manually!
+isort:skip_file
+"""
+
+import builtins
+import typing
+
+import google.protobuf.descriptor
+import google.protobuf.message
+import tensorflow.core.framework.allocation_description_pb2
+import tensorflow.core.framework.tensor_shape_pb2
+import tensorflow.core.framework.types_pb2
+
+DESCRIPTOR: google.protobuf.descriptor.FileDescriptor
+
+@typing.final
+class TensorDescription(google.protobuf.message.Message):
+    DESCRIPTOR: google.protobuf.descriptor.Descriptor
+
+    DTYPE_FIELD_NUMBER: builtins.int
+    SHAPE_FIELD_NUMBER: builtins.int
+    ALLOCATION_DESCRIPTION_FIELD_NUMBER: builtins.int
+    dtype: tensorflow.core.framework.types_pb2.DataType.ValueType
+    """Data type of tensor elements"""
+    @property
+    def shape(self) -> tensorflow.core.framework.tensor_shape_pb2.TensorShapeProto:
+        """Shape of the tensor."""
+
+    @property
+    def allocation_description(self) -> tensorflow.core.framework.allocation_description_pb2.AllocationDescription:
+        """Information about the size and allocator used for the data"""
+
+    def __init__(
+        self,
+        *,
+        dtype: tensorflow.core.framework.types_pb2.DataType.ValueType | None = ...,
+        shape: tensorflow.core.framework.tensor_shape_pb2.TensorShapeProto | None = ...,
+        allocation_description: tensorflow.core.framework.allocation_description_pb2.AllocationDescription | None = ...,
+    ) -> None: ...
+    def HasField(
+        self, field_name: typing.Literal["allocation_description", b"allocation_description", "shape", b"shape"]
+    ) -> builtins.bool: ...
+    def ClearField(
+        self,
+        field_name: typing.Literal["allocation_description", b"allocation_description", "dtype", b"dtype", "shape", b"shape"],
+    ) -> None: ...
+
+global___TensorDescription = TensorDescription

package/dist/typeshed-fallback/stubs/tensorflow/tensorflow/core/framework/tensor_pb2.pyi

@@ -0,0 +1,235 @@
+"""
+@generated by mypy-protobuf.  Do not edit manually!
+isort:skip_file
+"""
+
+import builtins
+import collections.abc
+import typing
+
+import google.protobuf.descriptor
+import google.protobuf.internal.containers
+import google.protobuf.message
+import tensorflow.core.framework.resource_handle_pb2
+import tensorflow.core.framework.tensor_shape_pb2
+import tensorflow.core.framework.types_pb2
+
+DESCRIPTOR: google.protobuf.descriptor.FileDescriptor
+
+@typing.final
+class TensorProto(google.protobuf.message.Message):
+    """Protocol buffer representing a tensor."""
+
+    DESCRIPTOR: google.protobuf.descriptor.Descriptor
+
+    DTYPE_FIELD_NUMBER: builtins.int
+    TENSOR_SHAPE_FIELD_NUMBER: builtins.int
+    VERSION_NUMBER_FIELD_NUMBER: builtins.int
+    TENSOR_CONTENT_FIELD_NUMBER: builtins.int
+    HALF_VAL_FIELD_NUMBER: builtins.int
+    FLOAT_VAL_FIELD_NUMBER: builtins.int
+    DOUBLE_VAL_FIELD_NUMBER: builtins.int
+    INT_VAL_FIELD_NUMBER: builtins.int
+    STRING_VAL_FIELD_NUMBER: builtins.int
+    SCOMPLEX_VAL_FIELD_NUMBER: builtins.int
+    INT64_VAL_FIELD_NUMBER: builtins.int
+    BOOL_VAL_FIELD_NUMBER: builtins.int
+    DCOMPLEX_VAL_FIELD_NUMBER: builtins.int
+    RESOURCE_HANDLE_VAL_FIELD_NUMBER: builtins.int
+    VARIANT_VAL_FIELD_NUMBER: builtins.int
+    UINT32_VAL_FIELD_NUMBER: builtins.int
+    UINT64_VAL_FIELD_NUMBER: builtins.int
+    FLOAT8_VAL_FIELD_NUMBER: builtins.int
+    dtype: tensorflow.core.framework.types_pb2.DataType.ValueType
+    """Data type of the tensor."""
+    version_number: builtins.int
+    """Only one of the representations below is set, one of "tensor_contents" and
+    the "xxx_val" attributes.  We are not using oneof because as oneofs cannot
+    contain repeated fields it would require another extra set of messages.
+
+    Version number.
+
+    In version 0, if the "repeated xxx" representations contain only one
+    element, that element is repeated to fill the shape.  This makes it easy
+    to represent a constant Tensor with a single value.
+    """
+    tensor_content: builtins.bytes
+    """Serialized raw tensor content from either Tensor::AsProtoTensorContent or
+    memcpy in tensorflow::grpc::EncodeTensorToByteBuffer. This representation
+    can be used for all tensor types. The purpose of this representation is to
+    reduce serialization overhead during RPC call by avoiding serialization of
+    many repeated small items.
+    """
+    float8_val: builtins.bytes
+    """DT_FLOAT8_*, use variable-sized set of bytes
+    (i.e. the equivalent of repeated uint8, if such a thing existed).
+    """
+    @property
+    def tensor_shape(self) -> tensorflow.core.framework.tensor_shape_pb2.TensorShapeProto:
+        """Shape of the tensor.  TODO(touts): sort out the 0-rank issues."""
+
+    @property
+    def half_val(self) -> google.protobuf.internal.containers.RepeatedScalarFieldContainer[builtins.int]:
+        """Type specific representations that make it easy to create tensor protos in
+        all languages.  Only the representation corresponding to "dtype" can
+        be set.  The values hold the flattened representation of the tensor in
+        row major order.
+
+        DT_HALF, DT_BFLOAT16. Note that since protobuf has no int16 type, we'll
+        have some pointless zero padding for each value here.
+        """
+
+    @property
+    def float_val(self) -> google.protobuf.internal.containers.RepeatedScalarFieldContainer[builtins.float]:
+        """DT_FLOAT."""
+
+    @property
+    def double_val(self) -> google.protobuf.internal.containers.RepeatedScalarFieldContainer[builtins.float]:
+        """DT_DOUBLE."""
+
+    @property
+    def int_val(self) -> google.protobuf.internal.containers.RepeatedScalarFieldContainer[builtins.int]:
+        """DT_INT32, DT_INT16, DT_UINT16, DT_INT8, DT_UINT8."""
+
+    @property
+    def string_val(self) -> google.protobuf.internal.containers.RepeatedScalarFieldContainer[builtins.bytes]:
+        """DT_STRING"""
+
+    @property
+    def scomplex_val(self) -> google.protobuf.internal.containers.RepeatedScalarFieldContainer[builtins.float]:
+        """DT_COMPLEX64. scomplex_val(2*i) and scomplex_val(2*i+1) are real
+        and imaginary parts of i-th single precision complex.
+        """
+
+    @property
+    def int64_val(self) -> google.protobuf.internal.containers.RepeatedScalarFieldContainer[builtins.int]:
+        """DT_INT64"""
+
+    @property
+    def bool_val(self) -> google.protobuf.internal.containers.RepeatedScalarFieldContainer[builtins.bool]:
+        """DT_BOOL"""
+
+    @property
+    def dcomplex_val(self) -> google.protobuf.internal.containers.RepeatedScalarFieldContainer[builtins.float]:
+        """DT_COMPLEX128. dcomplex_val(2*i) and dcomplex_val(2*i+1) are real
+        and imaginary parts of i-th double precision complex.
+        """
+
+    @property
+    def resource_handle_val(
+        self,
+    ) -> google.protobuf.internal.containers.RepeatedCompositeFieldContainer[
+        tensorflow.core.framework.resource_handle_pb2.ResourceHandleProto
+    ]:
+        """DT_RESOURCE"""
+
+    @property
+    def variant_val(self) -> google.protobuf.internal.containers.RepeatedCompositeFieldContainer[global___VariantTensorDataProto]:
+        """DT_VARIANT"""
+
+    @property
+    def uint32_val(self) -> google.protobuf.internal.containers.RepeatedScalarFieldContainer[builtins.int]:
+        """DT_UINT32"""
+
+    @property
+    def uint64_val(self) -> google.protobuf.internal.containers.RepeatedScalarFieldContainer[builtins.int]:
+        """DT_UINT64"""
+
+    def __init__(
+        self,
+        *,
+        dtype: tensorflow.core.framework.types_pb2.DataType.ValueType | None = ...,
+        tensor_shape: tensorflow.core.framework.tensor_shape_pb2.TensorShapeProto | None = ...,
+        version_number: builtins.int | None = ...,
+        tensor_content: builtins.bytes | None = ...,
+        half_val: collections.abc.Iterable[builtins.int] | None = ...,
+        float_val: collections.abc.Iterable[builtins.float] | None = ...,
+        double_val: collections.abc.Iterable[builtins.float] | None = ...,
+        int_val: collections.abc.Iterable[builtins.int] | None = ...,
+        string_val: collections.abc.Iterable[builtins.bytes] | None = ...,
+        scomplex_val: collections.abc.Iterable[builtins.float] | None = ...,
+        int64_val: collections.abc.Iterable[builtins.int] | None = ...,
+        bool_val: collections.abc.Iterable[builtins.bool] | None = ...,
+        dcomplex_val: collections.abc.Iterable[builtins.float] | None = ...,
+        resource_handle_val: (
+            collections.abc.Iterable[tensorflow.core.framework.resource_handle_pb2.ResourceHandleProto] | None
+        ) = ...,
+        variant_val: collections.abc.Iterable[global___VariantTensorDataProto] | None = ...,
+        uint32_val: collections.abc.Iterable[builtins.int] | None = ...,
+        uint64_val: collections.abc.Iterable[builtins.int] | None = ...,
+        float8_val: builtins.bytes | None = ...,
+    ) -> None: ...
+    def HasField(self, field_name: typing.Literal["tensor_shape", b"tensor_shape"]) -> builtins.bool: ...
+    def ClearField(
+        self,
+        field_name: typing.Literal[
+            "bool_val",
+            b"bool_val",
+            "dcomplex_val",
+            b"dcomplex_val",
+            "double_val",
+            b"double_val",
+            "dtype",
+            b"dtype",
+            "float8_val",
+            b"float8_val",
+            "float_val",
+            b"float_val",
+            "half_val",
+            b"half_val",
+            "int64_val",
+            b"int64_val",
+            "int_val",
+            b"int_val",
+            "resource_handle_val",
+            b"resource_handle_val",
+            "scomplex_val",
+            b"scomplex_val",
+            "string_val",
+            b"string_val",
+            "tensor_content",
+            b"tensor_content",
+            "tensor_shape",
+            b"tensor_shape",
+            "uint32_val",
+            b"uint32_val",
+            "uint64_val",
+            b"uint64_val",
+            "variant_val",
+            b"variant_val",
+            "version_number",
+            b"version_number",
+        ],
+    ) -> None: ...
+
+global___TensorProto = TensorProto
+
+@typing.final
+class VariantTensorDataProto(google.protobuf.message.Message):
+    """Protocol buffer representing the serialization format of DT_VARIANT tensors."""
+
+    DESCRIPTOR: google.protobuf.descriptor.Descriptor
+
+    TYPE_NAME_FIELD_NUMBER: builtins.int
+    METADATA_FIELD_NUMBER: builtins.int
+    TENSORS_FIELD_NUMBER: builtins.int
+    type_name: builtins.str
+    """Name of the type of objects being serialized."""
+    metadata: builtins.bytes
+    """Portions of the object that are not Tensors."""
+    @property
+    def tensors(self) -> google.protobuf.internal.containers.RepeatedCompositeFieldContainer[global___TensorProto]:
+        """Tensors contained within objects being serialized."""
+
+    def __init__(
+        self,
+        *,
+        type_name: builtins.str | None = ...,
+        metadata: builtins.bytes | None = ...,
+        tensors: collections.abc.Iterable[global___TensorProto] | None = ...,
+    ) -> None: ...
+    def ClearField(
+        self, field_name: typing.Literal["metadata", b"metadata", "tensors", b"tensors", "type_name", b"type_name"]
+    ) -> None: ...
+
+global___VariantTensorDataProto = VariantTensorDataProto