Depthwise Convolution and space to Batch conversion

eva211 · June 7, 2019, 6:07am

Is depthwise convolution in TensorRT implemented making use of space to Batch conversion?
I am aware that Dilated convolutions make use of SpaceToBatch conversion in tensorflow.
Can anyone provide an insight into how depthwise layer is implemented in TensorRT.
I am trying to implement MobileNetv1 in TRT 5.0 and it is working fine. Wanted to know more about the internal working of depthwise convolution.

Topic		Replies	Views
When will dilated convolution be supported in TensorRT (Python), or SpaceToBatchND / BatchToSpaceND TensorRT	6	3305	June 28, 2019
TensorRT 3 RC and grouped convolutions TensorRT	6	3630	October 30, 2018
Does 3D convolution/deconvolution supportted in TensorRT ? TensorRT	0	558	June 4, 2019
Using dynamic shape with space_to_depth implementation TensorRT tensorrt	2	477	October 12, 2021
3D convolution/deconvolution support in TensorRT 3.0 ? TensorRT	4	1679	February 20, 2019
Is Deformable Convolution supported in TensorRT? TensorRT	4	3913	October 12, 2021
about Space to Batch and Batch to Space operation Jetson Nano	7	1399	October 18, 2021
question on 1x1 conv acceleration of tensorRT TensorRT	5	1037	August 1, 2019
TensorRT memory layout TensorRT	1	1209	November 25, 2019
Tensor RT implements convolution efficiently GPU-Accelerated Libraries	0	660	July 5, 2017

Depthwise Convolution and space to Batch conversion

Related topics