ONNX
59 public types in this namespace, organized by kind.
Models & Types (29)
| Type | Summary |
|---|---|
AttributeProto |
Attributes A named attribute containing either singular float, integer, string, graph, and tensor values, or repeated float, integer, string, graph, and tensor values. |
Dimension |
|
FunctionProto |
|
GraphProto |
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. |
Map |
map<K,V> |
ModelProto |
Models ModelProto is a top-level file/container format for bundling a ML model and associating its computation graph with metadata. |
NodeProto |
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. |
OnnxGraphSignature |
An immutable I/O signature for one graph, without guessed native architecture values. |
OnnxLayerInputs |
Named tensors flowing INTO a layer’s ONNX node(s). |
OnnxLayerOutputs |
Named tensors flowing OUT of a layer’s ONNX node(s). |
OnnxModelDownloader |
Downloads ONNX models from HuggingFace Hub and other repositories. |
OnnxModelMetadata |
Metadata about a loaded ONNX model. |
OnnxModel<T> |
A wrapper for ONNX models that provides easy-to-use inference with AiDotNet Tensor types. |
OnnxTensorInfo |
Information about an ONNX tensor (input or output). |
OnnxValueSignature |
An immutable I/O value signature read from a loaded ONNX graph. |
OperatorSetIdProto |
Operator Sets OperatorSets are uniquely identified by a (domain, opset_version) pair. |
Optional |
wrapper for Tensor, Sequence, or Map |
Segment |
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. |
Sequence |
repeated T |
SparseTensor |
|
SparseTensorProto |
A serialized sparse-tensor value |
StringStringEntryProto |
StringStringEntryProto follows the pattern for cross-proto-version maps. |
Tensor |
|
TensorAnnotation |
|
TensorProto |
Tensors A serialized tensor value. |
TensorShapeProto |
Defines a tensor shape. |
TrainingInfoProto |
Training information TrainingInfoProto stores information for training a model. |
TypeProto |
Types The standard ONNX data types. |
ValueInfoProto |
Defines information on value, including the name, the type, and the shape of the value. |
Enums (11)
| Type | Summary |
|---|---|
AttributeType |
Note: this enum is structurally identical to the OpSchema::AttrType enum defined in schema.h. |
DataLocation |
Location of the data for this tensor. |
DataType |
|
GraphOptimizationLevel |
Graph optimization levels for ONNX Runtime. |
OnnxExecutionProvider |
Specifies the execution provider (hardware accelerator) for ONNX model inference. |
OnnxLogLevel |
Log severity levels for ONNX Runtime. |
OnnxModelRole |
Identifies the graph’s role in a multimodal model. |
OperatorStatus |
Operator/function status. |
ValueOneofCase |
Enum of possible cases for the “value” oneof. |
ValueOneofCase |
Enum of possible cases for the “value” oneof. |
Version |
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. |
Structs (1)
| Type | Summary |
|---|---|
OnnxAxisSpec |
Per-axis shape descriptor for an ONNX TensorShapeProto.Dimension. |
Options & Configuration (3)
| Type | Summary |
|---|---|
OnnxExportOptions |
Optional configuration for ONNX export. |
OnnxModelOptions |
Configuration options for loading and running ONNX models. |
OnnxMultimodalConfiguration |
The effective host dimensions and immutable signatures of loaded multimodal graphs. |
Helpers & Utilities (14)
| Type | Summary |
|---|---|
AiModelResultOnnxExtensions |
Public-facing ONNX export API on AiModelResult. |
AudioGen |
Audio generation models. |
OnnxExporter |
Exports AiDotNet models to the ONNX format. |
OnnxGraphBuilder |
Thin facade over the vendored ONNX protobuf types that lets layer converters add nodes, initializers, inputs, and outputs to a model graph without touching the generated GraphProto directly. |
OnnxModelBuilder |
Internal builder for constructing ONNX model bytes. |
OnnxModelRepositories |
Common ONNX model repositories and their identifiers. |
OnnxReflection |
Holder for reflection information generated from onnx.proto3 |
OnnxTensorConverter |
Converts between AiDotNet Tensor types and ONNX Runtime tensor types. |
Tts |
Text-to-speech models. |
Types |
Container for nested types declared in the AttributeProto message type. |
Types |
Container for nested types declared in the TensorProto message type. |
Types |
Container for nested types declared in the TensorShapeProto message type. |
Types |
Container for nested types declared in the TypeProto message type. |
Whisper |
OpenAI Whisper speech recognition models. |
Exceptions (1)
| Type | Summary |
|---|---|
OnnxExportUnsupportedException |
Thrown when a model contains a layer or component that does not yet have an ONNX export converter. |