List view

[quant] Quantized AdaptivePool3d is much slower for ChannelsLast3d.

low priorityWe're unlikely to get around to doing this in the near futureWe're unlikely to get around to doing this in the near future

module: memory formatMemory format/layout related issues/changes (channels_last, nhwc)Memory format/layout related issues/changes (channels_last, nhwc)

module: performanceIssues related to performance, either of kernel code or framework glueIssues related to performance, either of kernel code or framework glue

oncall: quantizationQuantization support in PyTorchQuantization support in PyTorch

triagedThis issue has been looked at a team member, and triaged and prioritized into an appropriate moduleThis issue has been looked at a team member, and triaged and prioritized into an appropriate module

Status: Open.

#42779 In pytorch/pytorch;

· thinking-tower opened on Aug 8, 2020