Web Neural Network API

1. Introduction

The Web Neural Network API defines a web-friendly hardware-agnostic abstraction layer that makes use of Machine Learning capabilities of operating systems and underlying hardware platforms without being tied to platform-specific capabilities. The abstraction layer addresses the requirements of key Machine Learning JavaScript frameworks and also allows web developers familiar with the ML domain to write custom code without the help of libraries.

For an illustrated introduction, please see the explainer.

2. Use cases

2.1. Application Use Cases

This section illustrates application-level use cases for neural network inference hardware acceleration. All applications in those use cases can be built on top of pre-trained deep neural network (DNN) [models].

Note: Please be aware that some of the use cases described here, are by their very nature, privacy-invasive. Developers who are planning to use the API for such use cases should ensure that the API is being used to benefit users, for purposes that users understand, and approve. They should apply the Ethical Principles for Web Machine Learning [webmachinelearning-ethics] and implement appropriate privacy risk mitigations such as transparency, data minimisation, and users controls.

Note: § 3 Accessibility Considerations provides guidance on how to improve accessibility of these use cases.

2.1.1. Person Detection

A user opens a web-based video conferencing application, but she temporarily leaves from her room. The application is watching whether she is in front of her PC by using object detection (for example, using object detection approaches such as [SSD] or [YOLO] that use a single DNN) to detect regions in a camera input frame that include persons.

When she comes back, the application automatically detects her and notifies other online users that she is active now.

2.1.2. Semantic Segmentation

A user joins a teleconference via a web-based video conferencing application at her desk since no meeting room in her office is available. During the teleconference, she does not wish that her room and people in the background are visible. To protect the privacy of the other people and the surroundings, the application runs a machine learning model such as [DeepLabv3+], [MaskR-CNN] or [SegAny] to semantically split an image into segments and replaces segments that represent other people and background with another picture.

2.1.3. Skeleton Detection

A web-based video conferencing application tracks a pose of user’s skeleton by running a machine learning model, which allows for real-time human pose estimation, such as [PoseNet] to recognize her gesture and body language. When she raises her hand, her microphone is automatically unmuted and she can start speaking on the teleconference.

2.1.4. Face Recognition

There are multiple people in the conference room and they join an online meeting using a web-based video conferencing application. The application detects faces of participants by using object detection (for example, using object detection approaches such as [SSD]) and checks whether each face was present at the previous meeting or not by running a machine learning model such as [FaceNet], which verifies whether two faces would be identical or not.

2.1.5. Facial Landmark Detection

A user wants to find new glasses that beautifully fits her on an online glasses store. The online store offers web-based try-on simulator that runs a machine learning model such as Face Alignment Network [FAN] to detect facial landmarks like eyes, nose, mouth, etc. When she chooses a pair of glasses, the simulator properly renders the selected glasses on the detected position of eyes on her facial image.

2.1.6. Style Transfer

A user is looking for cosmetics on an online store and wondering which color may fit her face. The online store shows sample facial makeup images of cosmetics, and offers makeup simulator that runs a machine learning model like [ContextualLoss] or [PairedCycleGAN] to transfer the makeup style of the sample makeup image to her facial image. She can check how the selected makeup looks like on her face by the simulator.

2.1.7. Super Resolution

A web-based video conferencing is receiving a video stream from its peer, but the resolution of the video becomes lower due to network congestion. To prevent degradation of the perceived video quality, the application runs a machine learning model for super-resolution such as [SRGAN] to generate higher-resolution video frames.

2.1.8. Image Captioning

For better accessibility, a web-based presentation application provides automatic image captioning by running a machine learning model such as [im2txt] which predicts explanatory words of the presentation slides.

2.1.9. Text-to-image

Images are a core part of modern web experiences. An ability to generate images based on text input in a privacy-preserving manner enables visual personalization and adaptation of web applications and content. For example, a web application can use as an input a natural language description on the web page or a description provided by the user within a text prompt to produce an image matching the text description. This text-to-image use case enabled by latent diffusion model architecture [LDM] forms the basis for additional text-to-image use cases. For example, inpainting where a portion of an existing image on the web page is selectively modified using the newly generated content, or the converse, outpainting, where an original image is extended beyond its original dimensions filling the empty space with generated content.

2.1.10. Machine Translation

Multiple people from various countries are talking via a web-based real-time text chat application. The application translates their conversation by using a machine learning model such as [GNMT] or [OpenNMT], which translates every text into different language.

2.1.11. Emotion Analysis

A user is talking to her friend via a web-based real-time text chat application, and she is wondering how the friend feels because she cannot see the friend’s face. The application analyses the friend’s emotion by using a machine learning model such as [DeepMoji], which infers emotion from input texts, and displays an emoji that represents the estimated emotion.

2.1.12. Video Summarization

A web-based video conferencing application records received video streams, and it needs to reduce recorded video data to be stored. The application generates the short version of the recorded video by using a machine learning model for video summarization such as [Video-Summarization-with-LSTM].

2.1.13. Noise Suppression

A web-based video conferencing application records received audio streams, but usually the background noise is everywhere. The application leverages real-time noise suppression using Recurrent Neural Network such as [RNNoise] for suppressing background dynamic noise like baby cry or dog barking to improve audio experiences in video conferences.

2.1.14. Speech Recognition

Speech recognition, also known as speech to text, enables recognition and translation of spoken language into text. Example applications of speech recognition include transcription, automatic translation, multimodal interaction, real-time captioning and virtual assistants. Speech recognition improves accessibility of auditory content and makes it possible to interact with such content in a privacy-preserving manner in a textual form. Examples of common use cases include watching videos or participating in online meetings using real-time captioning. Models such as [Whisper] approach humans in their accuracy and robustness and are well positioned to improve accessibility of such use cases.

2.1.15. Text Generation

Various text generation use cases are enabled by large language models (LLM) that are able to perform tasks where a general ability to predict the next item in a text sequence is required. This class of models can translate texts, answer questions based on a text input, summarize a larger body of text, or generate text output based on a textual input. LLMs enable better performance compared to older models based on RNN, CNN, or LSTM architectures and further improve the performance of many other use cases discussed in this section. Examples of LLMs include [t5-small], [m2m100_418M], [gpt2], and [llama-2-7b].

2.1.16. Detecting fake video

A user is exposed to realistic fake videos generated by ‘deepfake’ on the web. The fake video can swap the speaker’s face into the president’s face to incite a user politically or to manipulate user’s opinion. The deepfake detection applications such as [FaceForensics++] analyze the videos and protect a user against the fake videos or images. When she watches a fake video on the web, the detection application alerts her of the fraud video in real-time.

2.2. Framework Use Cases

This section collects framework-level use cases for a dedicated low-level API for neural network inference hardware acceleration. It is expected that Machine Learning frameworks will be key consumers of the Web Neural Network API (WebNN API) and the low-level details exposed through the WebNN API are abstracted out from typical web developers. However, it is also expected that web developers with specific interest and competence in Machine Learning will want to interface with the WebNN API directly instead of a higher-level ML framework.

2.2.1. Custom Layer

A web application developer wants to run a DNN model on the WebNN API. However, she has found that some of activation functions like [LeakyReLU], [ELU], etc. are not included in the WebNN API. To address this issue, she constructs custom layers of the additional activation functions on top of the WebNN API. Note that the scope of custom layers may include convolution, normalization, etc. as well as activation.

2.2.2. Network Concatenation

A web application uses a DNN model, and its model data of upper convolutional layers and lower fully-connected layers are stored in separate files, since model data of the fully-connected layers are periodically updated due to fine tuning at the server side.

Therefore, the application downloads both partial model files at first and concatenates them into a single model. When the model is updated, the application downloads fine-tuned part of the model and replace only the fully-connected layers with it.

2.2.3. Performance Adaptation

A web application developer has a concern about performance of her DNN model on mobile devices. She has confirmed that it may run too slow on mobile devices which do not have GPU acceleration. To address this issue, her web application refers to the WebNN API to confirm whether acceleration is available or not, so that the application can display the warning for devices without acceleration.

After several weeks, she has developed a tiny DNN model that can even run on CPU. In order to accommodate CPU execution, she modifies the application so that the application loads the tiny model in the case of CPU-only devices.

2.2.4. Operation Level Execution

A JavaScript ML framework is responsible for loading, interpreting and executing a ML model. During the model execution phase, the framework iterates through the operations of the model and executes each operation on the hardware device, like CPU, GPU or ML accelerator. To avoid the unnecessary data copying across devices, the framework selects the same device to execute the operations. For a compute intensive operation, such as convolution 2D or matrix multiplication, the framework uses WebNN API to execute it with the ML-specific acceleration available on that selected device.

2.2.5. Integration with real-time video processing

The user experience of WebRTC-based video conferencing is enhanced using real-time video processing. For example, background blur implemented using a § 2.1.2 Semantic Segmentation model blurs the background in the user’s live camera feed. To satisfy the performance requirements of this use case, the WebNN API integrates with primitives from other Web APIs that make up the media pipeline to allow WebNN API-based transformation of real-time video streams.

3. Accessibility Considerations

This section provides guidance to web authors on how to improve accessibility of § 2.1 Application Use Cases enabled by neural network inference hardware acceleration. This guidance generalizes beyond the specific use cases outlined in this specification, and web authors are encouraged to consult [wcag] for further accessibility guidance and § 6 Ethical Considerations for digital accessibility in context of ethical principles.

§ 2.1.8 Image Captioning can be improved by ensuring the captions are surfaced to screen-reader and other Assistive Technology (AT) users. Web authors are encouraged to ensure the generated image captions are semantically linked to their respective images, either via the standard alt attribute, or other means which may depend on whether the descriptions are updated on initial page load, or later, as the result of user action.

§ 2.1.11 Emotion Analysis can mis-label and thus mis-classify users, leading to discriminatory experiences. Web authors are encouraged to expose confidence scores and give users an option to turn the feature off.

§ 2.1.13 Noise Suppression with aggressive filters can wipe out the speech of users with dysarthria, making captions and recognition fail. Web authors are encouraged to expose a bypass or sensitivity control, and not hard-wire noise suppression when live captions are active.

§ 2.2.5 Integration with real-time video processing with background-blur powered segmentation helps remove distractions, but can add too much delay that breaks lip-reading and live captions. Web authors are encouraged to provide an ability for user-facing keyboard- and screen-reader-operable “Background blur on/off” control, surfaced next to other accessibility/media settings.

§ 7.2 Device Selection allows web authors to indicate preferences for execution speed and power consumption. Implementers are encouraged to allow users to override the web author hint in browser UI to ensure that people on low-end or battery-sensitive devices can keep captions and other critical accessibility features responsive, especially on portable AAC or eye-gaze setups.

4. Security Considerations

This API is disabled by default in all cross-origin frames using the § 7.5 Permissions Policy Integration. This prevents third-party content from using this API unless the embedding page explicitly sets a policy that grants permission.

This API allows creation of an MLContext from a GPUDevice defined by WebGPU specification. See WebGPU Security Considerations for more information regarding security characteristics of this context.

This API provides an abstraction across GPU, CPU, and dedicated ML accelerator hardware. When using a GPU, denial of service considerations similar to WebGPU apply. When using a CPU or a dedicated ML accelerator, the types of potential resource contention are different and mitigations will be implementation and configuration dependent. Implementations should use whatever mechanisms are available from the platform to prevent sites from using an unfair amount of system resources. These compute units are shared resources, and the use of any compute API will affect overall performance on a fully-loaded system.

Once the graph is fully constructed and compiled, the input shapes into each of the operations in the graph are inferred and finalized. The bounds checking occurs when the compute method is invoked that executes the graph against the actual data. No actual data is bound to the compiled graph before this stage. It is the implementation’s responsibility to make sure proper bounds checking occurs against the shapes of the data already inferred by that time.

Document operations susceptible to out-of-bounds access as a guidance to implementers.

Implementations must defend against control-flow attacks based on changes to data considered to be constant. For example, optimizations in the underlying platform may assume that a weight remains unchanged throughout a computation. If the API allowed the contents of buffers holding weights to change during a computation then those optimization assumptions would be invalidated, causing undefined behavior in the underlying platform. The API mitigates this category of attacks from script by always copying or transferring buffers, but implementations should consider additional defenses such as process isolation of data assumed to be constant.

As a future-proofing measure, the API design allows certain operations that can be generically emulated to be deprecated for security, performance, or other reasons without breaking compatibility. This is made possible by high-level functions that are defined in terms of smaller primitive operations defined in this specifications. This enables a native implementation of a high-level function to be replaced with a polyfill implementation.

Investigate side channel attack feasibility considering the current state where CPU is shared between processes running renderers.

In order to not allow an attacker to target a specific implementation that may contain a flaw, the § 7.2 Device Selection mechanism is a hint only, and the concrete device selection is left to the implementation - a user agent could for instance choose never to run a model on a device with known vulnerabilities. As a further mitigation, no device enumeration mechanism is defined.

Hinting partially mitigates the concern. Investigate additional mitigations.

The API design minimizes the attack surface for the compiled computational graph. The MLGraphBuilder interface that hosts the various operations is a data definition API and as such doesn’t execute anything, only constructs data. What follows, is that the potential for an attack is limited to when binding the data to the graph before executing it by invoking the MLContext.dispatch() method. This enables implementers to focus on hardening the MLContext.dispatch() method. For example, by making sure it honors the boundary of data and fails appropriately when the bounds are not respected.

Purpose-built Web APIs for measuring high-resolution time mitigate against timing attacks using techniques such as resolution reduction, adding jitter, detection of abuse and API call throttling [hr-time-3]. The practical deployment of WebNN implementations are likely to bring enough jitter to make timing attacks impractical (e.g. because they would use IPC) but implementers are advised to consider and test their implementations against timing attacks.

Note: Security risks related to Unicode sequences are discussed in context of the label USVString definition.

4.1. Guidelines for new operations

This section is non-normative.

5. Privacy Considerations

This API provides a privacy improvement over cloud-based inference alternatives by keeping sensitive user data within the browser’s sandbox. Input data such as images, audio, video streams, and other personal information never leave the user’s device, eliminating risks associated with data transmission to remote servers and third-party data processing.

However, as a powerful local compute API that interacts closely with hardware acceleration capabilities, the WebNN API has to balance performance optimization with privacy protection. The API includes multiple privacy-preserving measures to mitigate against fingerprinting while still enabling effective machine learning inference capabilities.

5.1. Fingerprinting

By design, this API aims to expose the minimum amount of information necessary to address the identified § 2 Use cases with the best performance and reliability of results. First, the API mitigates against fingerprinting through standardization: by defining consistent behavior across diverse platform APIs and by minimizing information leakage about the underlying hardware variation across conformant implementations. This is achieved through:

• § 7.3 Operators that are hardware-agnostic and minimize the exposure of low-level details of the underlying platform, in line with the principle of data minimization.

• § 8.2.1 MLContextOptions API that allows a web developer to indicate preference for execution speed and power consumption, but does not expose the actual device selected for execution, nor does it allow a web developer to enumerate or select specific devices. This hinting mechanism does not add to the entropy.

• § 8.3.7 opSupportLimits() API that allows a web developer to query support for specific operators using an explicit query API instead of inferring this information using a side channel. This API can contribute to fingerprintability, but its entropy can be reduced by limiting the number of distinguishable configurations exposed through this API using buckets as appropriate.

• Standardized data types and tensor operations that work consistently across platforms.

• Consistent error handling across different backend implementations.

The overall design ensures that implementations maintain a consistent interface across different platforms while providing the necessary functionality. By abstracting platform-specific details, the API can provide privacy-preserving predictable behavior regardless of whether the underlying acceleration is provided by CPU, GPU, or dedicated ML hardware.

Note: MLContextOptions is under active development, and the design is expected to change, informed by further implementation experience and new use cases from the wider web community.

MLGraph.devices API extension has been proposed to expose the actual devices selected for execution after the graph is fully constructed and compiled. Privacy implications of this API extension are under investigation. [Issue #836]

5.2. Execution Time Analysis

The timing characteristics of operations can provide some indirect information about the underlying hardware performance, a feature inherent to any compute API. In certain circumstances an execution time analysis can reveal indirectly the performance of the underlying platform’s neural network hardware acceleration capabilities relative to another underlying platform. See also § 4 Security Considerations for further discussion on timing attacks.

Note: The group welcomes further input on the proposed execution time analysis fingerprinting vector and mitigations.

5.3. WebGPU Comparison

Unlike WebGPU, this API does not intrinsically support custom shader authoring; and as a result is not prone to timing attacks that rely on shader caches, or other persistent data. The API builds upon pre-existing shaders and lower level primitives of the browser or the underlying OS. Web developers who interface with GPUDevice are expected to be aware of WebGPU compilation cache considerations.

The WebGPU API identifies machine-specific artifacts as a privacy consideration. Similarly, the WebNN API’s compute unit scheduling may under certain circumstances introduce a fingerprint. However, similarly to WebGPU, such fingerprints are identical across most or all of the devices of each vendor, mitigating the concern. Furthermore, software implementations can be used to further eliminate such artifacts.

In general, implementers of this API are expected to apply WebGPU Privacy Considerations to their implementations where applicable.

6. Ethical Considerations

The Working Group has started documenting ethical issues associated with using Machine Learning on the Web, to help identify what mitigations its normative specifications should take into account. The Working Group publishes and maintains an Ethical Principles for Web Machine Learning document [webmachinelearning-ethics] open to contributions from the wider community via a dedicated GitHub repository.

7. Programming Model

7.1. Overview

At the heart of neural networks is a computational graph of mathematical operations. These operations are the building blocks of modern machine learning technologies in computer vision, natural language processing, and robotics. The WebNN API is a specification for constructing, compiling, and executing computational graphs of neural networks.

The MLGraph interface represents a compiled computational graph that is immutable (that is, a model).

The MLGraphBuilder interface serves as a builder (factory) to construct a computational graph (its graph) that is then compiled to create an MLGraph.

In WebNN, a computational graph is composed of operators which act on data, and are the nodes of the graph. MLOperands are a representation of data that flows within the computational graph, and are the edges of the graph. MLOperands include a computational graph’s input values for inference, constants (including trained weights) used for inference, intermediate values (often referred to as activations) computed during inference, as well as the output values of inference. An operator’s input is one or more MLOperands. An operator’s output is one or more MLOperands. Operators have operator-specific parameters that control their behavior, which can include zero or more activation functions.

A key part of the MLGraphBuilder interface are methods such as gemm() and relu() which create an operator which represents the actual operation to perform on the input data when the computation is run, and return a new MLOperand holding the operator. Methods that create an MLOperand connect any inputs and activations to the operator. Each method invocation returns a distinct new value, without changing the value of any other MLOperand.

An operator has a label, a string which may be included in diagnostics such as exception messages. When an operator is created its label is initialized in an implementation-defined manner and may include the passed label.

Consider adding a mechanism for reporting errors during dispatch(). [Issue #778]

At inference time, every MLOperand will be bound to a tensor (the actual data), which are essentially multidimensional arrays. The representation of the tensors is implementation dependent, but it typically includes the array data stored in some buffer (memory) and some metadata describing the array data (such as its shape).

Operations within the computational graph have functional semantics. This allows the implementation to potentially share the array data between multiple tensors. For example, the implementation of operations such as reshape, or slice may return a view of its input tensor that shares the same buffer as the input tensor. (In the case of reshape, the entire data is shared, while in the case of slice, a part of the input data is shared.) The implementation may use views, as above, for intermediate values.

Before the execution, the computation graph that is used to compute one or more specified outputs needs to be converted, compiled, and optimized. The key purpose of the compilation step is to enable optimizations that span two or more operations, such as operation or loop fusion. The user agent may also perform these optimizations during graph conversion.

The MLGraphBuilder.build() method compiles the graph in the background without blocking the calling thread, and returns a Promise that resolves to an MLGraph. Each MLGraphBuilder can build at most one MLGraph.

The MLGraph underlying implementation will be composed of platform-specific representations of operators and operands which correspond to the MLGraphBuilder’s operators and MLOperands, but which are not script-visible and may be compositions or decompositions of the graph as constructed by script.

Once the MLGraph is constructed, the MLContext.dispatch() method performs the execution of the graph asynchronously either on a parallel timeline in a separate worker thread for the CPU execution or on a GPU timeline in a GPU command queue. This method returns immediately without blocking the calling thread while the actual execution is offloaded to a different timeline. The caller supplies the input values using MLNamedTensors, binding the input MLOperands to their values. The caller also supplies MLNamedTensors for output MLOperands which will contain the result of graph execution, if successful, which may be read back to script using the MLContext.readTensor(tensor) method. This type of execution supports CPU, GPU, and NPU devices.

7.2. Device Selection

An MLContext interface represents a global state of neural network execution. One of the important context states is the underlying execution device that manages the resources and facilitates the compilation and the eventual execution of the neural network graph. In addition to the default method of creation with MLContextOptions, an MLContext could also be created from a specific GPUDevice that is already in use by the application.

In a situation when a GPU context executes a graph with a constant or an input in the system memory as an ArrayBufferView, the input content is automatically uploaded from the system memory to the GPU memory, and downloaded back to the system memory of an ArrayBufferView output buffer at the end of the graph execution. This data upload and download cycles will only occur whenever the execution device requires the data to be copied out of and back into the system memory, such as in the case of the GPU. It doesn’t occur when the device is a CPU device. Additionally, the result of the graph execution is in a known layout format. While the execution may be optimized for a native memory access pattern in an intermediate result within the graph, the output of the last operation of the graph must convert the content back to a known layout format at the end of the graph in order to maintain the expected behavior from the caller’s perspective.

For a history and rationale of this design, please see the device selection explainer.

7.3. Operators

This section is non-normative.

The WebNN API defines a set of operators required by well-known CNN and RNN, transformer and generative models that address key § 2.1 Application Use Cases. The details of each operator are defined in the normative sections of this specification, in alphabetical order by the operator name. These operators are grouped into categories based on their functionality in the following non-normative table to give a functional overview of the API surface.

Note: Some operators belong to multiple categories. For example, clamp() is both a math function and also used as an activation.

7.4. Task Source

The ML task source is a task source to be used for all tasks related to asynchronous compilation and execution of MLGraphs and creation of MLContexts.

7.5. Permissions Policy Integration

This specification defines a policy-controlled feature identified by the string "webnn". Its default allowlist is 'self'.

8. API

8.1. The navigator.ml interface

An ML object is available in the Window and WorkerGlobalScope contexts through the Navigator and WorkerNavigator interfaces respectively and is exposed via navigator.ml.

   {
 [, ]    ;
};
  ;
  ;

8.2. ML interface

  {
 ,
 ,
 
};

  {
   = "default";
   = ;
};

[, =(, )]
  {
 <> (   = {});
 <> ( );
};

8.2.1. MLContextOptions

Note: MLContextOptions is under active development, and the design is expected to change, informed by further implementation experience and new use cases from the wider web community. The Working Group is considering additional API controls to allow the definition of a fallback device, multiple devices in a preferred order, or an exclusion of a specific device. Other considerations under discussion include error handling, ultimate fallback, and quantized operators. Feedback is welcome on any of these design considerations from web developers, library authors, OS and hardware vendors, and other stakeholders via GitHub. See § 5 Privacy Considerations for additional discussion of fingerprinting considerations.

The powerPreference option is an MLPowerPreference and indicates the application’s preference as related to power consumption. It is one of the following:

"default"

Let the user agent select the most suitable behavior.

"high-performance"

Prioritizes execution speed over power consumption.

"low-power"

Prioritizes power consumption over other considerations such as execution speed.

The accelerated option indicates the application’s preference as related to massively parallel acceleration. This option has less priority than powerPreference. When set to true (by default), the underlying platform will attempt to use the available massively parallel accelerators, such as a GPU or NPU, also depending on the powerPreference. When set to false, the application indicates it prefers CPU inference. If there is contradictory input, for instance when powerPreference is "high-performance" and accelerated is false, then the implementation will choose the best available match in the underlying platform (for instance a high performance CPU mode, or will ignore accelerated as it has less priority than powerPreference).

8.2.2. createContext()

8.3. MLContext interface

 <, > ;

  {
  ;
};

[, =(, )]
  {
  ( ,  ,  );

 <> ( );
 <> (
  ,  );

 <> ( );
 <> ( ,  );

  ( ,  );

  ();

  ();

    ;
   <> ;
};

The context type is the type of the execution context that manages the resources and facilitates the compilation and execution of the neural network graph:

"default"

Context created per user preference options.

"webgpu"

Context created from WebGPU device.

8.3.1. dispatch()

Schedules the computational workload of a compiled MLGraph on the MLContext’s [[timeline]].

Note: dispatch() itself provides no signal that graph execution has completed. Rather, callers can await the results of reading back the output tensors. See § 8.3.1.1 Examples below.

When a constant operand is created using a tensor, it is legal for that tensor to be destroyed after build completes. Implementations are expected to ensure that the compiled graph remains valid and unaffected by such destruction.

8.3.1.1. Examples

8.3.2. createTensor()

Creates an MLTensor associated with this MLContext.

8.3.3. createConstantTensor()

Creates a constant MLTensor associated with this MLContext.

8.3.4. readTensor(tensor)

Reads back the [[data]] of an MLTensor from the MLContext.[[timeline]] to script.

8.3.5. readTensor(tensor, outputData)

Bring-your-own-buffer variant of readTensor(tensor). Reads back the [[data]] of an MLTensor into the provided buffer.

8.3.6. writeTensor()

Writes data to the [[data]] of an MLTensor on the MLContext’s [[timeline]].

Note: Similar to dispatch(), writeTensor() itself provides no signal that the write has completed. To inspect the contents of a tensor, callers can await the results of reading back the tensor.

8.3.7. opSupportLimits()

Note: The opSupportLimits() API is not intended to provide additional entropy for browser fingerprinting. In current implementations this feature support information can be inferred from the OS and browser version alone. If the diversity of future implementations warrants it, this API allows future implementations to add new privacy mitigations e.g. to bucket capabilities similar to WebGPU to reduce entropy.

See § 5 Privacy Considerations for additional discussion of fingerprinting considerations.

8.3.7.1. MLOpSupportLimits dictionary

  {
  ;
 []  ;
  ;
  ;
  ;
};

preferredInputLayout, of type MLInputOperandLayout

Preferred input layout for layout dependent operators like conv2d().

maxTensorByteLength, of type unsigned long long

The maximum supported length of tensors, in bytes.

input, of type MLTensorLimits

Support limits for input MLOperands for an MLGraph.

constant, of type MLTensorLimits

Support limits for constant MLOperands for an MLGraph.

output, of type MLTensorLimits

Support limits for output MLOperands for an MLGraph.

8.3.7.2. MLRankRange dictionary

  {
  ;
  ;
};

min, of type unsigned long

Minimum supported rank.

max, of type unsigned long

Maximum supported rank.

8.3.7.3. MLTensorLimits dictionary

 <> ;

  {
  ;
  ;
};

dataTypes, of type MLDataTypeList

Supported data types.

rankRange, of type MLRankRange

Minimum and maximum supported ranks.

8.3.7.4. MLBinarySupportLimits dictionary

  {
  ;
  ;
  ;
};

a, of type MLTensorLimits

MLTensorLimits for a operand.

b, of type MLTensorLimits

MLTensorLimits for b operand.

output, of type MLTensorLimits

MLTensorLimits for output operand.

8.3.7.5. MLSingleInputSupportLimits dictionary

  {
  ;
  ;
};

input, of type MLTensorLimits

MLTensorLimits for input operand.

output, of type MLTensorLimits

MLTensorLimits for output operand.

8.3.8. destroy()

The destroy() method can be called to release all resources associated with the context. Any outstanding compute requests and MLTensor creation/read/write requests will fail.

8.3.9. Errors

When a user agent determines that an MLContext is no longer available to fulfill requests, it must run the context lost steps for it.

message, of type DOMString

An implementation-defined message providing information about the error that occurred.

A MLContext is lost if its [[lost]] Promise is settled.

8.4. MLGraph interface

[, =(, )]
  {
  ();
};

8.4.1. destroy()

The destroy() method can be called to release all resources associated with the graph.

Note: Since no further workloads can be enqueued using this graph, implementations can free any additional resource allocations associated with this graph once all previously submitted workloads using it are complete.

8.5. MLOperandDescriptor dictionary

An MLOperandDescriptor describes the shape (dimensions) and data type of an operand. They are used to describe the inputs and constants for an MLGraph, and every MLOperand has an internal MLOperandDescriptor.

  {
 ,
 
};

  {
 ,
 ,
 ,
 ,
 ,
 ,
 ,
 
};

  {
   ;
  <[] > ;
};

dataType, of type MLOperandDataType

The operand data type.

shape, of type sequence<[EnforceRange] unsigned long>

The list of dimensions of the operand. It is empty for scalar operands.

A valid dimension is an integer greater than zero and in the range of long. Implementations may impose a smaller upper bound.

A valid tensor count is an integer greater than zero and less or equal to 8192. Implementations may impose a smaller upper bound.

Should 0-size dimensions be supported? [Issue #391]

8.6. MLOperand interface

An MLOperand represents an intermediary graph being constructed as a result of compositing parts of an operation into a fully composed operation.

For instance, an MLOperand can represent a constant feeding to an operation or the result from combining multiple constants together into an operation. See also § 7 Programming Model.

[, =(, )]
  {
    ;
   <> ;
};

  {
   = "";
};

 (  ) ;

An MLOperand’s dataType is its [[descriptor]].dataType.

An MLOperand’s shape is its [[descriptor]].shape.

An MLOperand’s rank is its shape’s size.

The dataType getter steps are to return this’s dataType.

The shape getter steps are to return this’s shape.

Since the [[builder]] object is bound by the MLGraphBuilder() constructor to an MLContext object, an MLOperand is also always bound to the same MLContext object.

If an operation supports only a subset of MLOperandDataTypes, the allowed data types for each of the operation’s input operands, including both positional arguments and options, are given as either an explicit list of MLOperandDataTypes, or a constraint that the operand’s dataType must be the same as the dataType of another input operand, or any to allow any MLOperandDataType.

Implementations may support fewer data types for operands than specified. This can be queried for each operation using the opSupportLimits() method on MLContext and inspecting the dataTypes value of the corresponding member for the operation.

Should we specify the subset of data types that must be supported for each operator?

If an operation requires input operands with a particular rank, the allowed ranks for each of the operation’s input operands, including both positional arguments and options, are given as an explicit rank (e.g. 1), or N to allow any dimensionality, or the same as another operand. More specific constraints are common, such as when an input operand’s shape must be unidirectionally broadcastable to or bidirectionally broadcastable with another input operand; in these cases, the allowed ranks are listed as a range, with specific validation given as steps in the operation.

Implementations may impose a more restricted lower bound and/or upper bound on the rank of operands than specified. This can be queried for each operation using the opSupportLimits() method on MLContext and inspecting the rankRange.min and rankRange.max values of the corresponding member for the operation.

MLOperatorOptions has the following members:

label, of type USVString, defaulting to ""

Optionally provided when an operator is created using MLGraphBuilder methods that create MLOperands. The implementation may use this value to initialize the operator’s label.

Note: The label is not intended to be a natural language string. It is a language-independent identifier, analogous to a variable name or error code, like "mul#1234".

Note: Implementations are encouraged to use the label provided by developers to enhance error messages and improve debuggability, including both synchronous errors during graph construction and for errors that occur during the asynchronous build() method.

When displaying labels provided by developers via label in debugging tools, logs, or error messages, implementations should sanitize the output to prevent security risks, such as injection of malicious Unicode sequences (e.g. Bidirectional Text Spoofing [UTR36], Source Code Spoofing [UTS55] and other concerns). For example, implementations should escape or filter control characters (e.g., U+202A to U+202E, U+2066 to U+2069) or use a safe rendering mechanism to neutralize potential spoofing.

8.6.1. Creating an MLOperand

To validate operand given MLGraphBuilder builder and MLOperand operand, return true if operand.[[builder]] is builder, and false otherwise.

8.6.1.1. MLNumber

MLNumber is used when specifying the type of a numeric option for an MLOperand which can be of any MLOperandDataType, including both 64-bit integer types ("uint64" and "int64") and 32-bit floating point ("float32"). Implementations process the value according to the corresponding MLOperandDataType. For example, if clamp(input, options) is called with an MLOperand with dataType "uint32", the MLNumber parameters are explicitly cast to unsigned long.

Support for unions of bigint and numeric types is new in [WEBIDL], and implementation support is also limited. Prototype implementations are encouraged to provide feedback for this approach. [whatwg/webidl Issue #1388]

8.7. MLTensorDescriptor dictionary

An MLTensorDescriptor describes the characteristics and capabilities of an MLTensor.

  :  {
   = ;
   = ;
};

readable, of type boolean, defaulting to false

Whether the tensor’s contents can be read via readTensor(tensor) or readTensor(tensor, outputData).

writable, of type boolean, defaulting to false

Whether the tensor’s contents can be written to via writeTensor().

8.8. MLTensor interface

The MLTensor interface represents a tensor which may be used as an input or output to an MLGraph. The memory backing an MLTensor should be allocated in an implementation-defined fashion according to the requirements of the MLContext and the MLTensorDescriptor used to create it. Operations involving the [[data]] of an MLTensor occur on the [[timeline]] of its associated MLContext.

The implementation-defined requirements of how an MLTensor is allocated may include constraints such as that the memory is allocated with a particular byte alignment or in a particular memory pool.

[, =(, )]
  {
    ;
   <> ;
    ;
    ;
    ;

  ();
};

An MLTensor’s dataType is its [[descriptor]]’s dataType.

An MLTensor’s shape is its [[descriptor]]’s shape.

The dataType getter steps are to return this’s dataType.

The shape getter steps are to return this’s shape.

The readable getter steps are to return this.[[descriptor]].readable.

The writable getter steps are to return this.[[descriptor]].writable.

The constant getter steps are to return this’s [[isConstant]].

8.8.1. Creating an MLTensor

An MLTensor is created by its associated MLContext.

8.8.2. destroy()

Releases the resources associated with the MLTensor. This method is idempotent.

Note: Since no further operations can be enqueued using this tensor, implementations can free any additional resource allocations associated with this tensor once all previously submitted operations using it are complete.

8.8.3. Creating a constant MLTensor

A constant MLTensor is created by its associated MLContext.

8.9. MLGraphBuilder interface

The MLGraphBuilder interface defines a set of operations as identified by the § 2 Use cases that can be composed into a computational graph. It also represents the intermediate state of a graph building session.

 <, > ;

[, =(, )]
  {
 // Construct the graph builder from the context.
 ( );

 // Create an operand for a graph input.
  ( ,  );

 // Create an operand for a graph constant.
  ( ,
  );

 // Create a scalar operand from the specified number of the specified type.
  ( ,  );

 // Create an operand from a specified constant tensor.
  ( );

 // Compile the graph up to the specified output operands asynchronously.
 <> ( );
};

8.9.1. MLGraphBuilder constructor

8.9.2. input operands

Create a named MLOperand based on a descriptor, that can be used as an input.

8.9.3. constant operands

8.9.3.1. constant(descriptor, buffer)

8.9.3.2. constant(tensor)

8.9.3.3. constant(dataType, value)

8.9.4. build method

NOTE: Specifying an input operand or constant operand as a graph output results in an error, as this is usually an incorrect usage of the API. Callers can work around this by introducing an identity() operator.

8.9.5. argMin/argMax operations

  :  {
   = ;
   = "int32";
};

   {
  ( , []  ,
    = {});
  ( , []  ,
    = {});
};

   {
  ;
  ;
};

MLArgMinMaxOptions has the following members:

keepDimensions, of type boolean, defaulting to false

If true, retains reduced dimensions with size 1.

outputDataType, of type MLOperandDataType, defaulting to "int32"

An MLOperandDataType. The output data type.

MLOpSupportLimits has the following members for argMin() and argMax():

argMin, of type MLSingleInputSupportLimits

Support limits for operator argMin().

argMax, of type MLSingleInputSupportLimits

Support limits for operator argMax().

8.9.6. batchNormalization

  :  {
  ;
  ;
 []   = 1;
   = 1e-5;
};

   {
  ( ,  ,  ,
    = {});
};

  {
  ;
  ;
  ;
  ;
  ;
  ;
};

   {
  ;
};

MLBatchNormalizationOptions has the following members:

scale, of type MLOperand

The 1-D tensor of the scaling values whose size is equal to the size of the input dimension denoted by axis.

bias, of type MLOperand

The 1-D tensor of the bias values whose size is equal to the size of the input dimension denoted by axis.

axis, of type unsigned long, defaulting to 1

The index to the feature count dimension of the input shape for which the mean and variance values are. Its value must be in the range [0, N-1] where N is the rank of the input tensor. The default value is 1, corresponding to the channel ("c") dimension in the "nchw" data layout.

epsilon, of type double, defaulting to 1e-5

A small value to prevent computational error due to divide-by-zero.

MLBatchNormalizationSupportLimits has the following members:

input, of type MLTensorLimits

MLTensorLimits for input operand.

mean, of type MLTensorLimits

MLTensorLimits for mean operand.

variance, of type MLTensorLimits

MLTensorLimits for variance operand.

scale, of type MLTensorLimits

MLTensorLimits for scale operand.

bias, of type MLTensorLimits

MLTensorLimits for bias operand.

output, of type MLTensorLimits

MLTensorLimits for output operand.

MLOpSupportLimits has the following members for batchNormalization():

batchNormalization, of type MLBatchNormalizationSupportLimits

Support limits for operator batchNormalization().

8.9.7. cast

   {
  ( ,
  ,
    = {});
};

   {
  ;
};

MLOpSupportLimits has the following members for cast():

cast, of type MLSingleInputSupportLimits

Support limits for operator cast().

Casting between MLOperandDataTypes is specified for some cases and implementation-defined in other cases, according to the following table:

NOTE: For example, casting -1 from "int8" to "uint8" is specified to yield 255. But casting -1 from "float32" to "uint8" is implementation-defined.

8.9.8. clamp

  :  {
  ;
  ;
};

   {
  ( ,    = {});
};

   {
  ;
};

MLClampOptions has the following members:

minValue, of type MLNumber

The minimum value of the range. When it is not specified, the clamping is not performed on the lower limit of the range.

maxValue, of type MLNumber

The maximum value of the range. When it is not specified, the clamping is not performed on the upper limit of the range.

MLOpSupportLimits has the following member for clamp():

clamp, of type MLSingleInputSupportLimits

Support limits for operator clamp().

8.9.9. concat

   {
  (<> ,
 []  ,
    = {});
};

  {
  ;
  ;
};

   {
  ;
};

MLConcatSupportLimits has the following members:

inputs, of type MLTensorLimits

MLTensorLimits for all input operands.

output, of type MLTensorLimits

MLTensorLimits for output operand.

MLOpSupportLimits has the following member for concat():

concat, of type MLConcatSupportLimits

Support limits for operator concat().

8.9.10. conv2d

  {
 ,
 ,
 ,
 
};

  :  {
 <[] > ;
 <[] > ;
 <[] > ;
 []   = 1;
   = "nchw";
   = "oihw";
  ;
};

   {
  ( ,
  ,
    = {});
};

  {
  ;
  ;
  ;
  ;
};

   {
  ;
};

MLConv2dOptions has the following members:

padding, of type sequence<[EnforceRange] unsigned long>

A list of length 4: [beginningHeight, endingHeight, beginningWidth, endingWidth]. Specifies the additional rows and columns added to the beginning and ending of each spatial dimension of the convolution input. The default value is [0, 0, 0, 0].

strides, of type sequence<[EnforceRange] unsigned long>

A list of length 2: [strideHeight, strideWidth]. Specifies the stride of the sliding window for each spatial dimension of the convolution input. The default value is [1, 1].

dilations, of type sequence<[EnforceRange] unsigned long>

A list of length 2: [dilationHeight, dilationWidth]. Specifies the dilation factor for each spatial dimension applied on the convolution filter (kernel). The default value is [1, 1].

groups, of type unsigned long, defaulting to 1

The number of groups that input channels and output channels are divided into.

inputLayout, of type MLInputOperandLayout, defaulting to "nchw"

Specifies the layout format of the input and output tensor as follows:

• "nchw"

  • input tensor: [batches, inputChannels, height, width]

  • output tensor: [batches, outputChannels, height, width]

• "nhwc":

  • input tensor: [batches, height, width, inputChannels]

  • output tensor: [batches, height, width, outputChannels]

filterLayout, of type MLConv2dFilterOperandLayout, defaulting to "oihw"

Specifies the layout format of the filter tensor as follows:

• "oihw": [outputChannels, inputChannels/groups, height, width]

• "hwio": [height, width, inputChannels/groups, outputChannels]

• "ohwi": [outputChannels, height, width, inputChannels/groups]

• "ihwo": [inputChannels/groups, height, width, outputChannels]

bias, of type MLOperand

An additional 1-D tensor with the shape of [outputChannels] whose values are to be added to the convolution result.

MLConv2dSupportLimits has the following members:

input, of type MLTensorLimits

MLTensorLimits for input operand.

filter, of type MLTensorLimits

MLTensorLimits for filter operand.

bias, of type MLTensorLimits

MLTensorLimits for bias operand.

output, of type MLTensorLimits

MLTensorLimits for output operand.

MLOpSupportLimits has the following member for conv2d():

conv2d, of type MLConv2dSupportLimits

Support limits for operator conv2d().

8.9.11. convTranspose2d

  {
 ,
 ,
 
};

  :  {
 <[] > ;
 <[] > ;
 <[] > ;
 <[] > ;
 <[] > ;
 []   = 1;
   = "nchw";
   = "iohw";
  ;
};

   {
  ( ,  ,
    = {});
};

   {
  ;
};

MLConvTranspose2dOptions has the following members:

padding, of type sequence<[EnforceRange] unsigned long>

A list of length 4: [beginningHeight, endingHeight, beginningWidth, endingWidth]. Specifies the additional rows and columns added to the beginning and ending of each spatial dimension of the convolution input. The default value is [0, 0, 0, 0].

strides, of type sequence<[EnforceRange] unsigned long>

A list of length 2: [strideHeight, strideWidth]. Specifies the stride of the sliding window for each spatial dimension of the convolution input. The default value is [1, 1].

dilations, of type sequence<[EnforceRange] unsigned long>

A list of length 2: [dilationHeight, dilationWidth]. Specifies the dilation factor for each spatial dimension applied on the convolution filter (kernel). The default value is [1, 1].

outputPadding, of type sequence<[EnforceRange] unsigned long>

A list of length 2. Specifies the padding values applied to each spatial dimension of the output tensor. The explicit padding values are needed to disambiguate the output tensor shape for transposed convolution when the value of the strides is greater than 1.

Note that these values are only used to disambiguate output shape when needed; it does not necessarily cause any padding value to be written to the output tensor.

The default value is [0, 0].

outputSizes, of type sequence<[EnforceRange] unsigned long>

A list of length 2. Specifies the sizes of the last two dimensions of the output tensor. When the output sizes are explicitly specified, the output padding values in outputPadding are ignored.

If not specified, the output sizes are automatically computed.

groups, of type unsigned long, defaulting to 1

The number of groups that input channels and output channels are divided into.

inputLayout, of type MLInputOperandLayout, defaulting to "nchw"

Specifies the layout format of the input and output tensor as follows:

• "nchw"

  • input tensor: [batches, inputChannels, height, width]

  • output tensor: [batches, outputChannels, height, width]

• "nhwc":

  • input tensor: [batches, height, width, inputChannels]

  • output tensor: [batches, height, width, outputChannels]

filterLayout, of type MLConvTranspose2dFilterOperandLayout, defaulting to "iohw"

Specifies the layout format of the filter tensor as follows:

• "iohw": [inputChannels, outputChannels/groups, height, width]

• "hwoi": [height, width, outputChannels/groups, inputChannels]

• "ohwi": [outputChannels/groups, height, width, inputChannels]

bias, of type MLOperand

An additional 1-D tensor with the shape of [outputChannels] whose values are to be added to the convolution result.

MLOpSupportLimits has the following member for convTranspose2d():

convTranspose2d, of type MLConv2dSupportLimits

Support limits for operator convTranspose2d().

8.9.12. cumulativeSum

  :  {
   = ;
   = ;
};

   {
  ( ,
  ,
    = {});
};

   {
  ;
};

MLCumulativeSumOptions has the following members:

exclusive, of type boolean, defaulting to false

Whether to include or exclude the current value in the output, meaning inclusive prefix sum or exclusive prefix sum [Prefix-sum]. Given input [1,2,3,4], inclusive summation would yield an output of [1,3,6,10] whereas exclusive would yield [0,1,3,6]. The default is inclusive.

reversed, of type boolean, defaulting to false

Whether to reverse the summation direction along the active axis to instead start from the high coordinate to low coordinate. Given input [1,2,3,4], inclusive forward summation would yield an output of [1,3,6,10] whereas inclusive backward summation would yield [10,9,7,4]. The default is forward.

MLOpSupportLimits has the following member for cumulativeSum():

cumulativeSum, of type MLSingleInputSupportLimits

Support limits for operator cumulativeSum().

8.9.13. Element-wise binary operations

The operation will be broadcast according to [numpy-broadcasting-rule]. The input tensors must be bidirectionally broadcastable. The rank of the output tensor is the maximum rank of the input tensors. For each dimension of the output tensor, its size is the maximum size along that dimension of the input tensors.

   {
  ( ,  ,    = {});
  ( ,  ,    = {});
  ( ,  ,    = {});
  ( ,  ,    = {});
  ( ,  ,    = {});
  ( ,  ,    = {});
  ( ,  ,    = {});
};

   {
  ;
  ;
  ;
  ;
  ;
  ;
  ;
};

MLOpSupportLimits has the following members for element-wise binary operations:

add, of type MLBinarySupportLimits

Support limits for operator add().

sub, of type MLBinarySupportLimits

Support limits for operator sub().

mul, of type MLBinarySupportLimits

Support limits for operator mul().

div, of type MLBinarySupportLimits

Support limits for operator div().

max, of type MLBinarySupportLimits

Support limits for operator max().

min, of type MLBinarySupportLimits

Support limits for operator min().

pow, of type MLBinarySupportLimits

Support limits for operator pow().

8.9.14. Element-wise logical operations

For multiple-operand operations, the operation will be broadcast according to [numpy-broadcasting-rule]. The input tensors must be bidirectionally broadcastable. The rank of the output tensor is the maximum rank of the input tensors. For each dimension of the output tensor, its size is the maximum size along that dimension of the input tensors.

   {
  ( ,
  ,
    = {});
  ( ,
  ,
    = {});
  ( ,
  ,
    = {});
  ( ,
  ,
    = {});
  ( ,
  ,
    = {});
  ( ,
  ,
    = {});
  ( ,    = {});
  ( ,
  ,
    = {});
  ( ,
  ,
    = {});
  ( ,
  ,
    = {});
  ( ,    = {});
  ( ,    = {});
};

  {
  ;
  ;
};

   {
  ;
  ;
  ;
  ;
  ;
  ;
  ;
  ;
  ;
  ;
  ;
  ;
};

MLLogicalNotSupportLimits has the following members:

a, of type MLTensorLimits

MLTensorLimits for a operand.

output, of type MLTensorLimits

MLTensorLimits for output operand.

MLOpSupportLimits has the following members for element-wise logical operations:

equal, of type MLBinarySupportLimits

Support limits for operator equal().

notEqual, of type MLBinarySupportLimits

Support limits for operator notEqual().

greater, of type MLBinarySupportLimits

Support limits for operator greater().

greaterOrEqual, of type MLBinarySupportLimits

Support limits for operator greaterOrEqual().

lesser, of type MLBinarySupportLimits

Support limits for operator lesser().

lesserOrEqual, of type MLBinarySupportLimits

Support limits for operator lesserOrEqual().

logicalNot, of type MLLogicalNotSupportLimits

Support limits for operator logicalNot().

logicalAnd, of type MLBinarySupportLimits

Support limits for operator logicalAnd().

logicalOr, of type MLBinarySupportLimits

Support limits for operator logicalOr().

logicalXor, of type MLBinarySupportLimits

Support limits for operator logicalXor().

isNaN, of type MLLogicalNotSupportLimits

Support limits for operator isNaN().

isInfinite, of type MLLogicalNotSupportLimits

Support limits for operator isInfinite().

8.9.15. Element-wise unary operations

   {
  ( ,    = {});
  ( ,    = {});
  ( ,    = {});
  ( ,    = {});
  ( ,    = {});
  ( ,    = {});
  ( ,    = {});
  ( ,    = {});
  ( ,    = {});
  ( ,    = {});
  ( ,    = {});
  ( ,    = {});
  ( ,    = {});
  ( ,    = {});
  ( ,    = {});
};

   {
  ;
  ;
  ;
  ;
  ;
  ;
  ;
  ;
  ;
  ;
  ;
  ;
  ;
  ;
  ;
};

MLOpSupportLimits has the following members for element-wise unary operations:

abs, of type MLSingleInputSupportLimits

Support limits for operator abs().

ceil, of type MLSingleInputSupportLimits

Support limits for operator ceil().

cos, of type MLSingleInputSupportLimits

Support limits for operator cos().

erf, of type MLSingleInputSupportLimits

Support limits for operator erf().

exp, of type MLSingleInputSupportLimits

Support limits for operator exp().

floor, of type MLSingleInputSupportLimits

Support limits for operator floor().

identity, of type MLSingleInputSupportLimits

Support limits for operator identity().

log, of type MLSingleInputSupportLimits

Support limits for operator log().

neg, of type MLSingleInputSupportLimits

Support limits for operator neg().

reciprocal, of type MLSingleInputSupportLimits

Support limits for operator reciprocal().

roundEven, of type MLSingleInputSupportLimits

Support limits for operator roundEven().

sin, of type MLSingleInputSupportLimits

Support limits for operator sin().

sign, of type MLSingleInputSupportLimits

Support limits for operator sign().

sqrt, of type MLSingleInputSupportLimits

Support limits for operator sqrt().

tan, of type MLSingleInputSupportLimits

Support limits for operator tan().

8.9.16. dequantizeLinear

   {
  ( ,
  ,
  ,
    = {});
};

  {
  ;
  ;
  ;
  ;
};

   {
  ;
};

MLQuantizeDequantizeLinearSupportLimits has the following members:

input, of type MLTensorLimits

MLTensorLimits for input operand.

scale, of type MLTensorLimits

MLTensorLimits for scale operand.

zeroPoint, of type MLTensorLimits

MLTensorLimits for zeroPoint operand.

output, of type MLTensorLimits

MLTensorLimits for output operand.

MLOpSupportLimits has the following member for dequantizeLinear():

dequantizeLinear, of type MLQuantizeDequantizeLinearSupportLimits

Support limits for operator dequantizeLinear().

8.9.17. quantizeLinear

   {
  ( ,
  ,
  ,
    = {});
};

   {
  ;
};

MLOpSupportLimits has the following member for quantizeLinear():

quantizeLinear, of type MLQuantizeDequantizeLinearSupportLimits

Support limits for operator quantizeLinear().

8.9.18. elu

  :  {
   = 1;
};

   {
  ( ,    = {});
};

   {
  ;
};

MLEluOptions has the following members:

alpha, of type double, defaulting to 1

A scalar multiplier.

MLOpSupportLimits has the following members for elu():

elu, of type MLSingleInputSupportLimits

Support limits for operator elu().

8.9.19. expand

   {
  ( ,
 <[] > ,
    = {});
};

   {
  ;
};

MLOpSupportLimits has the following members for expand():

expand, of type MLSingleInputSupportLimits

Support limits for operator expand().

8.9.20. gather

  :  {
 []   = 0;
};

   {
  ( ,
  ,
    = {});
};

  {
  ;
  ;
  ;
};

   {
  ;
};

MLGatherOptions has the following members:

axis, of type unsigned long, defaulting to 0

The axis along which the gathered values are obtained. Its value must be in the range [0, N-1] where N is the rank of the input tensor.

MLGatherSupportLimits has the following members:

input, of type MLTensorLimits

MLTensorLimits for input operand.

indices, of type MLTensorLimits

MLTensorLimits for indices operand.

output, of type MLTensorLimits

MLTensorLimits for output operand.

MLOpSupportLimits has the following members for gather():

gather, of type MLGatherSupportLimits

Support limits for operator gather().

8.9.21. gatherElements

   {
  ( ,
  ,
    = {});
};

   {
  ;
};

MLOpSupportLimits has the following members for gatherElements():

gatherElements, of type MLGatherSupportLimits

Support limits for operator gatherElements().

8.9.22. gatherND

   {
  ( ,
  ,
    = {});
};

   {
  ;
};

MLOpSupportLimits has the following members for gatherND():

gatherND, of type MLGatherSupportLimits

Support limits for operator gatherND().

8.9.23. gelu

   {
  ( ,    = {});
};

   {
  ;
};

MLOpSupportLimits has the following member for gelu():

gelu, of type MLSingleInputSupportLimits

Support limits for operator gelu().

8.9.24. gemm

  :  {
  ;
   = 1.0;
   = 1.0;
   = ;
   = ;
};

   {
  ( ,  ,    = {});
};

  {
  ;
  ;
  ;
  ;
};

   {
  ;
};

MLGemmOptions has the following members:

c, of type MLOperand

The third input tensor. It is either a scalar, or of the shape that is unidirectionally broadcastable to the shape [M, N]. When it is not specified, the computation is done as if c is a scalar 0.0.

alpha, of type double, defaulting to 1.0

A multiplier for the first input.

beta, of type double, defaulting to 1.0

A multiplier for the third input c.

aTranspose, of type boolean, defaulting to false

Indicates if the first input is transposed prior to calculating the output.

bTranspose, of type boolean, defaulting to false

Indicates if the second input is transposed prior to calculating the output.

MLGemmSupportLimits has the following members:

a, of type MLTensorLimits

MLTensorLimits for a operand.

b, of type MLTensorLimits

MLTensorLimits for b operand.

c, of type MLTensorLimits

MLTensorLimits for c operand.

output, of type MLTensorLimits

MLTensorLimits for output operand.

MLOpSupportLimits has the following member for gemm():

gemm, of type MLGemmSupportLimits

Support limits for operator gemm().

8.9.25. gru

  {
 , // update-reset-new gate ordering
  // reset-update-new gate ordering
};

  {
 ,
 ,
 
};

  {
 ,
 ,
 
};

  :  {
  ;
  ;
  ;
   = ;
   = ;
   = "forward";
   = "zrn";
 <> ;
};

   {
 <> ( ,
  ,
  ,
 []  ,
 []  ,
    = {});
};

  {
  ;
  ;
  ;
  ;
  ;
  ;
  ;
  ;
};

   {
  ;
};

MLGruOptions has the following members:

bias, of type MLOperand

The 2-D input bias tensor of shape [numDirections, 3 * hiddenSize]. The ordering of the bias vectors in the second dimension of the tensor shape is specified according to layout.

recurrentBias, of type MLOperand

The 2-D recurrent bias tensor of shape [numDirections, 3 * hiddenSize]. The ordering of the bias vectors in the second dimension of the tensor shape is specified according to layout.

initialHiddenState, of type MLOperand

The 3-D initial hidden state tensor of shape [numDirections, batchSize, hiddenSize]. When not specified, implementations must use a tensor filled with zero.

resetAfter, of type boolean, defaulting to true

Indicates whether to apply the reset gate after or before matrix multiplication.

returnSequence, of type boolean, defaulting to false

Indicates whether to also return the entire sequence with every output from each time step in it in addition to the output of the last time step.

direction, of type MLRecurrentNetworkDirection, defaulting to "forward"

The processing direction of the input sequence. When set to "both", the size of the first dimension of the weight and the bias tensor shapes must be 2, and the input is processed in both directions.

layout, of type MLGruWeightLayout, defaulting to "zrn"

The ordering of the weight and bias vectors for the internal gates of GRU, specifically the update (z), reset (r), and new (n) gate, as indicated in the second dimension of the weight and bias tensor shape.

activations, of type sequence<MLRecurrentNetworkActivation>

Specifies a pair of activation functions with the first function used for the update and reset gate, and the second used for the new gate. When not specified, defaults to the "sigmoid" and "tanh" functions, respectively.

MLGruSupportLimits has the following members:

input, of type MLTensorLimits

MLTensorLimits for input operand.

weight, of type MLTensorLimits

MLTensorLimits for weight operand.

recurrentWeight, of type MLTensorLimits

MLTensorLimits for recurrentWeight operand.

bias, of type MLTensorLimits

MLTensorLimits for bias operand.

recurrentBias, of type MLTensorLimits

MLTensorLimits for recurrentBias operand.

initialHiddenState, of type MLTensorLimits

MLTensorLimits for initialHiddenState operand.

output0, of type MLTensorLimits

MLTensorLimits for all the output operands[0].

output1, of type MLTensorLimits

MLTensorLimits for all the output operands[1].

MLOpSupportLimits has the following member for gru():

gru, of type MLGruSupportLimits

Support limits for operator gru().

8.9.26. gruCell

  :  {
  ;
  ;
   = ;
   = "zrn";
 <> ;
};

   {
  ( ,
  ,
  ,
  ,
 []  ,
    = {});
};

  {
  ;
  ;
  ;
  ;
  ;
  ;
  ;
};

   {
  ;
};

MLGruCellOptions has the following members:

bias, of type MLOperand

The 1-D input bias tensor of shape [3 * hiddenSize]. The ordering of the bias vectors in the second dimension of the tensor shape is specified according to layout.

recurrentBias, of type MLOperand

The 1-D recurrent bias tensor of shape [3 * hiddenSize]. The ordering of the bias vectors in the second dimension of the tensor shape is specified according to layout.

resetAfter, of type boolean, defaulting to true

Indicates whether to apply the reset gate after or before matrix multiplication.

layout, of type MLGruWeightLayout, defaulting to "zrn"

The ordering of the weight and bias vectors for the internal gates of GRU, specifically the update (z), reset (r), and new (n) gate, as indicated in the second dimension of the weight and bias tensor shape.

activations, of type sequence<MLRecurrentNetworkActivation>

Specifies a pair of activation functions with the first function used for the update and reset gate, and the second used for the new gate. When not specified, defaults to the "sigmoid" and "tanh" functions, respectively.

MLGruCellSupportLimits has the following members;

input, of type MLTensorLimits

MLTensorLimits for input operand.

weight, of type MLTensorLimits

MLTensorLimits for weight operand.

recurrentWeight, of type MLTensorLimits

MLTensorLimits for recurrentWeight operand.

hiddenState, of type MLTensorLimits

MLTensorLimits for hiddenState operand.

bias, of type MLTensorLimits

MLTensorLimits for bias operand.

recurrentBias, of type MLTensorLimits

MLTensorLimits for recurrentBias operand.

output, of type MLTensorLimits

MLTensorLimits for output operand.

MLOpSupportLimits has the following member for gruCell():

gruCell, of type MLGruCellSupportLimits

Support limits for operator gruCell().

8.9.27. hardSigmoid

  :  {
   = 0.2;
   = 0.5;
};

   {
  ( ,    = {});
};

   {
  ;
};

MLHardSigmoidOptions has the following members:

alpha, of type double, defaulting to 0.2

A scalar multiplier.

beta, of type double, defaulting to 0.5

A scalar addition.

MLOpSupportLimits has the following member for hardSigmoid():

hardSigmoid, of type MLSingleInputSupportLimits

Support limits for operator hardSigmoid().

8.9.28. hardSwish

   {
  ( ,    = {});
};

   {
  ;
};

MLOpSupportLimits has the following member for hardSwish():

hardSwish, of type MLSingleInputSupportLimits

Support limits for operator hardSwish().

8.9.29. instanceNormalization

  :  {
  ;
  ;
   = 1e-5;
   = "nchw";
};

   {
  (
  ,
    = {});
};

  {
  ;
  ;
  ;
  ;
};

   {
  ;
};

MLInstanceNormalizationOptions has the following members:

scale, of type MLOperand

The 1-D tensor of the scaling values whose size is equal to the number of channels, i.e. the size of the feature dimension of the input. For example, for an input tensor with "nchw" layout, the size is equal to input’s shape[1].

bias, of type MLOperand

The 1-D tensor of the bias values whose size is equal to the size of the feature dimension of the input. For example, for an input tensor with "nchw" layout, the size is equal to input’s shape[1].

epsilon, of type double, defaulting to 1e-5

A small value to prevent computational error due to divide-by-zero.

layout, of type MLInputOperandLayout, defaulting to "nchw"

The layout format of the input.

MLNormalizationSupportLimits has the following members:

input, of type MLTensorLimits

MLTensorLimits for input operand.

scale, of type MLTensorLimits

MLTensorLimits for scale operand.

bias, of type MLTensorLimits

MLTensorLimits for bias operand.

output, of type MLTensorLimits

MLTensorLimits for output operand.

MLOpSupportLimits has the following member for instanceNormalization():

instanceNormalization, of type MLNormalizationSupportLimits

Support limits for operator instanceNormalization().

8.9.30. layerNormalization

  :  {
  ;
  ;
 <[] > ;
   = 1e-5;
};

   {
  ( ,
    = {});
};

   {
  ;
};

MLLayerNormalizationOptions has the following members:

scale, of type MLOperand

The N-D tensor of the scaling values whose shape is determined by the axes member in that each value in axes indicates the dimension of the input tensor with scaling values. For example, for an axes values of [1,2,3], the shape of this tensor is the list of the corresponding sizes of the input dimension 1, 2 and 3. When this member is not present, the scaling value is assumed to be 1.

bias, of type MLOperand

The N-D tensor of the bias values whose shape is determined by the axes member in that each value in axes indicates the dimension of the input tensor with bias values. For example, for an axes values of [1,2,3], the shape of this tensor is the list of the corresponding sizes of the input dimension 1, 2 and 3. When this member is not present, the bias value is assumed to be 0.

axes, of type sequence<[EnforceRange] unsigned long>

The indices to the input dimensions to reduce. When this member is not present, it is treated as if all dimensions except the first were given (e.g. for a 4-D input tensor, axes = [1,2,3]). That is, the reduction for the mean and variance values are calculated across all the input features for each independent batch. If empty, no dimensions are reduced.

epsilon, of type double, defaulting to 1e-5

A small value to prevent computational error due to divide-by-zero.

MLOpSupportLimits has the following member for layerNormalization():

layerNormalization, of type MLNormalizationSupportLimits

Support limits for operator layerNormalization().

8.9.31. leakyRelu

  :  {
   = 0.01;
};

   {
  ( ,    = {});
};

   {
  ;
};

MLLeakyReluOptions has the following members:

alpha, of type double, defaulting to 0.01

A scalar multiplier.

MLOpSupportLimits has the following member for leakyRelu():

leakyRelu, of type MLSingleInputSupportLimits

Support limits for operator leakyRelu().

8.9.32. linear

  :  {
   = 1;
   = 0;
};

   {
  ( ,    = {});
};

   {
  ;
};

MLLinearOptions has the following members:

alpha, of type double, defaulting to 1

A scalar multiplier.

beta, of type double, defaulting to 0

A scalar addition.

MLOpSupportLimits has the following member for linear():

linear, of type MLSingleInputSupportLimits

Support limits for operator linear().

8.9.33. lstm

  {
 , // input-output-forget-cell gate ordering
  // input-forget-cell-output gate ordering
};

  :  {
  ;
  ;
  ;
  ;
  ;
   = ;
   = "forward";
   = "iofg";
 <> ;
};

   {
 <> ( ,
  ,
  ,
 []  ,
 []  ,
    = {});
};

  {
  ;
  ;
  ;
  ;
  ;
  ;
  ;
  ;
  ;
  ;
  ;
};

   {
  ;
};

MLLstmOptions has the following members:

bias, of type MLOperand

The 2-D input bias tensor of shape [numDirections, 4 * hiddenSize]. The ordering of the bias vectors in the second dimension of the tensor shape is specified according to layout.

recurrentBias, of type MLOperand

The 2-D recurrent bias tensor of shape [numDirections, 4 * hiddenSize]. The ordering of the bias vectors in the first dimension of the tensor shape is specified according to layout.

peepholeWeight, of type MLOperand

The 2-D weight tensor for peepholes of shape [numDirections, 3 * hiddenSize]. The pack ordering of the weight vectors is for the input (i), output (o), and forget (f) gate, respectively.

initialHiddenState, of type MLOperand

The 3-D initial hidden state tensor of shape [numDirections, batchSize, hiddenSize]. When not specified, implementations must use a tensor filled with zero.

initialCellState, of type MLOperand

The 3-D initial hidden state tensor of shape [numDirections, batchSize, hiddenSize]. When not specified, implementations must use a tensor filled with zero.

returnSequence, of type boolean, defaulting to false

Indicates whether to also return the entire sequence with every output from each time step in it in addition to the output of the last time step.

direction, of type MLRecurrentNetworkDirection, defaulting to "forward"

The processing direction of the input sequence. When set to "both", the size of the first dimension of the weight and the bias tensor shapes must be 2, and the input is processed in both directions.

layout, of type MLLstmWeightLayout, defaulting to "iofg"

The ordering of the weight and bias vectors for the internal gates of LSTM, specifically the input (i), output (o), forget (f), and cell (g) gate, as indicated in the first dimension of the weight and bias tensor shapes.

activations, of type sequence<MLRecurrentNetworkActivation>

A list of three activation functions, the first one is used for the input (i), forget (f), and output (o) gate, the second one is used for the cell (g) gate, and the last used for filtering the output cell state before combining it with the result of the output gate to form the output hidden state. When not specified, defaults to a sequence of the "sigmoid", "tanh", and "tanh" functions, respectively.

MLLstmSupportLimits has the following members:

input, of type MLTensorLimits

MLTensorLimits for input operand.

weight, of type MLTensorLimits

MLTensorLimits for weight operand.

recurrentWeight, of type MLTensorLimits

MLTensorLimits for recurrentWeight operand.

bias, of type MLTensorLimits

MLTensorLimits for bias operand.

recurrentBias, of type MLTensorLimits

MLTensorLimits for recurrentBias operand.

peepholeWeight, of type MLTensorLimits

MLTensorLimits for peepholeWeight operand.

initialHiddenState, of type MLTensorLimits

MLTensorLimits for initialHiddenState operand.

initialCellState, of type MLTensorLimits

MLTensorLimits for initialCellState operand.

output0, of type MLTensorLimits

MLTensorLimits for all the output operands[0].

output1, of type MLTensorLimits

MLTensorLimits for all the output operands[1].

output2, of type MLTensorLimits

MLTensorLimits for all the output operands[2].

MLOpSupportLimits has the following member for lstm():

lstm, of type MLLstmSupportLimits

Support limits for operator lstm().

8.9.34. lstmCell

  :  {
  ;
  ;
  ;
   = "iofg";
 <> ;
};

   {
 <> ( ,
  ,
  ,
  ,
  ,
 []  ,
    = {});
};

  {
  ;
  ;
  ;
  ;
  ;
  ;
  ;
  ;
  ;
  ;
};

   {
  ;
};

MLLstmCellOptions has the following members:

bias, of type MLOperand

The 1-D input bias tensor of shape [4 * hiddenSize]. The ordering of the bias vectors in the first dimension of the tensor shape is specified according to layout.

recurrentBias, of type MLOperand

The 1-D recurrent bias tensor of shape [4 * hiddenSize]. The ordering of the bias vectors in the first dimension of the tensor shape is specified according to layout.

peepholeWeight, of type MLOperand

The 1-D weight tensor for peepholes of shape [3 * hiddenSize]. The pack ordering of the weight vectors is for the input (i), output (o), and forget (f) gate, respectively.

layout, of type MLLstmWeightLayout, defaulting to "iofg"

The ordering of the weight and bias vectors for the internal gates of LSTM, specifically the input (i), output (o), forget (f), and cell (g) gate, as indicated in the first dimension of the weight and bias tensor shapes.

activations, of type sequence<MLRecurrentNetworkActivation>

A list of three activation functions, the first one is used for the input (i), forget (f), and output (o) gate, the second one is used for the cell (g) gate, and the last used for filtering the output cell state before combining it with the result of the output gate to form the output hidden state. When not specified, defaults to a sequence of the "sigmoid", "tanh", and "tanh" functions, respectively.

MLLstmCellSupportLimits has the following members:

input, of type MLTensorLimits

MLTensorLimits for input operand.

weight, of type MLTensorLimits

MLTensorLimits for weight operand.

recurrentWeight, of type MLTensorLimits

MLTensorLimits for recurrentWeight operand.

hiddenState, of type MLTensorLimits

MLTensorLimits for hiddenState operand.

cellState, of type MLTensorLimits

MLTensorLimits for cellState operand.

bias, of type MLTensorLimits

MLTensorLimits for bias operand.

recurrentBias, of type MLTensorLimits

MLTensorLimits for recurrentBias operand.

peepholeWeight, of type MLTensorLimits

MLTensorLimits for peepholeWeight operand.

output0, of type MLTensorLimits

MLTensorLimits for all the output operands[0].

output1, of type MLTensorLimits

MLTensorLimits for all the output operands[1].

MLOpSupportLimits has the following member for lstmCell():

lstmCell, of type MLLstmCellSupportLimits

Support limits for operator lstmCell().

8.9.35. matmul

   {
  ( ,  ,    = {});
};

   {
  ;
};

MLOpSupportLimits has the following member for matmul():

matmul, of type MLBinarySupportLimits

Support limits for operator matmul().

8.9.36. pad

  {
 ,
 ,
 
};

  :  {
   = "constant";
   = 0;
};

   {
  ( ,
 <[] > ,
 <[] > ,
    = {});
};

   {
  ;
};

MLPadOptions has the following members:

mode, of type MLPaddingMode, defaulting to "constant"

The different ways to pad the tensor.

value, of type MLNumber, defaulting to 0

The padding value when mode is set to "constant".

MLOpSupportLimits has the following member for pad():

pad, of type MLSingleInputSupportLimits

Support limits for operator pad().

8.9.37. Pooling operations

  {
 ,
 
};

  :  {
 <[] > ;
 <[] > ;
 <[] > ;
 <[] > ;
   = "nchw";
   = "floor";
 <[] > ;
};

   {
  ( ,    = {});
  ( ,    = {});
  ( ,    = {});
};

   {
  ;
  ;
  ;
};

MLPool2dOptions has the following members:

windowDimensions, of type sequence<[EnforceRange] unsigned long>

A list of length 2: [windowHeight, windowWidth]. Specifies the dimensions of the sliding window. The default value for the window dimensions are the height and width dimensions of the input shape.

padding, of type sequence<[EnforceRange] unsigned long>

A list of length 4: [beginningHeight, endingHeight, beginningWidth, endingWidth]. Specifies the additional rows and columns added to the beginning and ending of each spatial dimension of the convolution input. The default value is [0,0,0,0].

strides, of type sequence<[EnforceRange] unsigned long>

A list of length 2: [strideHeight, strideWidth]. Specifies the stride of the sliding window for each spatial dimension of the convolution input. The default value is [1,1].

dilations, of type sequence<[EnforceRange] unsigned long>

A list of length 2: [dilationHeight, dilationWidth]. Specifies the dilation factor for each spatial dimension applied on the convolution filter (kernel). The default value is [1,1].

layout, of type MLInputOperandLayout, defaulting to "nchw"

Specifies the layout format of the input and output tensor as follows:

• "nchw"

  • input tensor: [batches, inputChannels, height, width]

  • output tensor: [batches, outputChannels, height, width]

• "nhwc":

  • input tensor: [batches, height, width, inputChannels]

  • output tensor: [batches, height, width, outputChannels]

outputShapeRounding, of type MLRoundingType, defaulting to "floor"

The rounding function used to compute the output shape, depending on whether full or partial window results are desired.

outputSizes, of type sequence<[EnforceRange] unsigned long>

A list of length 2: [outputHeight, outputWidth] Specifies the sizes of the two spatial dimensions of the output tensor. When the output sizes are explicitly specified, the outputShapeRounding is ignored. If not specified, the output sizes are automatically computed.

MLOpSupportLimits has the following members for pooling operations:

averagePool2d, of type MLSingleInputSupportLimits

Support limits for operator averagePool2d().

l2Pool2d, of type MLSingleInputSupportLimits

Support limits for operator l2Pool2d().

maxPool2d, of type MLSingleInputSupportLimits

Support limits for operator maxPool2d().

buildermaxPool2dinput

8.9.37.1. averagePool2d

8.9.37.2. l2Pool2d

8.9.37.3. maxPool2d

8.9.38. prelu

The operation will be broadcast according to [numpy-broadcasting-rule]. The input tensors must be bidirectionally broadcastable. The rank of the output tensor is the maximum rank of the input tensors. For each dimension of the output tensor, its size is the maximum size along that dimension of the input tensors.

   {
  ( ,
  ,
    = {});
};

  {
  ;
  ;
  ;
};

   {
  ;
};

MLPreluSupportLimits has the following members:

input, of type MLTensorLimits

MLTensorLimits for input operand.

slope, of type MLTensorLimits

MLTensorLimits for slope operand.

output, of type MLTensorLimits

MLTensorLimits for output operand.

MLOpSupportLimits has the following member for prelu():

prelu, of type MLPreluSupportLimits

Support limits for operator prelu().