[Relay][Frontend][TFLite] Add parser support for squared difference #4652
Conversation
|
Could you check this PR, @kevinthesun @FrozenGene |
| @@ -723,6 +723,14 @@ def _test_greater(data): | |||
| """ One iteration of greater """ | |||
| return _test_elemwise(math_ops.greater, data) | |||
|
|
|||
| ####################################################################### | |||
| # Squared_difference | |||
| # ------- | |||
There was a problem hiding this comment.
Align - the same length of squared_difference
python/tvm/relay/frontend/tflite.py
Outdated
| # Check if the input tensor is quantized, call QNN op | ||
| if self.is_quantized(op): | ||
| raise tvm.error.OpNotImplemented( | ||
| 'TFlite quantized greater operator is not supported yet.') |
There was a problem hiding this comment.
comment is wrong. Not greater op.
python/tvm/relay/frontend/tflite.py
Outdated
| raise tvm.error.OpNotImplemented( | ||
| 'TFlite quantized greater operator is not supported yet.') | ||
| difference = self._convert_elemwise(_op.subtract, op) | ||
| out = _op.multiply(difference, difference) |
There was a problem hiding this comment.
maybe could consider using power.
python/tvm/relay/frontend/tflite.py
Outdated
| raise tvm.error.OpNotImplemented( | ||
| 'TFlite quantized squared difference operator is not supported yet.') | ||
| difference = self._convert_elemwise(_op.subtract, op) | ||
| out = _op.power(difference, relay.const(2, 'float32')) |
There was a problem hiding this comment.
It it better not make it be relay.const(2, 'float32')). Because let us imagine if op's data is fp16 (in the future), 2 shouldn't be float32. So we could do like this: call get_tensor_type_str(self.get_output_tensors(op)[0].tensor.Type()) to get type, then call _op.power. Or if you think it is complex, you could simply call _op.multiple(difference, difference), I think it is OK. Previously, I mention _op.power, because we could have intrinsic of power to handle it. However, I think it shouldn't be bottleneck after thinking twice.
There was a problem hiding this comment.
Thanks for your suggestion. I think _op.power is more appropriate here and I have already fixed it.
|
Thanks @wyc-ruiker LGTM now. Let us wait CI's result. |
|
Thanks @wyc-ruiker This is merged. |
FrozenGene
added
status: accepted
and removed
status: need update
…pache#4652) * [Relay][Frontend][TFLite] Add parser support for squared difference * fix some error * fix exp_type * add comment
…pache#4652) * [Relay][Frontend][TFLite] Add parser support for squared difference * fix some error * fix exp_type * add comment
…pache#4652) * [Relay][Frontend][TFLite] Add parser support for squared difference * fix some error * fix exp_type * add comment
Thanks for contributing to TVM! Please refer to guideline https://docs.tvm.ai/contribute/ for useful information and tips. After the pull request is submitted, please request code reviews from Reviewers by @ them in the pull request thread.