Coverage for /pythoncovmergedfiles/medio/medio/usr/local/lib/python3.8/site-packages/tensorflow/python/ops/gen_encode_proto_ops.py: 25%
83 statements
« prev ^ index » next coverage.py v7.4.0, created at 2024-01-03 07:57 +0000
« prev ^ index » next coverage.py v7.4.0, created at 2024-01-03 07:57 +0000
1"""Python wrappers around TensorFlow ops.
3This file is MACHINE GENERATED! Do not edit.
4"""
6import collections
8from tensorflow.python import pywrap_tfe as pywrap_tfe
9from tensorflow.python.eager import context as _context
10from tensorflow.python.eager import core as _core
11from tensorflow.python.eager import execute as _execute
12from tensorflow.python.framework import dtypes as _dtypes
13from tensorflow.security.fuzzing.py import annotation_types as _atypes
15from tensorflow.python.framework import op_def_registry as _op_def_registry
16from tensorflow.python.framework import ops as _ops
17from tensorflow.python.framework import op_def_library as _op_def_library
18from tensorflow.python.util.deprecation import deprecated_endpoints
19from tensorflow.python.util import dispatch as _dispatch
20from tensorflow.python.util.tf_export import tf_export
22from typing import TypeVar
24@_dispatch.add_fallback_dispatch_list
25@_dispatch.add_type_based_api_dispatcher
26@tf_export('io.encode_proto')
27def encode_proto(sizes, values, field_names, message_type, descriptor_source="local://", name=None):
28 r"""The op serializes protobuf messages provided in the input tensors.
30 The types of the tensors in `values` must match the schema for the fields
31 specified in `field_names`. All the tensors in `values` must have a common
32 shape prefix, *batch_shape*.
34 The `sizes` tensor specifies repeat counts for each field. The repeat count
35 (last dimension) of a each tensor in `values` must be greater than or equal
36 to corresponding repeat count in `sizes`.
38 A `message_type` name must be provided to give context for the field names.
39 The actual message descriptor can be looked up either in the linked-in
40 descriptor pool or a filename provided by the caller using the
41 `descriptor_source` attribute.
43 For the most part, the mapping between Proto field types and TensorFlow dtypes
44 is straightforward. However, there are a few special cases:
46 - A proto field that contains a submessage or group can only be converted
47 to `DT_STRING` (the serialized submessage). This is to reduce the complexity
48 of the API. The resulting string can be used as input to another instance of
49 the decode_proto op.
51 - TensorFlow lacks support for unsigned integers. The ops represent uint64
52 types as a `DT_INT64` with the same twos-complement bit pattern (the obvious
53 way). Unsigned int32 values can be represented exactly by specifying type
54 `DT_INT64`, or using twos-complement if the caller specifies `DT_INT32` in
55 the `output_types` attribute.
57 The `descriptor_source` attribute selects the source of protocol
58 descriptors to consult when looking up `message_type`. This may be:
60 - An empty string or "local://", in which case protocol descriptors are
61 created for C++ (not Python) proto definitions linked to the binary.
63 - A file, in which case protocol descriptors are created from the file,
64 which is expected to contain a `FileDescriptorSet` serialized as a string.
65 NOTE: You can build a `descriptor_source` file using the `--descriptor_set_out`
66 and `--include_imports` options to the protocol compiler `protoc`.
68 - A "bytes://<bytes>", in which protocol descriptors are created from `<bytes>`,
69 which is expected to be a `FileDescriptorSet` serialized as a string.
71 Args:
72 sizes: A `Tensor` of type `int32`.
73 Tensor of int32 with shape `[batch_shape, len(field_names)]`.
74 values: A list of `Tensor` objects.
75 List of tensors containing values for the corresponding field.
76 field_names: A list of `strings`.
77 List of strings containing proto field names.
78 message_type: A `string`. Name of the proto message type to decode.
79 descriptor_source: An optional `string`. Defaults to `"local://"`.
80 name: A name for the operation (optional).
82 Returns:
83 A `Tensor` of type `string`.
84 """
85 _ctx = _context._context or _context.context()
86 tld = _ctx._thread_local_data
87 if tld.is_eager:
88 try:
89 _result = pywrap_tfe.TFE_Py_FastPathExecute(
90 _ctx, "EncodeProto", name, sizes, values, "field_names", field_names,
91 "message_type", message_type, "descriptor_source", descriptor_source)
92 return _result
93 except _core._NotOkStatusException as e:
94 _ops.raise_from_not_ok_status(e, name)
95 except _core._FallbackException:
96 pass
97 try:
98 _result = _dispatcher_for_encode_proto(
99 (sizes, values, field_names, message_type, descriptor_source,
100 name,), None)
101 if _result is not NotImplemented:
102 return _result
103 return encode_proto_eager_fallback(
104 sizes, values, field_names=field_names, message_type=message_type,
105 descriptor_source=descriptor_source, name=name, ctx=_ctx)
106 except _core._SymbolicException:
107 pass # Add nodes to the TensorFlow graph.
108 except (TypeError, ValueError):
109 _result = _dispatch.dispatch(
110 encode_proto, (), dict(sizes=sizes, values=values,
111 field_names=field_names,
112 message_type=message_type,
113 descriptor_source=descriptor_source,
114 name=name)
115 )
116 if _result is not _dispatch.OpDispatcher.NOT_SUPPORTED:
117 return _result
118 raise
119 else:
120 _result = _dispatcher_for_encode_proto(
121 (sizes, values, field_names, message_type, descriptor_source, name,),
122 None)
123 if _result is not NotImplemented:
124 return _result
125 # Add nodes to the TensorFlow graph.
126 if not isinstance(field_names, (list, tuple)):
127 raise TypeError(
128 "Expected list for 'field_names' argument to "
129 "'encode_proto' Op, not %r." % field_names)
130 field_names = [_execute.make_str(_s, "field_names") for _s in field_names]
131 message_type = _execute.make_str(message_type, "message_type")
132 if descriptor_source is None:
133 descriptor_source = "local://"
134 descriptor_source = _execute.make_str(descriptor_source, "descriptor_source")
135 try:
136 _, _, _op, _outputs = _op_def_library._apply_op_helper(
137 "EncodeProto", sizes=sizes, values=values, field_names=field_names,
138 message_type=message_type,
139 descriptor_source=descriptor_source, name=name)
140 except (TypeError, ValueError):
141 _result = _dispatch.dispatch(
142 encode_proto, (), dict(sizes=sizes, values=values,
143 field_names=field_names,
144 message_type=message_type,
145 descriptor_source=descriptor_source,
146 name=name)
147 )
148 if _result is not _dispatch.OpDispatcher.NOT_SUPPORTED:
149 return _result
150 raise
151 _result = _outputs[:]
152 if _execute.must_record_gradient():
153 _attrs = ("field_names", _op.get_attr("field_names"), "message_type",
154 _op.get_attr("message_type"), "descriptor_source",
155 _op.get_attr("descriptor_source"), "Tinput_types",
156 _op.get_attr("Tinput_types"))
157 _inputs_flat = _op.inputs
158 _execute.record_gradient(
159 "EncodeProto", _inputs_flat, _attrs, _result)
160 _result, = _result
161 return _result
163EncodeProto = tf_export("raw_ops.EncodeProto")(_ops.to_raw_op(encode_proto))
164_dispatcher_for_encode_proto = encode_proto._tf_type_based_dispatcher.Dispatch
167def encode_proto_eager_fallback(sizes, values, field_names, message_type, descriptor_source, name, ctx):
168 if not isinstance(field_names, (list, tuple)):
169 raise TypeError(
170 "Expected list for 'field_names' argument to "
171 "'encode_proto' Op, not %r." % field_names)
172 field_names = [_execute.make_str(_s, "field_names") for _s in field_names]
173 message_type = _execute.make_str(message_type, "message_type")
174 if descriptor_source is None:
175 descriptor_source = "local://"
176 descriptor_source = _execute.make_str(descriptor_source, "descriptor_source")
177 _attr_Tinput_types, values = _execute.convert_to_mixed_eager_tensors(values, ctx)
178 sizes = _ops.convert_to_tensor(sizes, _dtypes.int32)
179 _inputs_flat = [sizes] + list(values)
180 _attrs = ("field_names", field_names, "message_type", message_type,
181 "descriptor_source", descriptor_source, "Tinput_types", _attr_Tinput_types)
182 _result = _execute.execute(b"EncodeProto", 1, inputs=_inputs_flat,
183 attrs=_attrs, ctx=ctx, name=name)
184 if _execute.must_record_gradient():
185 _execute.record_gradient(
186 "EncodeProto", _inputs_flat, _attrs, _result)
187 _result, = _result
188 return _result