diff --git a/docs/modelserving/observability/prometheus_metrics.md b/docs/modelserving/observability/prometheus_metrics.md index d7f5c3892..9508e2094 100644 --- a/docs/modelserving/observability/prometheus_metrics.md +++ b/docs/modelserving/observability/prometheus_metrics.md @@ -28,7 +28,7 @@ spec: predictor: sklearn: protocolVersion: v2 - storageUri: "gs://seldon-models/sklearn/mms/lr_model" + storageUri: "gs://seldon-models/sklearn/iris" ``` The default values for `serving.kserve.io/enable-prometheus-scraping` can be set in the `inferenceservice-config` configmap. See [the docs](https://github.com/kserve/kserve/blob/master/qpext/README.md#configs) for more info. diff --git a/docs/modelserving/v1beta1/onnx/README.md b/docs/modelserving/v1beta1/onnx/README.md new file mode 100644 index 000000000..7993ecefb --- /dev/null +++ b/docs/modelserving/v1beta1/onnx/README.md @@ -0,0 +1,67 @@ + +# Deploy InferenceService with ONNX model +## Setup +1. Your ~/.kube/config should point to a cluster with [KServe installed](https://github.com/kserve/kserve#installation). +2. Your cluster's Istio Ingress gateway must be [network accessible](https://istio.io/latest/docs/tasks/traffic-management/ingress/ingress-control/). + +## Create the InferenceService + +=== "New Schema" + + ```yaml + apiVersion: "serving.kserve.io/v1beta1" + kind: "InferenceService" + metadata: + name: "style-sample" + spec: + predictor: + model: + protocolVersion: v2 + modelFormat: + name: onnx + storageUri: "gs://kfserving-examples/models/onnx" + ``` + +=== "Old Schema" + + ```yaml + apiVersion: "serving.kserve.io/v1beta1" + kind: "InferenceService" + metadata: + name: "style-sample" + spec: + predictor: + onnx: + storageUri: "gs://kfserving-examples/models/onnx" + ``` +!!! Note + For the default kserve installation, While using new schema, you must specify **protocolVersion** as v2 for onnx. Otherwise, you will get a no runtime found error. +Expected Output +``` +$ inferenceservice.serving.kserve.io/style-sample configured +``` + +## Run a sample inference + +1. Setup env vars +The first step is to [determine the ingress IP and ports](https://kserve.github.io/website/master/get_started/first_isvc/#4-determine-the-ingress-ip-and-ports) and set `INGRESS_HOST` and `INGRESS_PORT` + +``` +export ISVC_NAME=style-sample +export SERVICE_HOSTNAME=$(kubectl get inferenceservice ${ISVC_NAME} -o jsonpath='{.status.url}' | cut -d "/" -f 3) +``` +2. Verify the service is healthy +``` +curl -v -H "Host:${SERVICE_HOSTNAME}" http://localhost:8080//v2/health/ready +``` +3. Install dependencies +``` +pip install -r requirements.txt +``` +4. Run the [sample notebook](mosaic-onnx.ipynb) in jupyter +``` +jupyter notebook +``` + +## Uploading your own model +The sample model for the example in this readme is already uploaded and available for use. However if you would like to modify the example to use your own ONNX model, all you need to do is to upload your model as `model.onnx` to S3, GCS or an Azure Blob. diff --git a/docs/modelserving/v1beta1/onnx/assets/onnx_ml_pb2.py b/docs/modelserving/v1beta1/onnx/assets/onnx_ml_pb2.py new file mode 100644 index 000000000..b7d159254 --- /dev/null +++ b/docs/modelserving/v1beta1/onnx/assets/onnx_ml_pb2.py @@ -0,0 +1,1581 @@ +# Generated by the protocol buffer compiler. DO NOT EDIT! +# source: onnx-ml.proto + +import sys +_b=sys.version_info[0]<3 and (lambda x:x) or (lambda x:x.encode('latin1')) +from google.protobuf.internal import enum_type_wrapper +from google.protobuf import descriptor as _descriptor +from google.protobuf import message as _message +from google.protobuf import reflection as _reflection +from google.protobuf import symbol_database as _symbol_database +# @@protoc_insertion_point(imports) + +_sym_db = _symbol_database.Default() + + + + +DESCRIPTOR = _descriptor.FileDescriptor( + name='onnx-ml.proto', + package='onnx', + syntax='proto3', + serialized_options=None, + serialized_pb=_b('\n\ronnx-ml.proto\x12\x04onnx\"\xe0\x03\n\x0e\x41ttributeProto\x12\x0c\n\x04name\x18\x01 \x01(\t\x12\x15\n\rref_attr_name\x18\x15 \x01(\t\x12\x12\n\ndoc_string\x18\r \x01(\t\x12\x30\n\x04type\x18\x14 \x01(\x0e\x32\".onnx.AttributeProto.AttributeType\x12\t\n\x01\x66\x18\x02 \x01(\x02\x12\t\n\x01i\x18\x03 \x01(\x03\x12\t\n\x01s\x18\x04 \x01(\x0c\x12\x1c\n\x01t\x18\x05 \x01(\x0b\x32\x11.onnx.TensorProto\x12\x1b\n\x01g\x18\x06 \x01(\x0b\x32\x10.onnx.GraphProto\x12\x0e\n\x06\x66loats\x18\x07 \x03(\x02\x12\x0c\n\x04ints\x18\x08 \x03(\x03\x12\x0f\n\x07strings\x18\t \x03(\x0c\x12\"\n\x07tensors\x18\n \x03(\x0b\x32\x11.onnx.TensorProto\x12 \n\x06graphs\x18\x0b \x03(\x0b\x32\x10.onnx.GraphProto\"\x91\x01\n\rAttributeType\x12\r\n\tUNDEFINED\x10\x00\x12\t\n\x05\x46LOAT\x10\x01\x12\x07\n\x03INT\x10\x02\x12\n\n\x06STRING\x10\x03\x12\n\n\x06TENSOR\x10\x04\x12\t\n\x05GRAPH\x10\x05\x12\n\n\x06\x46LOATS\x10\x06\x12\x08\n\x04INTS\x10\x07\x12\x0b\n\x07STRINGS\x10\x08\x12\x0b\n\x07TENSORS\x10\t\x12\n\n\x06GRAPHS\x10\n\"Q\n\x0eValueInfoProto\x12\x0c\n\x04name\x18\x01 \x01(\t\x12\x1d\n\x04type\x18\x02 \x01(\x0b\x32\x0f.onnx.TypeProto\x12\x12\n\ndoc_string\x18\x03 \x01(\t\"\x96\x01\n\tNodeProto\x12\r\n\x05input\x18\x01 \x03(\t\x12\x0e\n\x06output\x18\x02 \x03(\t\x12\x0c\n\x04name\x18\x03 \x01(\t\x12\x0f\n\x07op_type\x18\x04 \x01(\t\x12\x0e\n\x06\x64omain\x18\x07 \x01(\t\x12\'\n\tattribute\x18\x05 \x03(\x0b\x32\x14.onnx.AttributeProto\x12\x12\n\ndoc_string\x18\x06 \x01(\t\"\xbb\x02\n\nModelProto\x12\x12\n\nir_version\x18\x01 \x01(\x03\x12.\n\x0copset_import\x18\x08 \x03(\x0b\x32\x18.onnx.OperatorSetIdProto\x12\x15\n\rproducer_name\x18\x02 \x01(\t\x12\x18\n\x10producer_version\x18\x03 \x01(\t\x12\x0e\n\x06\x64omain\x18\x04 \x01(\t\x12\x15\n\rmodel_version\x18\x05 \x01(\x03\x12\x12\n\ndoc_string\x18\x06 \x01(\t\x12\x1f\n\x05graph\x18\x07 \x01(\x0b\x32\x10.onnx.GraphProto\x12&\n\tfunctions\x18\x64 \x03(\x0b\x32\x13.onnx.FunctionProto\x12\x34\n\x0emetadata_props\x18\x0e \x03(\x0b\x32\x1c.onnx.StringStringEntryProto\"4\n\x16StringStringEntryProto\x12\x0b\n\x03key\x18\x01 \x01(\t\x12\r\n\x05value\x18\x02 \x01(\t\"k\n\x10TensorAnnotation\x12\x13\n\x0btensor_name\x18\x01 \x01(\t\x12\x42\n\x1cquant_parameter_tensor_names\x18\x02 \x03(\x0b\x32\x1c.onnx.StringStringEntryProto\"\xa3\x02\n\nGraphProto\x12\x1d\n\x04node\x18\x01 \x03(\x0b\x32\x0f.onnx.NodeProto\x12\x0c\n\x04name\x18\x02 \x01(\t\x12&\n\x0binitializer\x18\x05 \x03(\x0b\x32\x11.onnx.TensorProto\x12\x12\n\ndoc_string\x18\n \x01(\t\x12#\n\x05input\x18\x0b \x03(\x0b\x32\x14.onnx.ValueInfoProto\x12$\n\x06output\x18\x0c \x03(\x0b\x32\x14.onnx.ValueInfoProto\x12(\n\nvalue_info\x18\r \x03(\x0b\x32\x14.onnx.ValueInfoProto\x12\x37\n\x17quantization_annotation\x18\x0e \x03(\x0b\x32\x16.onnx.TensorAnnotation\"\xb8\x05\n\x0bTensorProto\x12\x0c\n\x04\x64ims\x18\x01 \x03(\x03\x12\x11\n\tdata_type\x18\x02 \x01(\x05\x12*\n\x07segment\x18\x03 \x01(\x0b\x32\x19.onnx.TensorProto.Segment\x12\x16\n\nfloat_data\x18\x04 \x03(\x02\x42\x02\x10\x01\x12\x16\n\nint32_data\x18\x05 \x03(\x05\x42\x02\x10\x01\x12\x13\n\x0bstring_data\x18\x06 \x03(\x0c\x12\x16\n\nint64_data\x18\x07 \x03(\x03\x42\x02\x10\x01\x12\x0c\n\x04name\x18\x08 \x01(\t\x12\x12\n\ndoc_string\x18\x0c \x01(\t\x12\x10\n\x08raw_data\x18\t \x01(\x0c\x12\x33\n\rexternal_data\x18\r \x03(\x0b\x32\x1c.onnx.StringStringEntryProto\x12\x35\n\rdata_location\x18\x0e \x01(\x0e\x32\x1e.onnx.TensorProto.DataLocation\x12\x17\n\x0b\x64ouble_data\x18\n \x03(\x01\x42\x02\x10\x01\x12\x17\n\x0buint64_data\x18\x0b \x03(\x04\x42\x02\x10\x01\x1a%\n\x07Segment\x12\r\n\x05\x62\x65gin\x18\x01 \x01(\x03\x12\x0b\n\x03\x65nd\x18\x02 \x01(\x03\"\xda\x01\n\x08\x44\x61taType\x12\r\n\tUNDEFINED\x10\x00\x12\t\n\x05\x46LOAT\x10\x01\x12\t\n\x05UINT8\x10\x02\x12\x08\n\x04INT8\x10\x03\x12\n\n\x06UINT16\x10\x04\x12\t\n\x05INT16\x10\x05\x12\t\n\x05INT32\x10\x06\x12\t\n\x05INT64\x10\x07\x12\n\n\x06STRING\x10\x08\x12\x08\n\x04\x42OOL\x10\t\x12\x0b\n\x07\x46LOAT16\x10\n\x12\n\n\x06\x44OUBLE\x10\x0b\x12\n\n\x06UINT32\x10\x0c\x12\n\n\x06UINT64\x10\r\x12\r\n\tCOMPLEX64\x10\x0e\x12\x0e\n\nCOMPLEX128\x10\x0f\x12\x0c\n\x08\x42\x46LOAT16\x10\x10\")\n\x0c\x44\x61taLocation\x12\x0b\n\x07\x44\x45\x46\x41ULT\x10\x00\x12\x0c\n\x08\x45XTERNAL\x10\x01\"\x95\x01\n\x10TensorShapeProto\x12-\n\x03\x64im\x18\x01 \x03(\x0b\x32 .onnx.TensorShapeProto.Dimension\x1aR\n\tDimension\x12\x13\n\tdim_value\x18\x01 \x01(\x03H\x00\x12\x13\n\tdim_param\x18\x02 \x01(\tH\x00\x12\x12\n\ndenotation\x18\x03 \x01(\tB\x07\n\x05value\"\xc2\x04\n\tTypeProto\x12-\n\x0btensor_type\x18\x01 \x01(\x0b\x32\x16.onnx.TypeProto.TensorH\x00\x12\x31\n\rsequence_type\x18\x04 \x01(\x0b\x32\x18.onnx.TypeProto.SequenceH\x00\x12\'\n\x08map_type\x18\x05 \x01(\x0b\x32\x13.onnx.TypeProto.MapH\x00\x12-\n\x0bopaque_type\x18\x07 \x01(\x0b\x32\x16.onnx.TypeProto.OpaqueH\x00\x12:\n\x12sparse_tensor_type\x18\x08 \x01(\x0b\x32\x1c.onnx.TypeProto.SparseTensorH\x00\x12\x12\n\ndenotation\x18\x06 \x01(\t\x1a\x42\n\x06Tensor\x12\x11\n\telem_type\x18\x01 \x01(\x05\x12%\n\x05shape\x18\x02 \x01(\x0b\x32\x16.onnx.TensorShapeProto\x1a.\n\x08Sequence\x12\"\n\telem_type\x18\x01 \x01(\x0b\x32\x0f.onnx.TypeProto\x1a<\n\x03Map\x12\x10\n\x08key_type\x18\x01 \x01(\x05\x12#\n\nvalue_type\x18\x02 \x01(\x0b\x32\x0f.onnx.TypeProto\x1a&\n\x06Opaque\x12\x0e\n\x06\x64omain\x18\x01 \x01(\t\x12\x0c\n\x04name\x18\x02 \x01(\t\x1aH\n\x0cSparseTensor\x12\x11\n\telem_type\x18\x01 \x01(\x05\x12%\n\x05shape\x18\x02 \x01(\x0b\x32\x16.onnx.TensorShapeProtoB\x07\n\x05value\"5\n\x12OperatorSetIdProto\x12\x0e\n\x06\x64omain\x18\x01 \x01(\t\x12\x0f\n\x07version\x18\x02 \x01(\x03\"\xbf\x01\n\rFunctionProto\x12\x0c\n\x04name\x18\x01 \x01(\t\x12\x15\n\rsince_version\x18\x02 \x01(\x03\x12$\n\x06status\x18\x03 \x01(\x0e\x32\x14.onnx.OperatorStatus\x12\r\n\x05input\x18\x04 \x03(\t\x12\x0e\n\x06output\x18\x05 \x03(\t\x12\x11\n\tattribute\x18\x06 \x03(\t\x12\x1d\n\x04node\x18\x07 \x03(\x0b\x32\x0f.onnx.NodeProto\x12\x12\n\ndoc_string\x18\x08 \x01(\t*\x97\x01\n\x07Version\x12\x12\n\x0e_START_VERSION\x10\x00\x12\x19\n\x15IR_VERSION_2017_10_10\x10\x01\x12\x19\n\x15IR_VERSION_2017_10_30\x10\x02\x12\x18\n\x14IR_VERSION_2017_11_3\x10\x03\x12\x18\n\x14IR_VERSION_2019_1_22\x10\x04\x12\x0e\n\nIR_VERSION\x10\x05*.\n\x0eOperatorStatus\x12\x10\n\x0c\x45XPERIMENTAL\x10\x00\x12\n\n\x06STABLE\x10\x01\x62\x06proto3') +) + +_VERSION = _descriptor.EnumDescriptor( + name='Version', + full_name='onnx.Version', + filename=None, + file=DESCRIPTOR, + values=[ + _descriptor.EnumValueDescriptor( + name='_START_VERSION', index=0, number=0, + serialized_options=None, + type=None), + _descriptor.EnumValueDescriptor( + name='IR_VERSION_2017_10_10', index=1, number=1, + serialized_options=None, + type=None), + _descriptor.EnumValueDescriptor( + name='IR_VERSION_2017_10_30', index=2, number=2, + serialized_options=None, + type=None), + _descriptor.EnumValueDescriptor( + name='IR_VERSION_2017_11_3', index=3, number=3, + serialized_options=None, + type=None), + _descriptor.EnumValueDescriptor( + name='IR_VERSION_2019_1_22', index=4, number=4, + serialized_options=None, + type=None), + _descriptor.EnumValueDescriptor( + name='IR_VERSION', index=5, number=5, + serialized_options=None, + type=None), + ], + containing_type=None, + serialized_options=None, + serialized_start=3199, + serialized_end=3350, +) +_sym_db.RegisterEnumDescriptor(_VERSION) + +Version = enum_type_wrapper.EnumTypeWrapper(_VERSION) +_OPERATORSTATUS = _descriptor.EnumDescriptor( + name='OperatorStatus', + full_name='onnx.OperatorStatus', + filename=None, + file=DESCRIPTOR, + values=[ + _descriptor.EnumValueDescriptor( + name='EXPERIMENTAL', index=0, number=0, + serialized_options=None, + type=None), + _descriptor.EnumValueDescriptor( + name='STABLE', index=1, number=1, + serialized_options=None, + type=None), + ], + containing_type=None, + serialized_options=None, + serialized_start=3352, + serialized_end=3398, +) +_sym_db.RegisterEnumDescriptor(_OPERATORSTATUS) + +OperatorStatus = enum_type_wrapper.EnumTypeWrapper(_OPERATORSTATUS) +_START_VERSION = 0 +IR_VERSION_2017_10_10 = 1 +IR_VERSION_2017_10_30 = 2 +IR_VERSION_2017_11_3 = 3 +IR_VERSION_2019_1_22 = 4 +IR_VERSION = 5 +EXPERIMENTAL = 0 +STABLE = 1 + + +_ATTRIBUTEPROTO_ATTRIBUTETYPE = _descriptor.EnumDescriptor( + name='AttributeType', + full_name='onnx.AttributeProto.AttributeType', + filename=None, + file=DESCRIPTOR, + values=[ + _descriptor.EnumValueDescriptor( + name='UNDEFINED', index=0, number=0, + serialized_options=None, + type=None), + _descriptor.EnumValueDescriptor( + name='FLOAT', index=1, number=1, + serialized_options=None, + type=None), + _descriptor.EnumValueDescriptor( + name='INT', index=2, number=2, + serialized_options=None, + type=None), + _descriptor.EnumValueDescriptor( + name='STRING', index=3, number=3, + serialized_options=None, + type=None), + _descriptor.EnumValueDescriptor( + name='TENSOR', index=4, number=4, + serialized_options=None, + type=None), + _descriptor.EnumValueDescriptor( + name='GRAPH', index=5, number=5, + serialized_options=None, + type=None), + _descriptor.EnumValueDescriptor( + name='FLOATS', index=6, number=6, + serialized_options=None, + type=None), + _descriptor.EnumValueDescriptor( + name='INTS', index=7, number=7, + serialized_options=None, + type=None), + _descriptor.EnumValueDescriptor( + name='STRINGS', index=8, number=8, + serialized_options=None, + type=None), + _descriptor.EnumValueDescriptor( + name='TENSORS', index=9, number=9, + serialized_options=None, + type=None), + _descriptor.EnumValueDescriptor( + name='GRAPHS', index=10, number=10, + serialized_options=None, + type=None), + ], + containing_type=None, + serialized_options=None, + serialized_start=359, + serialized_end=504, +) +_sym_db.RegisterEnumDescriptor(_ATTRIBUTEPROTO_ATTRIBUTETYPE) + +_TENSORPROTO_DATATYPE = _descriptor.EnumDescriptor( + name='DataType', + full_name='onnx.TensorProto.DataType', + filename=None, + file=DESCRIPTOR, + values=[ + _descriptor.EnumValueDescriptor( + name='UNDEFINED', index=0, number=0, + serialized_options=None, + type=None), + _descriptor.EnumValueDescriptor( + name='FLOAT', index=1, number=1, + serialized_options=None, + type=None), + _descriptor.EnumValueDescriptor( + name='UINT8', index=2, number=2, + serialized_options=None, + type=None), + _descriptor.EnumValueDescriptor( + name='INT8', index=3, number=3, + serialized_options=None, + type=None), + _descriptor.EnumValueDescriptor( + name='UINT16', index=4, number=4, + serialized_options=None, + type=None), + _descriptor.EnumValueDescriptor( + name='INT16', index=5, number=5, + serialized_options=None, + type=None), + _descriptor.EnumValueDescriptor( + name='INT32', index=6, number=6, + serialized_options=None, + type=None), + _descriptor.EnumValueDescriptor( + name='INT64', index=7, number=7, + serialized_options=None, + type=None), + _descriptor.EnumValueDescriptor( + name='STRING', index=8, number=8, + serialized_options=None, + type=None), + _descriptor.EnumValueDescriptor( + name='BOOL', index=9, number=9, + serialized_options=None, + type=None), + _descriptor.EnumValueDescriptor( + name='FLOAT16', index=10, number=10, + serialized_options=None, + type=None), + _descriptor.EnumValueDescriptor( + name='DOUBLE', index=11, number=11, + serialized_options=None, + type=None), + _descriptor.EnumValueDescriptor( + name='UINT32', index=12, number=12, + serialized_options=None, + type=None), + _descriptor.EnumValueDescriptor( + name='UINT64', index=13, number=13, + serialized_options=None, + type=None), + _descriptor.EnumValueDescriptor( + name='COMPLEX64', index=14, number=14, + serialized_options=None, + type=None), + _descriptor.EnumValueDescriptor( + name='COMPLEX128', index=15, number=15, + serialized_options=None, + type=None), + _descriptor.EnumValueDescriptor( + name='BFLOAT16', index=16, number=16, + serialized_options=None, + type=None), + ], + containing_type=None, + serialized_options=None, + serialized_start=1953, + serialized_end=2171, +) +_sym_db.RegisterEnumDescriptor(_TENSORPROTO_DATATYPE) + +_TENSORPROTO_DATALOCATION = _descriptor.EnumDescriptor( + name='DataLocation', + full_name='onnx.TensorProto.DataLocation', + filename=None, + file=DESCRIPTOR, + values=[ + _descriptor.EnumValueDescriptor( + name='DEFAULT', index=0, number=0, + serialized_options=None, + type=None), + _descriptor.EnumValueDescriptor( + name='EXTERNAL', index=1, number=1, + serialized_options=None, + type=None), + ], + containing_type=None, + serialized_options=None, + serialized_start=2173, + serialized_end=2214, +) +_sym_db.RegisterEnumDescriptor(_TENSORPROTO_DATALOCATION) + + +_ATTRIBUTEPROTO = _descriptor.Descriptor( + name='AttributeProto', + full_name='onnx.AttributeProto', + filename=None, + file=DESCRIPTOR, + containing_type=None, + fields=[ + _descriptor.FieldDescriptor( + name='name', full_name='onnx.AttributeProto.name', index=0, + number=1, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=_b("").decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='ref_attr_name', full_name='onnx.AttributeProto.ref_attr_name', index=1, + number=21, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=_b("").decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='doc_string', full_name='onnx.AttributeProto.doc_string', index=2, + number=13, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=_b("").decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='type', full_name='onnx.AttributeProto.type', index=3, + number=20, type=14, cpp_type=8, label=1, + has_default_value=False, default_value=0, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='f', full_name='onnx.AttributeProto.f', index=4, + number=2, type=2, cpp_type=6, label=1, + has_default_value=False, default_value=float(0), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='i', full_name='onnx.AttributeProto.i', index=5, + number=3, type=3, cpp_type=2, label=1, + has_default_value=False, default_value=0, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='s', full_name='onnx.AttributeProto.s', index=6, + number=4, type=12, cpp_type=9, label=1, + has_default_value=False, default_value=_b(""), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='t', full_name='onnx.AttributeProto.t', index=7, + number=5, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='g', full_name='onnx.AttributeProto.g', index=8, + number=6, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='floats', full_name='onnx.AttributeProto.floats', index=9, + number=7, type=2, cpp_type=6, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='ints', full_name='onnx.AttributeProto.ints', index=10, + number=8, type=3, cpp_type=2, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='strings', full_name='onnx.AttributeProto.strings', index=11, + number=9, type=12, cpp_type=9, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='tensors', full_name='onnx.AttributeProto.tensors', index=12, + number=10, type=11, cpp_type=10, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='graphs', full_name='onnx.AttributeProto.graphs', index=13, + number=11, type=11, cpp_type=10, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + _ATTRIBUTEPROTO_ATTRIBUTETYPE, + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=24, + serialized_end=504, +) + + +_VALUEINFOPROTO = _descriptor.Descriptor( + name='ValueInfoProto', + full_name='onnx.ValueInfoProto', + filename=None, + file=DESCRIPTOR, + containing_type=None, + fields=[ + _descriptor.FieldDescriptor( + name='name', full_name='onnx.ValueInfoProto.name', index=0, + number=1, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=_b("").decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='type', full_name='onnx.ValueInfoProto.type', index=1, + number=2, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='doc_string', full_name='onnx.ValueInfoProto.doc_string', index=2, + number=3, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=_b("").decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=506, + serialized_end=587, +) + + +_NODEPROTO = _descriptor.Descriptor( + name='NodeProto', + full_name='onnx.NodeProto', + filename=None, + file=DESCRIPTOR, + containing_type=None, + fields=[ + _descriptor.FieldDescriptor( + name='input', full_name='onnx.NodeProto.input', index=0, + number=1, type=9, cpp_type=9, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='output', full_name='onnx.NodeProto.output', index=1, + number=2, type=9, cpp_type=9, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='name', full_name='onnx.NodeProto.name', index=2, + number=3, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=_b("").decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='op_type', full_name='onnx.NodeProto.op_type', index=3, + number=4, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=_b("").decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='domain', full_name='onnx.NodeProto.domain', index=4, + number=7, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=_b("").decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='attribute', full_name='onnx.NodeProto.attribute', index=5, + number=5, type=11, cpp_type=10, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='doc_string', full_name='onnx.NodeProto.doc_string', index=6, + number=6, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=_b("").decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=590, + serialized_end=740, +) + + +_MODELPROTO = _descriptor.Descriptor( + name='ModelProto', + full_name='onnx.ModelProto', + filename=None, + file=DESCRIPTOR, + containing_type=None, + fields=[ + _descriptor.FieldDescriptor( + name='ir_version', full_name='onnx.ModelProto.ir_version', index=0, + number=1, type=3, cpp_type=2, label=1, + has_default_value=False, default_value=0, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='opset_import', full_name='onnx.ModelProto.opset_import', index=1, + number=8, type=11, cpp_type=10, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='producer_name', full_name='onnx.ModelProto.producer_name', index=2, + number=2, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=_b("").decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='producer_version', full_name='onnx.ModelProto.producer_version', index=3, + number=3, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=_b("").decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='domain', full_name='onnx.ModelProto.domain', index=4, + number=4, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=_b("").decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='model_version', full_name='onnx.ModelProto.model_version', index=5, + number=5, type=3, cpp_type=2, label=1, + has_default_value=False, default_value=0, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='doc_string', full_name='onnx.ModelProto.doc_string', index=6, + number=6, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=_b("").decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='graph', full_name='onnx.ModelProto.graph', index=7, + number=7, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='functions', full_name='onnx.ModelProto.functions', index=8, + number=100, type=11, cpp_type=10, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='metadata_props', full_name='onnx.ModelProto.metadata_props', index=9, + number=14, type=11, cpp_type=10, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=743, + serialized_end=1058, +) + + +_STRINGSTRINGENTRYPROTO = _descriptor.Descriptor( + name='StringStringEntryProto', + full_name='onnx.StringStringEntryProto', + filename=None, + file=DESCRIPTOR, + containing_type=None, + fields=[ + _descriptor.FieldDescriptor( + name='key', full_name='onnx.StringStringEntryProto.key', index=0, + number=1, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=_b("").decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='value', full_name='onnx.StringStringEntryProto.value', index=1, + number=2, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=_b("").decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=1060, + serialized_end=1112, +) + + +_TENSORANNOTATION = _descriptor.Descriptor( + name='TensorAnnotation', + full_name='onnx.TensorAnnotation', + filename=None, + file=DESCRIPTOR, + containing_type=None, + fields=[ + _descriptor.FieldDescriptor( + name='tensor_name', full_name='onnx.TensorAnnotation.tensor_name', index=0, + number=1, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=_b("").decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='quant_parameter_tensor_names', full_name='onnx.TensorAnnotation.quant_parameter_tensor_names', index=1, + number=2, type=11, cpp_type=10, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=1114, + serialized_end=1221, +) + + +_GRAPHPROTO = _descriptor.Descriptor( + name='GraphProto', + full_name='onnx.GraphProto', + filename=None, + file=DESCRIPTOR, + containing_type=None, + fields=[ + _descriptor.FieldDescriptor( + name='node', full_name='onnx.GraphProto.node', index=0, + number=1, type=11, cpp_type=10, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='name', full_name='onnx.GraphProto.name', index=1, + number=2, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=_b("").decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='initializer', full_name='onnx.GraphProto.initializer', index=2, + number=5, type=11, cpp_type=10, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='doc_string', full_name='onnx.GraphProto.doc_string', index=3, + number=10, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=_b("").decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='input', full_name='onnx.GraphProto.input', index=4, + number=11, type=11, cpp_type=10, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='output', full_name='onnx.GraphProto.output', index=5, + number=12, type=11, cpp_type=10, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='value_info', full_name='onnx.GraphProto.value_info', index=6, + number=13, type=11, cpp_type=10, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='quantization_annotation', full_name='onnx.GraphProto.quantization_annotation', index=7, + number=14, type=11, cpp_type=10, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=1224, + serialized_end=1515, +) + + +_TENSORPROTO_SEGMENT = _descriptor.Descriptor( + name='Segment', + full_name='onnx.TensorProto.Segment', + filename=None, + file=DESCRIPTOR, + containing_type=None, + fields=[ + _descriptor.FieldDescriptor( + name='begin', full_name='onnx.TensorProto.Segment.begin', index=0, + number=1, type=3, cpp_type=2, label=1, + has_default_value=False, default_value=0, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='end', full_name='onnx.TensorProto.Segment.end', index=1, + number=2, type=3, cpp_type=2, label=1, + has_default_value=False, default_value=0, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=1913, + serialized_end=1950, +) + +_TENSORPROTO = _descriptor.Descriptor( + name='TensorProto', + full_name='onnx.TensorProto', + filename=None, + file=DESCRIPTOR, + containing_type=None, + fields=[ + _descriptor.FieldDescriptor( + name='dims', full_name='onnx.TensorProto.dims', index=0, + number=1, type=3, cpp_type=2, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='data_type', full_name='onnx.TensorProto.data_type', index=1, + number=2, type=5, cpp_type=1, label=1, + has_default_value=False, default_value=0, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='segment', full_name='onnx.TensorProto.segment', index=2, + number=3, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='float_data', full_name='onnx.TensorProto.float_data', index=3, + number=4, type=2, cpp_type=6, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=_b('\020\001'), file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='int32_data', full_name='onnx.TensorProto.int32_data', index=4, + number=5, type=5, cpp_type=1, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=_b('\020\001'), file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='string_data', full_name='onnx.TensorProto.string_data', index=5, + number=6, type=12, cpp_type=9, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='int64_data', full_name='onnx.TensorProto.int64_data', index=6, + number=7, type=3, cpp_type=2, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=_b('\020\001'), file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='name', full_name='onnx.TensorProto.name', index=7, + number=8, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=_b("").decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='doc_string', full_name='onnx.TensorProto.doc_string', index=8, + number=12, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=_b("").decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='raw_data', full_name='onnx.TensorProto.raw_data', index=9, + number=9, type=12, cpp_type=9, label=1, + has_default_value=False, default_value=_b(""), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='external_data', full_name='onnx.TensorProto.external_data', index=10, + number=13, type=11, cpp_type=10, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='data_location', full_name='onnx.TensorProto.data_location', index=11, + number=14, type=14, cpp_type=8, label=1, + has_default_value=False, default_value=0, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='double_data', full_name='onnx.TensorProto.double_data', index=12, + number=10, type=1, cpp_type=5, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=_b('\020\001'), file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='uint64_data', full_name='onnx.TensorProto.uint64_data', index=13, + number=11, type=4, cpp_type=4, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=_b('\020\001'), file=DESCRIPTOR), + ], + extensions=[ + ], + nested_types=[_TENSORPROTO_SEGMENT, ], + enum_types=[ + _TENSORPROTO_DATATYPE, + _TENSORPROTO_DATALOCATION, + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=1518, + serialized_end=2214, +) + + +_TENSORSHAPEPROTO_DIMENSION = _descriptor.Descriptor( + name='Dimension', + full_name='onnx.TensorShapeProto.Dimension', + filename=None, + file=DESCRIPTOR, + containing_type=None, + fields=[ + _descriptor.FieldDescriptor( + name='dim_value', full_name='onnx.TensorShapeProto.Dimension.dim_value', index=0, + number=1, type=3, cpp_type=2, label=1, + has_default_value=False, default_value=0, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='dim_param', full_name='onnx.TensorShapeProto.Dimension.dim_param', index=1, + number=2, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=_b("").decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='denotation', full_name='onnx.TensorShapeProto.Dimension.denotation', index=2, + number=3, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=_b("").decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + _descriptor.OneofDescriptor( + name='value', full_name='onnx.TensorShapeProto.Dimension.value', + index=0, containing_type=None, fields=[]), + ], + serialized_start=2284, + serialized_end=2366, +) + +_TENSORSHAPEPROTO = _descriptor.Descriptor( + name='TensorShapeProto', + full_name='onnx.TensorShapeProto', + filename=None, + file=DESCRIPTOR, + containing_type=None, + fields=[ + _descriptor.FieldDescriptor( + name='dim', full_name='onnx.TensorShapeProto.dim', index=0, + number=1, type=11, cpp_type=10, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + ], + extensions=[ + ], + nested_types=[_TENSORSHAPEPROTO_DIMENSION, ], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=2217, + serialized_end=2366, +) + + +_TYPEPROTO_TENSOR = _descriptor.Descriptor( + name='Tensor', + full_name='onnx.TypeProto.Tensor', + filename=None, + file=DESCRIPTOR, + containing_type=None, + fields=[ + _descriptor.FieldDescriptor( + name='elem_type', full_name='onnx.TypeProto.Tensor.elem_type', index=0, + number=1, type=5, cpp_type=1, label=1, + has_default_value=False, default_value=0, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='shape', full_name='onnx.TypeProto.Tensor.shape', index=1, + number=2, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=2648, + serialized_end=2714, +) + +_TYPEPROTO_SEQUENCE = _descriptor.Descriptor( + name='Sequence', + full_name='onnx.TypeProto.Sequence', + filename=None, + file=DESCRIPTOR, + containing_type=None, + fields=[ + _descriptor.FieldDescriptor( + name='elem_type', full_name='onnx.TypeProto.Sequence.elem_type', index=0, + number=1, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=2716, + serialized_end=2762, +) + +_TYPEPROTO_MAP = _descriptor.Descriptor( + name='Map', + full_name='onnx.TypeProto.Map', + filename=None, + file=DESCRIPTOR, + containing_type=None, + fields=[ + _descriptor.FieldDescriptor( + name='key_type', full_name='onnx.TypeProto.Map.key_type', index=0, + number=1, type=5, cpp_type=1, label=1, + has_default_value=False, default_value=0, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='value_type', full_name='onnx.TypeProto.Map.value_type', index=1, + number=2, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=2764, + serialized_end=2824, +) + +_TYPEPROTO_OPAQUE = _descriptor.Descriptor( + name='Opaque', + full_name='onnx.TypeProto.Opaque', + filename=None, + file=DESCRIPTOR, + containing_type=None, + fields=[ + _descriptor.FieldDescriptor( + name='domain', full_name='onnx.TypeProto.Opaque.domain', index=0, + number=1, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=_b("").decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='name', full_name='onnx.TypeProto.Opaque.name', index=1, + number=2, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=_b("").decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=2826, + serialized_end=2864, +) + +_TYPEPROTO_SPARSETENSOR = _descriptor.Descriptor( + name='SparseTensor', + full_name='onnx.TypeProto.SparseTensor', + filename=None, + file=DESCRIPTOR, + containing_type=None, + fields=[ + _descriptor.FieldDescriptor( + name='elem_type', full_name='onnx.TypeProto.SparseTensor.elem_type', index=0, + number=1, type=5, cpp_type=1, label=1, + has_default_value=False, default_value=0, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='shape', full_name='onnx.TypeProto.SparseTensor.shape', index=1, + number=2, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=2866, + serialized_end=2938, +) + +_TYPEPROTO = _descriptor.Descriptor( + name='TypeProto', + full_name='onnx.TypeProto', + filename=None, + file=DESCRIPTOR, + containing_type=None, + fields=[ + _descriptor.FieldDescriptor( + name='tensor_type', full_name='onnx.TypeProto.tensor_type', index=0, + number=1, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='sequence_type', full_name='onnx.TypeProto.sequence_type', index=1, + number=4, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='map_type', full_name='onnx.TypeProto.map_type', index=2, + number=5, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='opaque_type', full_name='onnx.TypeProto.opaque_type', index=3, + number=7, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='sparse_tensor_type', full_name='onnx.TypeProto.sparse_tensor_type', index=4, + number=8, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='denotation', full_name='onnx.TypeProto.denotation', index=5, + number=6, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=_b("").decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + ], + extensions=[ + ], + nested_types=[_TYPEPROTO_TENSOR, _TYPEPROTO_SEQUENCE, _TYPEPROTO_MAP, _TYPEPROTO_OPAQUE, _TYPEPROTO_SPARSETENSOR, ], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + _descriptor.OneofDescriptor( + name='value', full_name='onnx.TypeProto.value', + index=0, containing_type=None, fields=[]), + ], + serialized_start=2369, + serialized_end=2947, +) + + +_OPERATORSETIDPROTO = _descriptor.Descriptor( + name='OperatorSetIdProto', + full_name='onnx.OperatorSetIdProto', + filename=None, + file=DESCRIPTOR, + containing_type=None, + fields=[ + _descriptor.FieldDescriptor( + name='domain', full_name='onnx.OperatorSetIdProto.domain', index=0, + number=1, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=_b("").decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='version', full_name='onnx.OperatorSetIdProto.version', index=1, + number=2, type=3, cpp_type=2, label=1, + has_default_value=False, default_value=0, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=2949, + serialized_end=3002, +) + + +_FUNCTIONPROTO = _descriptor.Descriptor( + name='FunctionProto', + full_name='onnx.FunctionProto', + filename=None, + file=DESCRIPTOR, + containing_type=None, + fields=[ + _descriptor.FieldDescriptor( + name='name', full_name='onnx.FunctionProto.name', index=0, + number=1, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=_b("").decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='since_version', full_name='onnx.FunctionProto.since_version', index=1, + number=2, type=3, cpp_type=2, label=1, + has_default_value=False, default_value=0, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='status', full_name='onnx.FunctionProto.status', index=2, + number=3, type=14, cpp_type=8, label=1, + has_default_value=False, default_value=0, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='input', full_name='onnx.FunctionProto.input', index=3, + number=4, type=9, cpp_type=9, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='output', full_name='onnx.FunctionProto.output', index=4, + number=5, type=9, cpp_type=9, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='attribute', full_name='onnx.FunctionProto.attribute', index=5, + number=6, type=9, cpp_type=9, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='node', full_name='onnx.FunctionProto.node', index=6, + number=7, type=11, cpp_type=10, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='doc_string', full_name='onnx.FunctionProto.doc_string', index=7, + number=8, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=_b("").decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=3005, + serialized_end=3196, +) + +_ATTRIBUTEPROTO.fields_by_name['type'].enum_type = _ATTRIBUTEPROTO_ATTRIBUTETYPE +_ATTRIBUTEPROTO.fields_by_name['t'].message_type = _TENSORPROTO +_ATTRIBUTEPROTO.fields_by_name['g'].message_type = _GRAPHPROTO +_ATTRIBUTEPROTO.fields_by_name['tensors'].message_type = _TENSORPROTO +_ATTRIBUTEPROTO.fields_by_name['graphs'].message_type = _GRAPHPROTO +_ATTRIBUTEPROTO_ATTRIBUTETYPE.containing_type = _ATTRIBUTEPROTO +_VALUEINFOPROTO.fields_by_name['type'].message_type = _TYPEPROTO +_NODEPROTO.fields_by_name['attribute'].message_type = _ATTRIBUTEPROTO +_MODELPROTO.fields_by_name['opset_import'].message_type = _OPERATORSETIDPROTO +_MODELPROTO.fields_by_name['graph'].message_type = _GRAPHPROTO +_MODELPROTO.fields_by_name['functions'].message_type = _FUNCTIONPROTO +_MODELPROTO.fields_by_name['metadata_props'].message_type = _STRINGSTRINGENTRYPROTO +_TENSORANNOTATION.fields_by_name['quant_parameter_tensor_names'].message_type = _STRINGSTRINGENTRYPROTO +_GRAPHPROTO.fields_by_name['node'].message_type = _NODEPROTO +_GRAPHPROTO.fields_by_name['initializer'].message_type = _TENSORPROTO +_GRAPHPROTO.fields_by_name['input'].message_type = _VALUEINFOPROTO +_GRAPHPROTO.fields_by_name['output'].message_type = _VALUEINFOPROTO +_GRAPHPROTO.fields_by_name['value_info'].message_type = _VALUEINFOPROTO +_GRAPHPROTO.fields_by_name['quantization_annotation'].message_type = _TENSORANNOTATION +_TENSORPROTO_SEGMENT.containing_type = _TENSORPROTO +_TENSORPROTO.fields_by_name['segment'].message_type = _TENSORPROTO_SEGMENT +_TENSORPROTO.fields_by_name['external_data'].message_type = _STRINGSTRINGENTRYPROTO +_TENSORPROTO.fields_by_name['data_location'].enum_type = _TENSORPROTO_DATALOCATION +_TENSORPROTO_DATATYPE.containing_type = _TENSORPROTO +_TENSORPROTO_DATALOCATION.containing_type = _TENSORPROTO +_TENSORSHAPEPROTO_DIMENSION.containing_type = _TENSORSHAPEPROTO +_TENSORSHAPEPROTO_DIMENSION.oneofs_by_name['value'].fields.append( + _TENSORSHAPEPROTO_DIMENSION.fields_by_name['dim_value']) +_TENSORSHAPEPROTO_DIMENSION.fields_by_name['dim_value'].containing_oneof = _TENSORSHAPEPROTO_DIMENSION.oneofs_by_name['value'] +_TENSORSHAPEPROTO_DIMENSION.oneofs_by_name['value'].fields.append( + _TENSORSHAPEPROTO_DIMENSION.fields_by_name['dim_param']) +_TENSORSHAPEPROTO_DIMENSION.fields_by_name['dim_param'].containing_oneof = _TENSORSHAPEPROTO_DIMENSION.oneofs_by_name['value'] +_TENSORSHAPEPROTO.fields_by_name['dim'].message_type = _TENSORSHAPEPROTO_DIMENSION +_TYPEPROTO_TENSOR.fields_by_name['shape'].message_type = _TENSORSHAPEPROTO +_TYPEPROTO_TENSOR.containing_type = _TYPEPROTO +_TYPEPROTO_SEQUENCE.fields_by_name['elem_type'].message_type = _TYPEPROTO +_TYPEPROTO_SEQUENCE.containing_type = _TYPEPROTO +_TYPEPROTO_MAP.fields_by_name['value_type'].message_type = _TYPEPROTO +_TYPEPROTO_MAP.containing_type = _TYPEPROTO +_TYPEPROTO_OPAQUE.containing_type = _TYPEPROTO +_TYPEPROTO_SPARSETENSOR.fields_by_name['shape'].message_type = _TENSORSHAPEPROTO +_TYPEPROTO_SPARSETENSOR.containing_type = _TYPEPROTO +_TYPEPROTO.fields_by_name['tensor_type'].message_type = _TYPEPROTO_TENSOR +_TYPEPROTO.fields_by_name['sequence_type'].message_type = _TYPEPROTO_SEQUENCE +_TYPEPROTO.fields_by_name['map_type'].message_type = _TYPEPROTO_MAP +_TYPEPROTO.fields_by_name['opaque_type'].message_type = _TYPEPROTO_OPAQUE +_TYPEPROTO.fields_by_name['sparse_tensor_type'].message_type = _TYPEPROTO_SPARSETENSOR +_TYPEPROTO.oneofs_by_name['value'].fields.append( + _TYPEPROTO.fields_by_name['tensor_type']) +_TYPEPROTO.fields_by_name['tensor_type'].containing_oneof = _TYPEPROTO.oneofs_by_name['value'] +_TYPEPROTO.oneofs_by_name['value'].fields.append( + _TYPEPROTO.fields_by_name['sequence_type']) +_TYPEPROTO.fields_by_name['sequence_type'].containing_oneof = _TYPEPROTO.oneofs_by_name['value'] +_TYPEPROTO.oneofs_by_name['value'].fields.append( + _TYPEPROTO.fields_by_name['map_type']) +_TYPEPROTO.fields_by_name['map_type'].containing_oneof = _TYPEPROTO.oneofs_by_name['value'] +_TYPEPROTO.oneofs_by_name['value'].fields.append( + _TYPEPROTO.fields_by_name['opaque_type']) +_TYPEPROTO.fields_by_name['opaque_type'].containing_oneof = _TYPEPROTO.oneofs_by_name['value'] +_TYPEPROTO.oneofs_by_name['value'].fields.append( + _TYPEPROTO.fields_by_name['sparse_tensor_type']) +_TYPEPROTO.fields_by_name['sparse_tensor_type'].containing_oneof = _TYPEPROTO.oneofs_by_name['value'] +_FUNCTIONPROTO.fields_by_name['status'].enum_type = _OPERATORSTATUS +_FUNCTIONPROTO.fields_by_name['node'].message_type = _NODEPROTO +DESCRIPTOR.message_types_by_name['AttributeProto'] = _ATTRIBUTEPROTO +DESCRIPTOR.message_types_by_name['ValueInfoProto'] = _VALUEINFOPROTO +DESCRIPTOR.message_types_by_name['NodeProto'] = _NODEPROTO +DESCRIPTOR.message_types_by_name['ModelProto'] = _MODELPROTO +DESCRIPTOR.message_types_by_name['StringStringEntryProto'] = _STRINGSTRINGENTRYPROTO +DESCRIPTOR.message_types_by_name['TensorAnnotation'] = _TENSORANNOTATION +DESCRIPTOR.message_types_by_name['GraphProto'] = _GRAPHPROTO +DESCRIPTOR.message_types_by_name['TensorProto'] = _TENSORPROTO +DESCRIPTOR.message_types_by_name['TensorShapeProto'] = _TENSORSHAPEPROTO +DESCRIPTOR.message_types_by_name['TypeProto'] = _TYPEPROTO +DESCRIPTOR.message_types_by_name['OperatorSetIdProto'] = _OPERATORSETIDPROTO +DESCRIPTOR.message_types_by_name['FunctionProto'] = _FUNCTIONPROTO +DESCRIPTOR.enum_types_by_name['Version'] = _VERSION +DESCRIPTOR.enum_types_by_name['OperatorStatus'] = _OPERATORSTATUS +_sym_db.RegisterFileDescriptor(DESCRIPTOR) + +AttributeProto = _reflection.GeneratedProtocolMessageType('AttributeProto', (_message.Message,), dict( + DESCRIPTOR = _ATTRIBUTEPROTO, + __module__ = 'onnx_ml_pb2' + # @@protoc_insertion_point(class_scope:onnx.AttributeProto) + )) +_sym_db.RegisterMessage(AttributeProto) + +ValueInfoProto = _reflection.GeneratedProtocolMessageType('ValueInfoProto', (_message.Message,), dict( + DESCRIPTOR = _VALUEINFOPROTO, + __module__ = 'onnx_ml_pb2' + # @@protoc_insertion_point(class_scope:onnx.ValueInfoProto) + )) +_sym_db.RegisterMessage(ValueInfoProto) + +NodeProto = _reflection.GeneratedProtocolMessageType('NodeProto', (_message.Message,), dict( + DESCRIPTOR = _NODEPROTO, + __module__ = 'onnx_ml_pb2' + # @@protoc_insertion_point(class_scope:onnx.NodeProto) + )) +_sym_db.RegisterMessage(NodeProto) + +ModelProto = _reflection.GeneratedProtocolMessageType('ModelProto', (_message.Message,), dict( + DESCRIPTOR = _MODELPROTO, + __module__ = 'onnx_ml_pb2' + # @@protoc_insertion_point(class_scope:onnx.ModelProto) + )) +_sym_db.RegisterMessage(ModelProto) + +StringStringEntryProto = _reflection.GeneratedProtocolMessageType('StringStringEntryProto', (_message.Message,), dict( + DESCRIPTOR = _STRINGSTRINGENTRYPROTO, + __module__ = 'onnx_ml_pb2' + # @@protoc_insertion_point(class_scope:onnx.StringStringEntryProto) + )) +_sym_db.RegisterMessage(StringStringEntryProto) + +TensorAnnotation = _reflection.GeneratedProtocolMessageType('TensorAnnotation', (_message.Message,), dict( + DESCRIPTOR = _TENSORANNOTATION, + __module__ = 'onnx_ml_pb2' + # @@protoc_insertion_point(class_scope:onnx.TensorAnnotation) + )) +_sym_db.RegisterMessage(TensorAnnotation) + +GraphProto = _reflection.GeneratedProtocolMessageType('GraphProto', (_message.Message,), dict( + DESCRIPTOR = _GRAPHPROTO, + __module__ = 'onnx_ml_pb2' + # @@protoc_insertion_point(class_scope:onnx.GraphProto) + )) +_sym_db.RegisterMessage(GraphProto) + +TensorProto = _reflection.GeneratedProtocolMessageType('TensorProto', (_message.Message,), dict( + + Segment = _reflection.GeneratedProtocolMessageType('Segment', (_message.Message,), dict( + DESCRIPTOR = _TENSORPROTO_SEGMENT, + __module__ = 'onnx_ml_pb2' + # @@protoc_insertion_point(class_scope:onnx.TensorProto.Segment) + )) + , + DESCRIPTOR = _TENSORPROTO, + __module__ = 'onnx_ml_pb2' + # @@protoc_insertion_point(class_scope:onnx.TensorProto) + )) +_sym_db.RegisterMessage(TensorProto) +_sym_db.RegisterMessage(TensorProto.Segment) + +TensorShapeProto = _reflection.GeneratedProtocolMessageType('TensorShapeProto', (_message.Message,), dict( + + Dimension = _reflection.GeneratedProtocolMessageType('Dimension', (_message.Message,), dict( + DESCRIPTOR = _TENSORSHAPEPROTO_DIMENSION, + __module__ = 'onnx_ml_pb2' + # @@protoc_insertion_point(class_scope:onnx.TensorShapeProto.Dimension) + )) + , + DESCRIPTOR = _TENSORSHAPEPROTO, + __module__ = 'onnx_ml_pb2' + # @@protoc_insertion_point(class_scope:onnx.TensorShapeProto) + )) +_sym_db.RegisterMessage(TensorShapeProto) +_sym_db.RegisterMessage(TensorShapeProto.Dimension) + +TypeProto = _reflection.GeneratedProtocolMessageType('TypeProto', (_message.Message,), dict( + + Tensor = _reflection.GeneratedProtocolMessageType('Tensor', (_message.Message,), dict( + DESCRIPTOR = _TYPEPROTO_TENSOR, + __module__ = 'onnx_ml_pb2' + # @@protoc_insertion_point(class_scope:onnx.TypeProto.Tensor) + )) + , + + Sequence = _reflection.GeneratedProtocolMessageType('Sequence', (_message.Message,), dict( + DESCRIPTOR = _TYPEPROTO_SEQUENCE, + __module__ = 'onnx_ml_pb2' + # @@protoc_insertion_point(class_scope:onnx.TypeProto.Sequence) + )) + , + + Map = _reflection.GeneratedProtocolMessageType('Map', (_message.Message,), dict( + DESCRIPTOR = _TYPEPROTO_MAP, + __module__ = 'onnx_ml_pb2' + # @@protoc_insertion_point(class_scope:onnx.TypeProto.Map) + )) + , + + Opaque = _reflection.GeneratedProtocolMessageType('Opaque', (_message.Message,), dict( + DESCRIPTOR = _TYPEPROTO_OPAQUE, + __module__ = 'onnx_ml_pb2' + # @@protoc_insertion_point(class_scope:onnx.TypeProto.Opaque) + )) + , + + SparseTensor = _reflection.GeneratedProtocolMessageType('SparseTensor', (_message.Message,), dict( + DESCRIPTOR = _TYPEPROTO_SPARSETENSOR, + __module__ = 'onnx_ml_pb2' + # @@protoc_insertion_point(class_scope:onnx.TypeProto.SparseTensor) + )) + , + DESCRIPTOR = _TYPEPROTO, + __module__ = 'onnx_ml_pb2' + # @@protoc_insertion_point(class_scope:onnx.TypeProto) + )) +_sym_db.RegisterMessage(TypeProto) +_sym_db.RegisterMessage(TypeProto.Tensor) +_sym_db.RegisterMessage(TypeProto.Sequence) +_sym_db.RegisterMessage(TypeProto.Map) +_sym_db.RegisterMessage(TypeProto.Opaque) +_sym_db.RegisterMessage(TypeProto.SparseTensor) + +OperatorSetIdProto = _reflection.GeneratedProtocolMessageType('OperatorSetIdProto', (_message.Message,), dict( + DESCRIPTOR = _OPERATORSETIDPROTO, + __module__ = 'onnx_ml_pb2' + # @@protoc_insertion_point(class_scope:onnx.OperatorSetIdProto) + )) +_sym_db.RegisterMessage(OperatorSetIdProto) + +FunctionProto = _reflection.GeneratedProtocolMessageType('FunctionProto', (_message.Message,), dict( + DESCRIPTOR = _FUNCTIONPROTO, + __module__ = 'onnx_ml_pb2' + # @@protoc_insertion_point(class_scope:onnx.FunctionProto) + )) +_sym_db.RegisterMessage(FunctionProto) + + +_TENSORPROTO.fields_by_name['float_data']._options = None +_TENSORPROTO.fields_by_name['int32_data']._options = None +_TENSORPROTO.fields_by_name['int64_data']._options = None +_TENSORPROTO.fields_by_name['double_data']._options = None +_TENSORPROTO.fields_by_name['uint64_data']._options = None +# @@protoc_insertion_point(module_scope) diff --git a/docs/modelserving/v1beta1/onnx/assets/predict_pb2.py b/docs/modelserving/v1beta1/onnx/assets/predict_pb2.py new file mode 100644 index 000000000..c71c138ac --- /dev/null +++ b/docs/modelserving/v1beta1/onnx/assets/predict_pb2.py @@ -0,0 +1,215 @@ +# Generated by the protocol buffer compiler. DO NOT EDIT! +# source: predict.proto + +import sys +_b=sys.version_info[0]<3 and (lambda x:x) or (lambda x:x.encode('latin1')) +from google.protobuf import descriptor as _descriptor +from google.protobuf import message as _message +from google.protobuf import reflection as _reflection +from google.protobuf import symbol_database as _symbol_database +# @@protoc_insertion_point(imports) + +_sym_db = _symbol_database.Default() + + +import assets.onnx_ml_pb2 as onnx__ml__pb2 + + +DESCRIPTOR = _descriptor.FileDescriptor( + name='predict.proto', + package='onnxruntime.server', + syntax='proto3', + serialized_options=None, + serialized_pb=_b('\n\rpredict.proto\x12\x12onnxruntime.server\x1a\ronnx-ml.proto\"\xaf\x01\n\x0ePredictRequest\x12>\n\x06inputs\x18\x02 \x03(\x0b\x32..onnxruntime.server.PredictRequest.InputsEntry\x12\x15\n\routput_filter\x18\x03 \x03(\t\x1a@\n\x0bInputsEntry\x12\x0b\n\x03key\x18\x01 \x01(\t\x12 \n\x05value\x18\x02 \x01(\x0b\x32\x11.onnx.TensorProto:\x02\x38\x01J\x04\x08\x01\x10\x02\"\x97\x01\n\x0fPredictResponse\x12\x41\n\x07outputs\x18\x01 \x03(\x0b\x32\x30.onnxruntime.server.PredictResponse.OutputsEntry\x1a\x41\n\x0cOutputsEntry\x12\x0b\n\x03key\x18\x01 \x01(\t\x12 \n\x05value\x18\x02 \x01(\x0b\x32\x11.onnx.TensorProto:\x02\x38\x01\x62\x06proto3') + , + dependencies=[onnx__ml__pb2.DESCRIPTOR,]) + + + + +_PREDICTREQUEST_INPUTSENTRY = _descriptor.Descriptor( + name='InputsEntry', + full_name='onnxruntime.server.PredictRequest.InputsEntry', + filename=None, + file=DESCRIPTOR, + containing_type=None, + fields=[ + _descriptor.FieldDescriptor( + name='key', full_name='onnxruntime.server.PredictRequest.InputsEntry.key', index=0, + number=1, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=_b("").decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='value', full_name='onnxruntime.server.PredictRequest.InputsEntry.value', index=1, + number=2, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=_b('8\001'), + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=158, + serialized_end=222, +) + +_PREDICTREQUEST = _descriptor.Descriptor( + name='PredictRequest', + full_name='onnxruntime.server.PredictRequest', + filename=None, + file=DESCRIPTOR, + containing_type=None, + fields=[ + _descriptor.FieldDescriptor( + name='inputs', full_name='onnxruntime.server.PredictRequest.inputs', index=0, + number=2, type=11, cpp_type=10, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='output_filter', full_name='onnxruntime.server.PredictRequest.output_filter', index=1, + number=3, type=9, cpp_type=9, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + ], + extensions=[ + ], + nested_types=[_PREDICTREQUEST_INPUTSENTRY, ], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=53, + serialized_end=228, +) + + +_PREDICTRESPONSE_OUTPUTSENTRY = _descriptor.Descriptor( + name='OutputsEntry', + full_name='onnxruntime.server.PredictResponse.OutputsEntry', + filename=None, + file=DESCRIPTOR, + containing_type=None, + fields=[ + _descriptor.FieldDescriptor( + name='key', full_name='onnxruntime.server.PredictResponse.OutputsEntry.key', index=0, + number=1, type=9, cpp_type=9, label=1, + has_default_value=False, default_value=_b("").decode('utf-8'), + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + _descriptor.FieldDescriptor( + name='value', full_name='onnxruntime.server.PredictResponse.OutputsEntry.value', index=1, + number=2, type=11, cpp_type=10, label=1, + has_default_value=False, default_value=None, + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + ], + extensions=[ + ], + nested_types=[], + enum_types=[ + ], + serialized_options=_b('8\001'), + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=317, + serialized_end=382, +) + +_PREDICTRESPONSE = _descriptor.Descriptor( + name='PredictResponse', + full_name='onnxruntime.server.PredictResponse', + filename=None, + file=DESCRIPTOR, + containing_type=None, + fields=[ + _descriptor.FieldDescriptor( + name='outputs', full_name='onnxruntime.server.PredictResponse.outputs', index=0, + number=1, type=11, cpp_type=10, label=3, + has_default_value=False, default_value=[], + message_type=None, enum_type=None, containing_type=None, + is_extension=False, extension_scope=None, + serialized_options=None, file=DESCRIPTOR), + ], + extensions=[ + ], + nested_types=[_PREDICTRESPONSE_OUTPUTSENTRY, ], + enum_types=[ + ], + serialized_options=None, + is_extendable=False, + syntax='proto3', + extension_ranges=[], + oneofs=[ + ], + serialized_start=231, + serialized_end=382, +) + +_PREDICTREQUEST_INPUTSENTRY.fields_by_name['value'].message_type = onnx__ml__pb2._TENSORPROTO +_PREDICTREQUEST_INPUTSENTRY.containing_type = _PREDICTREQUEST +_PREDICTREQUEST.fields_by_name['inputs'].message_type = _PREDICTREQUEST_INPUTSENTRY +_PREDICTRESPONSE_OUTPUTSENTRY.fields_by_name['value'].message_type = onnx__ml__pb2._TENSORPROTO +_PREDICTRESPONSE_OUTPUTSENTRY.containing_type = _PREDICTRESPONSE +_PREDICTRESPONSE.fields_by_name['outputs'].message_type = _PREDICTRESPONSE_OUTPUTSENTRY +DESCRIPTOR.message_types_by_name['PredictRequest'] = _PREDICTREQUEST +DESCRIPTOR.message_types_by_name['PredictResponse'] = _PREDICTRESPONSE +_sym_db.RegisterFileDescriptor(DESCRIPTOR) + +PredictRequest = _reflection.GeneratedProtocolMessageType('PredictRequest', (_message.Message,), dict( + + InputsEntry = _reflection.GeneratedProtocolMessageType('InputsEntry', (_message.Message,), dict( + DESCRIPTOR = _PREDICTREQUEST_INPUTSENTRY, + __module__ = 'predict_pb2' + # @@protoc_insertion_point(class_scope:onnxruntime.server.PredictRequest.InputsEntry) + )) + , + DESCRIPTOR = _PREDICTREQUEST, + __module__ = 'predict_pb2' + # @@protoc_insertion_point(class_scope:onnxruntime.server.PredictRequest) + )) +_sym_db.RegisterMessage(PredictRequest) +_sym_db.RegisterMessage(PredictRequest.InputsEntry) + +PredictResponse = _reflection.GeneratedProtocolMessageType('PredictResponse', (_message.Message,), dict( + + OutputsEntry = _reflection.GeneratedProtocolMessageType('OutputsEntry', (_message.Message,), dict( + DESCRIPTOR = _PREDICTRESPONSE_OUTPUTSENTRY, + __module__ = 'predict_pb2' + # @@protoc_insertion_point(class_scope:onnxruntime.server.PredictResponse.OutputsEntry) + )) + , + DESCRIPTOR = _PREDICTRESPONSE, + __module__ = 'predict_pb2' + # @@protoc_insertion_point(class_scope:onnxruntime.server.PredictResponse) + )) +_sym_db.RegisterMessage(PredictResponse) +_sym_db.RegisterMessage(PredictResponse.OutputsEntry) + + +_PREDICTREQUEST_INPUTSENTRY._options = None +_PREDICTRESPONSE_OUTPUTSENTRY._options = None +# @@protoc_insertion_point(module_scope) diff --git a/docs/modelserving/v1beta1/onnx/image.jpg b/docs/modelserving/v1beta1/onnx/image.jpg new file mode 100644 index 000000000..88e8a50a6 Binary files /dev/null and b/docs/modelserving/v1beta1/onnx/image.jpg differ diff --git a/docs/modelserving/v1beta1/onnx/mosaic-onnx.ipynb b/docs/modelserving/v1beta1/onnx/mosaic-onnx.ipynb new file mode 100644 index 000000000..4e1d44ce3 --- /dev/null +++ b/docs/modelserving/v1beta1/onnx/mosaic-onnx.ipynb @@ -0,0 +1,197 @@ +{ + "cells": [ + { + "attachments": {}, + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# Predict an ONNX InferenceService" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "metadata": {}, + "source": [ + "This example assumes you have already deployed the sample ONNX Inference Service. \n", + "\n", + "Deploy the sample ONNX InferenceSevice by following the instructions in the [README](https://github.com/kserve/kserve/blob/master/docs/samples/v1beta1/onnx/README.md)" + ] + }, + { + "cell_type": "code", + "execution_count": 26, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "env: MODEL_NAME=style\n", + "env: INGRESS_HOST=localhost\n", + "env: INGRESS_PORT=8080\n", + "env: SERVICE_HOSTNAME=style-sample.default.example.com\n" + ] + } + ], + "source": [ + "\n", + "%env MODEL_NAME=style\n", + "HOSTNAME=!(kubectl get inferenceservice \"style-sample\" -o jsonpath='{.status.url}' | cut -d \"/\" -f 3)\n", + "%env INGRESS_HOST=localhost\n", + "%env INGRESS_PORT=8080\n", + "%env SERVICE_HOSTNAME={HOSTNAME[0]}" + ] + }, + { + "cell_type": "code", + "execution_count": 29, + "metadata": {}, + "outputs": [], + "source": [ + "from PIL import Image\n", + "import numpy as np\n", + "import requests\n", + "import json\n", + "import os" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# load & resize image\n", + "image = Image.open(\"image.jpg\")\n", + "image = image.resize((224,224), Image.LANCZOS)\n", + "image" + ] + }, + { + "cell_type": "code", + "execution_count": 30, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "(1, 3, 224, 224)" + ] + }, + "execution_count": 30, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "# preprocess image data\n", + "norm_img_data = np.array(image).astype('float32')\n", + "norm_img_data = np.transpose(norm_img_data, [2, 0, 1])\n", + "norm_img_data = np.expand_dims(norm_img_data, axis=0)\n", + "np.shape(norm_img_data)" + ] + }, + { + "cell_type": "code", + "execution_count": 31, + "metadata": {}, + "outputs": [], + "source": [ + "# Create request message to be sent to the predictor\n", + "message_data = {}\n", + "inputs = {}\n", + "message_data[\"inputs\"] = []\n", + "inputs[\"name\"]=\"input1\"\n", + "inputs[\"shape\"]=norm_img_data.shape\n", + "inputs[\"datatype\"]=\"FP32\" # as the given onnx model expects float32\n", + "inputs[\"data\"]=norm_img_data.tolist()\n", + "message_data[\"inputs\"].append(inputs)" + ] + }, + { + "cell_type": "code", + "execution_count": 32, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "200\n" + ] + } + ], + "source": [ + "# Call predictor\n", + "\n", + "service_hostname=os.environ[\"SERVICE_HOSTNAME\"]\n", + "model_name=os.environ[\"MODEL_NAME\"]\n", + "ingress_ip=\"localhost\"\n", + "ingress_port=os.environ[\"INGRESS_PORT\"]\n", + "predictor_url = f\"http://{ingress_ip}:{ingress_port}/v2/models/{model_name}/infer\"\n", + "request_headers = {'Content-Type': 'application/json', 'Accept': 'application/json', 'Host': service_hostname}\n", + "response = requests.post(predictor_url, headers=request_headers, data=json.dumps(message_data))\n", + "print(response.status_code)" + ] + }, + { + "cell_type": "code", + "execution_count": 33, + "metadata": {}, + "outputs": [], + "source": [ + "response_message = json.loads(response.text)\n", + "output1 = np.array(response_message[\"outputs\"][0]['data'], dtype=np.float32)\n", + "output1 = output1.reshape(3,224,224)" + ] + }, + { + "cell_type": "code", + "execution_count": 34, + "metadata": {}, + "outputs": [ + { + "data": { + "image/jpeg": "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", + "image/png": "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", + "text/plain": [ + "" + ] + }, + "execution_count": 34, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "# postprocess\n", + "result = np.clip(output1, 0, 255)\n", + "result = result.transpose(1,2,0).astype(\"uint8\")\n", + "img = Image.fromarray(result)\n", + "img" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.7.0" + } + }, + "nbformat": 4, + "nbformat_minor": 2 +} diff --git a/docs/modelserving/v1beta1/onnx/onnx.yaml b/docs/modelserving/v1beta1/onnx/onnx.yaml new file mode 100644 index 000000000..319ab4bdb --- /dev/null +++ b/docs/modelserving/v1beta1/onnx/onnx.yaml @@ -0,0 +1,11 @@ +apiVersion: "serving.kserve.io/v1beta1" +kind: "InferenceService" +metadata: + name: "style-sample" +spec: + predictor: + model: + protocolVersion: v2 + modelFormat: + name: onnx + storageUri: "gs://kfserving-examples/models/onnx" diff --git a/docs/modelserving/v1beta1/onnx/requirements.txt b/docs/modelserving/v1beta1/onnx/requirements.txt new file mode 100644 index 000000000..26ba69045 --- /dev/null +++ b/docs/modelserving/v1beta1/onnx/requirements.txt @@ -0,0 +1,5 @@ +jupyter +numpy +pillow +protobuf +requests \ No newline at end of file diff --git a/docs/modelserving/v1beta1/sklearn/v2/README.md b/docs/modelserving/v1beta1/sklearn/v2/README.md index 54a439fc4..fc41264f8 100644 --- a/docs/modelserving/v1beta1/sklearn/v2/README.md +++ b/docs/modelserving/v1beta1/sklearn/v2/README.md @@ -85,6 +85,14 @@ For this, you will just need to use **version `v1beta1`** of the !!! Note For `V2 protocol (open inference protocol)` if `runtime` field is not provided then, by default `mlserver` runtime is used. +Note that this makes the following assumptions: + +- Your model weights (i.e. your `model.joblib` file) have already been uploaded + to a "model repository" (GCS in this example) and can be accessed as + `gs://seldon-models/sklearn/iris`. +- There is a K8s cluster available, accessible through `kubectl`. +- KServe has already been [installed in your cluster](../../../../get_started/README.md). + === "kubectl" ```bash diff --git a/mkdocs.yml b/mkdocs.yml index fd5d9d982..2b67d8a82 100644 --- a/mkdocs.yml +++ b/mkdocs.yml @@ -34,6 +34,7 @@ nav: - LightGBM: modelserving/v1beta1/lightgbm/README.md - Paddle: modelserving/v1beta1/paddle/README.md - MLFlow: modelserving/v1beta1/mlflow/v2/README.md + - ONNX: modelserving/v1beta1/onnx/README.md - Multi-Framework Serving Runtimes: - Nvidia Triton: - Torchscript: modelserving/v1beta1/triton/torchscript/README.md