Generated using TypeDoc
Set or get a number specifying the timeout for initialization of WebAssembly backend, in milliseconds. A zero +value indicates no timeout is set.
+set or get number of thread(s). If omitted or set to 0, number of thread(s) will be determined by system. If set +to 1, no worker thread will be spawned.
+This setting is available only when WebAssembly multithread feature is available in current context.
+Set or get a boolean value indicating whether to proxy the execution of main thread to a worker thread.
+set or get a boolean value indicating whether to enable SIMD. If set to false, SIMD will be forcely disabled.
+This setting is available only when WebAssembly SIMD feature is available in current context.
+Set a custom URL prefix to the .wasm files or a set of overrides for each .wasm file. The override path should be +an absolute path.
+Generated using TypeDoc
Set or get whether enable async download.
+Set or get the WebGL Context ID (webgl or webgl2).
+Set or get the maximum batch size for matmul. 0 means to disable batching.
+Set or get the packed texture mode
+Set or get the texture cache mode.
+Generated using TypeDoc
Generated using TypeDoc
Generated using TypeDoc
Generated using TypeDoc
Generated using TypeDoc
Generated using TypeDoc
A set of configurations for inference run behavior
+Set a single run configuration entry. See +https://github.com/microsoft/onnxruntime/blob/master/include/onnxruntime/core/session/ +onnxruntime_run_options_config_keys.h
+This setting is available only in WebAssembly backend. Will support Node.js binding and react-native later
+Log severity level. See +https://github.com/microsoft/onnxruntime/blob/master/include/onnxruntime/core/common/logging/severity.h
+This setting is available only in ONNXRuntime (Node.js binding and react-native) or WebAssembly backend
+Log verbosity level.
+This setting is available only in WebAssembly backend. Will support Node.js binding and react-native later
+A tag for the Run() calls using this
+This setting is available only in ONNXRuntime (Node.js binding and react-native) or WebAssembly backend
+Terminate all incomplete OrtRun calls as soon as possible if true
+This setting is available only in WebAssembly backend. Will support Node.js binding and react-native later
+Generated using TypeDoc
A set of configurations for session behavior.
+Whether enable CPU memory arena.
+This setting is available only in ONNXRuntime (Node.js binding and react-native) or WebAssembly backend
+Whether enable memory pattern.
+This setting is available only in ONNXRuntime (Node.js binding and react-native) or WebAssembly backend
+Wether enable profiling.
+This setting is a placeholder for a future use.
+Execution mode.
+This setting is available only in ONNXRuntime (Node.js binding and react-native) or WebAssembly backend
+An array of execution provider options.
+An execution provider option can be a string indicating the name of the execution provider, +or an object of corresponding type.
+Store configurations for a session. See +https://github.com/microsoft/onnxruntime/blob/master/include/onnxruntime/core/session/ +onnxruntime_session_options_config_keys.h
+This setting is available only in WebAssembly backend. Will support Node.js binding and react-native later
+The optimization level.
+This setting is available only in ONNXRuntime (Node.js binding and react-native) or WebAssembly backend
+The inter OP threads number.
+This setting is available only in ONNXRuntime (Node.js binding and react-native).
+The intra OP threads number.
+This setting is available only in ONNXRuntime (Node.js binding and react-native).
+Log ID.
+This setting is available only in ONNXRuntime (Node.js binding and react-native) or WebAssembly backend
+Log severity level. See +https://github.com/microsoft/onnxruntime/blob/master/include/onnxruntime/core/common/logging/severity.h
+This setting is available only in ONNXRuntime (Node.js binding and react-native) or WebAssembly backend
+Log verbosity level.
+This setting is available only in WebAssembly backend. Will support Node.js binding and react-native later
+File prefix for profiling.
+This setting is a placeholder for a future use.
+Generated using TypeDoc
Generated using TypeDoc
Generated using TypeDoc
Generated using TypeDoc
Represent a runtime instance of an ONNX model.
+Get input names of the loaded model.
+Get output names of the loaded model.
+End profiling.
+Execute the model asynchronously with the given feeds and options.
+Representation of the model input. See type description of InferenceSession.InputType for detail.
Optional. A set of options that controls the behavior of model inference.
+A promise that resolves to a map, which uses output names as keys and OnnxValue as corresponding values.
+Execute the model asynchronously with the given feeds, fetches and options.
+Representation of the model input. See type description of InferenceSession.InputType for detail.
Representation of the model output. See type description of InferenceSession.OutputType for
+detail.
Optional. A set of options that controls the behavior of model inference.
+A promise that resolves to a map, which uses output names as keys and OnnxValue as corresponding values.
+Start profiling.
+Generated using TypeDoc
Create a new inference session and load model asynchronously from an ONNX model file.
+The URI or file path of the model to load.
+specify configuration for creating a new inference session.
+A promise that resolves to an InferenceSession object.
+Create a new inference session and load model asynchronously from an array bufer.
+An ArrayBuffer representation of an ONNX model.
+specify configuration for creating a new inference session.
+A promise that resolves to an InferenceSession object.
+Create a new inference session and load model asynchronously from segment of an array bufer.
+An ArrayBuffer representation of an ONNX model.
+The beginning of the specified portion of the array buffer.
+The length in bytes of the array buffer.
+specify configuration for creating a new inference session.
+A promise that resolves to an InferenceSession object.
+Create a new inference session and load model asynchronously from a Uint8Array.
+A Uint8Array representation of an ONNX model.
+specify configuration for creating a new inference session.
+A promise that resolves to an InferenceSession object.
+Generated using TypeDoc
Generated using TypeDoc
Generated using TypeDoc
Represent multi-dimensional arrays to feed to or fetch from model inferencing.
+Get the buffer data of the tensor.
+Get the dimensions of the tensor.
+Get the number of elements in the tensor.
+Get the data type of the tensor.
+Create a new tensor with the same data buffer and specified dims.
+New dimensions. Size should match the old one.
+Generated using TypeDoc
Construct a new string tensor object from the given type, data and dims.
+Specify the element type.
+Specify the tensor data
+Specify the dimension of the tensor. If omitted, a 1-D tensor is assumed.
+Construct a new bool tensor object from the given type, data and dims.
+Specify the element type.
+Specify the tensor data
+Specify the dimension of the tensor. If omitted, a 1-D tensor is assumed.
+Construct a new numeric tensor object from the given type, data and dims.
+Specify the element type.
+Specify the tensor data
+Specify the dimension of the tensor. If omitted, a 1-D tensor is assumed.
+Construct a new float32 tensor object from the given data and dims.
+Specify the tensor data
+Specify the dimension of the tensor. If omitted, a 1-D tensor is assumed.
+Construct a new int8 tensor object from the given data and dims.
+Specify the tensor data
+Specify the dimension of the tensor. If omitted, a 1-D tensor is assumed.
+Construct a new uint8 tensor object from the given data and dims.
+Specify the tensor data
+Specify the dimension of the tensor. If omitted, a 1-D tensor is assumed.
+Construct a new uint16 tensor object from the given data and dims.
+Specify the tensor data
+Specify the dimension of the tensor. If omitted, a 1-D tensor is assumed.
+Construct a new int16 tensor object from the given data and dims.
+Specify the tensor data
+Specify the dimension of the tensor. If omitted, a 1-D tensor is assumed.
+Construct a new int32 tensor object from the given data and dims.
+Specify the tensor data
+Specify the dimension of the tensor. If omitted, a 1-D tensor is assumed.
+Construct a new int64 tensor object from the given data and dims.
+Specify the tensor data
+Specify the dimension of the tensor. If omitted, a 1-D tensor is assumed.
+Construct a new string tensor object from the given data and dims.
+Specify the tensor data
+Specify the dimension of the tensor. If omitted, a 1-D tensor is assumed.
+Construct a new bool tensor object from the given data and dims.
+Specify the tensor data
+Specify the dimension of the tensor. If omitted, a 1-D tensor is assumed.
+Construct a new float64 tensor object from the given data and dims.
+Specify the tensor data
+Specify the dimension of the tensor. If omitted, a 1-D tensor is assumed.
+Construct a new uint32 tensor object from the given data and dims.
+Specify the tensor data
+Specify the dimension of the tensor. If omitted, a 1-D tensor is assumed.
+Construct a new uint64 tensor object from the given data and dims.
+Specify the tensor data
+Specify the dimension of the tensor. If omitted, a 1-D tensor is assumed.
+Construct a new tensor object from the given type, data and dims.
+Specify the element type.
+Specify the tensor data
+Specify the dimension of the tensor. If omitted, a 1-D tensor is assumed.
+Construct a new tensor object from the given data and dims.
+Specify the tensor data
+Specify the dimension of the tensor. If omitted, a 1-D tensor is assumed.
+Generated using TypeDoc
Represent multi-dimensional arrays to feed to or fetch from model inferencing.
+Get the buffer data of the tensor.
+Get the dimensions of the tensor.
+Get the number of elements in the tensor.
+Get the data type of the tensor.
+Create a new tensor with the same data buffer and specified dims.
+New dimensions. Size should match the old one.
+Generated using TypeDoc
Generated using TypeDoc
A feeds (model inputs) is an object that uses input names as keys and OnnxValue as corresponding values.
+A fetches (model outputs) could be one of the following:
+A inferencing return type is an object that uses output names as keys and OnnxValue as corresponding values.
+Generated using TypeDoc
Generated using TypeDoc
ONNX Runtime JavaScript API
+ +ONNX Runtime JavaScript API is a unified API for all JavaScript usages, including the following NPM packages:
++- onnxruntime-node
+- onnxruntime-web
+- onnxruntime-react-native
+
+See also:
++- Get Started
+- Inference examples
+
+