Attributes
A named attribute containing either singular float, integer, string, graph,
and tensor values, or repeated float, integer, string, graph, and tensor values.
Note: this enum is structurally identical to the OpSchema::AttrType
enum defined in schema.h.
Attributes
A named attribute containing either singular float, integer, string, graph,
and tensor values, or repeated float, integer, string, graph, and tensor values.
Protobuf type onnx.FunctionProto
Protobuf type onnx.FunctionProto
Graphs
A graph defines the computational logic of a model and is comprised of a parameterized
list of nodes that form a directed acyclic graph based on their inputs and outputs.
Graphs
A graph defines the computational logic of a model and is comprised of a parameterized
list of nodes that form a directed acyclic graph based on their inputs and outputs.
Models
ModelProto is a top-level file/container format for bundling a ML model and
associating its computation graph with metadata.
Models
ModelProto is a top-level file/container format for bundling a ML model and
associating its computation graph with metadata.
Nodes
Computation graphs are made up of a DAG of nodes, which represent what is
commonly called a "layer" or "pipeline stage" in machine learning frameworks.
Nodes
Computation graphs are made up of a DAG of nodes, which represent what is
commonly called a "layer" or "pipeline stage" in machine learning frameworks.
Operator Sets
OperatorSets are uniquely identified by a (domain, opset_version) pair.
Operator Sets
OperatorSets are uniquely identified by a (domain, opset_version) pair.
Operator/function status.
A serialized sparse-tensor value
A serialized sparse-tensor value
StringStringEntryProto follows the pattern for cross-proto-version maps.
StringStringEntryProto follows the pattern for cross-proto-version maps.
Protobuf type onnx.TensorAnnotation
Protobuf type onnx.TensorAnnotation
Tensors
A serialized tensor value.
Tensors
A serialized tensor value.
Location of the data for this tensor.
Protobuf enum onnx.TensorProto.DataType
For very large tensors, we may want to store them in chunks, in which
case the following fields will specify the segment that is stored in
the current TensorProto.
For very large tensors, we may want to store them in chunks, in which
case the following fields will specify the segment that is stored in
the current TensorProto.
Protobuf type onnx.TensorShapeProto.Dimension
Protobuf type onnx.TensorShapeProto.Dimension
Training information
TrainingInfoProto stores information for training a model.
Training information
TrainingInfoProto stores information for training a model.
Types
The standard ONNX data types.
Types
The standard ONNX data types.
wrapper for Tensor, Sequence, or Map
wrapper for Tensor, Sequence, or Map
Protobuf type onnx.TypeProto.SparseTensor
Protobuf type onnx.TypeProto.SparseTensor
Protobuf type onnx.TypeProto.Tensor
Protobuf type onnx.TypeProto.Tensor
Defines information on value, including the name, the type, and
the shape of the value.
Defines information on value, including the name, the type, and
the shape of the value.
Versioning
ONNX versioning is specified in docs/IR.md and elaborated on in docs/Versioning.md
To be compatible with both proto2 and proto3, we will use a version number
that is not defined by the default value but an explicit enum number.