Sub (CPU)
=========

* Domain: ``ai.onnx``
* Device: ``CPU``
* Kernel: ``onnx_light_cpu::Sub``
* Supported types: ``FLOAT``, ``DOUBLE``, ``FLOAT16``, ``BFLOAT16``, ``INT8``, ``INT16``, ``INT32``, ``INT64``, ``UINT8``, ``UINT16``, ``UINT32``, ``UINT64``
* Opset bounds: since opset 7

Operator support
----------------

The standard ONNX schema and any standard shape inference come from onnx-light. No additional shape, peak-memory, fusion-pattern, or gradient implementation is registered by onnx-light-cpu for this operator.