| --- |
| library_name: kernels |
| license: apache-2.0 |
| tags: |
| - kernel |
| - webgpu |
| - wgsl |
| --- |
| # ai.onnx.Tile |
|
|
| `ai.onnx` · standard ONNX operator · ONNX opset ≥ 13 |
|
|
| ## Description |
|
|
| Constructs a tensor by tiling a given tensor: each dimension `i` of the input is repeated `repeats[i]` times, so `output_dim[i] = input_dim[i] * repeats[i]`. Equivalent to NumPy `tile` but without broadcasting. |
|
|
| See the [ONNX `Tile` spec](https://onnx.ai/onnx/operators/onnx__Tile.html) for the reference semantics. |
|
|
| ## Inputs |
|
|
| | Name | Logical dtype | WebGPU storage | Rank | Shape | Description | Presence | |
| | --- | --- | --- | --- | --- | --- | --- | |
| | `input` | `T` | runtime-selected; narrow integers and bool use 32-bit slots | — | — | Input tensor of any shape. | required | |
| | `repeats` | `S` | `uint32` | `1` | — | Logical int64 1-D tensor of length equal to the input rank, specifying non-negative repeat counts stored as uint32 by WebGPU. | required | |
|
|
| ## Outputs |
|
|
| | Name | Logical dtype | Rank | Shape | Description | Presence | |
| | --- | --- | --- | --- | --- | --- | |
| | `output` | `T` | same as `input` | — | Output tensor of the same type as the input, with each dimension scaled by the corresponding repeat count. | required | |
|
|
| ## Type constraints |
|
|
| | Variable | Allowed dtypes | |
| | --- | --- | |
| | `T` | `float32`, `float16`, `uint32`, `int32`, `int16`, `uint8`, `int8`, `bool`, `int64` | |
| | `S` | `int64` | |
|
|
| ## Files |
|
|
| - [`metadata.json`](build/webgpu/metadata.json) — kernel metadata (id, digests, per-variant templates, provenance) |
| - [`manifest.json`](build/webgpu/manifest.json) — the op contract (source of truth) |
| - [`test.json`](build/webgpu/test.json) — correctness cases |
| - [`bench.json`](build/webgpu/bench.json) — benchmark cases |
| - [`datamove-tile-vec4.wgsl.jinja`](build/webgpu/datamove-tile-vec4.wgsl.jinja) |
| - [`tile.wgsl.jinja`](build/webgpu/tile.wgsl.jinja) |
|
|
| ## Use with `@huggingface/kernels` |
|
|
| ```sh |
| npm install --save-exact @huggingface/kernels@0.0.1-preview.3 |
| ``` |
|
|
| Outputs with inferable metadata are allocated automatically. Explicit `outputs` entries request optional results or provide metadata that cannot be inferred from the supplied inputs and attributes. |
|
|
| This example supplies explicit metadata for: |
|
|
| - `output` |
|
|
| 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. |
|
|
| ```js |
| import { getKernel } from "@huggingface/kernels"; |
| |
| const kernel = await getKernel("webgpu-kernels/ai.onnx.Tile", { version: 1 }); |
| // Explicit destinations request optional results or supply metadata that cannot be inferred. |
| const { output } = await kernel({ |
| input: { data: inputData, shape: [3] }, |
| repeats: { data: repeatsData, shape: [1] }, |
| }, { |
| outputs: { output: { shape: [6], dtype: "int64" } }, |
| }); |
| ``` |
|
|
