| --- |
| library_name: kernels |
| license: apache-2.0 |
| tags: |
| - kernel |
| - webgpu |
| - wgsl |
| --- |
| # ai.onnx.ReduceMax |
|
|
| `ai.onnx` · standard ONNX operator · ONNX opset ≥ 20 |
|
|
| ## Description |
|
|
| Computes the maximum of input tensor elements along the specified axes. The output rank matches the input when `keepdims` is 1; reduced dimensions are pruned when `keepdims` is 0. Reduction over an empty set yields negative infinity when the dtype supports it, or the dtype's minimum value otherwise. For Boolean inputs, `false` is less than `true`. |
|
|
| See the [ONNX `ReduceMax` spec](https://onnx.ai/onnx/operators/onnx__ReduceMax.html) for the reference semantics. |
|
|
| ## Inputs |
|
|
| | Name | Bind key | Logical dtype | Rank | Shape | Description | Presence | |
| | --- | --- | --- | --- | --- | --- | --- | |
| | `data` | `x` | `T` | — | — | The input tensor to reduce. | required | |
|
|
| ## Outputs |
|
|
| | Name | Bind key | Logical dtype | Rank | Shape | Description | Presence | |
| | --- | --- | --- | --- | --- | --- | --- | |
| | `reduced` | `y` | `T` | derived | — | The reduced output tensor containing the maximum values. | required | |
|
|
| ## Attributes |
|
|
| Default values (overridable per request): |
|
|
| | Attribute | Default | Description | |
| | --- | --- | --- | |
| | `keepdims` | `1` | Whether to retain reduced dimensions in the output with size 1 (1) or prune them (0). | |
| | `noop_with_empty_axes` | `0` | When 1 and `axes` is empty, the op acts as an identity (no-op); when 0 (default), reduction happens over all axes. | |
| | `axes` | `[]` | Values of the optional ONNX `axes` tensor input, supplied through this request attribute; an empty list follows `noop_with_empty_axes`. | |
|
|
| ## Type constraints |
|
|
| | Variable | Allowed dtypes | |
| | --- | --- | |
| | `T` | `float32`, `float16`, `int32`, `uint32`, `int8`, `uint8`, `bool` | |
|
|
| ## Device requirements |
|
|
| Some implementation variants require `subgroups`. These are route-specific capabilities, not package-wide requirements; availability also depends on the request shape and dtype. |
|
|
| ## Files |
|
|
| - [`metadata.json`](build/webgpu/metadata.json) — kernel metadata (id, digests, 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 + tuning cases |
| - [`reduce-axis-split-reduce.wgsl.jinja`](build/webgpu/reduce-axis-split-reduce.wgsl.jinja) |
| - [`reduce-axis0-splitk-combine.wgsl.jinja`](build/webgpu/reduce-axis0-splitk-combine.wgsl.jinja) |
| - [`reduce-axis0-splitk-reduce.wgsl.jinja`](build/webgpu/reduce-axis0-splitk-reduce.wgsl.jinja) |
| - [`reduce-axis0-tilecols.wgsl.jinja`](build/webgpu/reduce-axis0-tilecols.wgsl.jinja) |
| - [`reduce-flat-partial.wgsl.jinja`](build/webgpu/reduce-flat-partial.wgsl.jinja) |
| - [`reduce-narrow-empty-identity.wgsl.jinja`](build/webgpu/reduce-narrow-empty-identity.wgsl.jinja) |
| - [`reduce-noop-empty-axes.wgsl.jinja`](build/webgpu/reduce-noop-empty-axes.wgsl.jinja) |
| - [`reduce-row-subgroup.wgsl.jinja`](build/webgpu/reduce-row-subgroup.wgsl.jinja) |
| - [`reduce-row-tree.wgsl.jinja`](build/webgpu/reduce-row-tree.wgsl.jinja) |
| - [`reduce-serial-axis.wgsl.jinja`](build/webgpu/reduce-serial-axis.wgsl.jinja) |
|
|
| ## Use with `@huggingface/kernels` |
|
|
| The loader automatically allocates outputs whose metadata it can derive from the manifest contract and this call. |
|
|
| The explicit `outputs` entries provide shape and logical dtype metadata for the results listed below: |
|
|
| - `y` |
|
|
| Each entry either requests an optional result or supplies metadata that cannot be inferred from the inputs. |
|
|
| The `version: 1` option selects the published kernel contract; it is independent of any operator opset, contrib `since_version`, or model 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.ReduceMax", { version: 1 }); |
| // Explicit destinations request optional results or supply metadata that cannot be inferred. |
| const { y } = await kernel({ x: { data: xData, shape: [] } }, { |
| outputs: { y: { shape: [], dtype: "float32" } }, |
| }); |
| ``` |
|
|
