Tina Jung

DequantizeLinear spec clarification: What happens if the subtraction overflows/underflows?

1

# Ask a Question ### Question [Dequantize linear](https://onnx.ai/onnx/operators/onnx__DequantizeLinear.html) defines the dequantization formula as `y = (x - x_zero_point) * x_scale` and that `x` and `x_zero_point` must have the same type...

Missing type support in parser for various types (float16, bfloat16, ...)

2

# Bug Report ### Is the issue related to model conversion? No. ### Describe the bug The parser has only partial support for data type parsing: https://github.com/onnx/onnx/blob/093a8d335a66ea136eb1f16b3a1ce6237ee353ab/onnx/defs/parser.cc#L436 The missing types...

## Issue I encounter an error when lowering a small onnx model that uses a pad with `reflect` mode to linalg. The error is the following: ``` error: failed to...