ai.onnx.Trilu

ai.onnx · standard ONNX operator · ONNX opset ≥ 14

Description

Returns the upper or lower triangular part of a 2-D matrix or batch of 2-D matrices of shape [*, N, M], zeroing all other elements. The diagonal offset k shifts the boundary: positive values move it above the main diagonal, negative values below it.

See the ONNX Trilu spec for the reference semantics.

Inputs

Outputs

Attributes

Default values (overridable per request):

Type constraints

Implementation variants

One implementation is selected per call from the device capabilities, the request shapes and the dtypes; these notes say what each one covers.

• mask_inner_packed_vec8_kinput — Processes eight float16 elements as four 32-bit storage words. Integer masks preserve retained bits and zero discarded halves; fully masked vectors skip the input load.
• mask_inner_packed_vec8 — Processes eight float16 elements as four 32-bit storage words. Integer masks preserve retained bits and zero discarded halves; fully masked vectors skip the input load.

Files

• metadata.json — kernel metadata (id, digests, per-variant templates, provenance)
• manifest.json — the op contract (source of truth)
• test.json — correctness cases
• bench.json — benchmark cases
• trilu-packed.wgsl.jinja
• trilu.wgsl.jinja

Use with @huggingface/kernels

npm install --save-exact @huggingface/kernels@0.0.1-preview.3

Required output shapes and logical data types are inferred from the supplied inputs and attributes; result tensors are allocated automatically.

The version: 1 option selects the published kernel contract; it is independent of any operator opset, contrib since_version, or model version. It follows the v1 branch as fixes land. To pin exact artifact bytes, pass a 40-character commit revision instead of version.

Replace each *Data placeholder with a typed array containing the corresponding input data.

import { getKernel } from "@huggingface/kernels";

const kernel = await getKernel("webgpu-kernels/ai.onnx.Trilu", { version: 1 });
const { output } = await kernel({ input: { data: inputData, shape: [2, 2] } });