whcao
whcao
Hi! Maybe it is a little hard to use FasterRCNN as a teacher to distill a yolo v3 student. To the best of my knowledge, there are no available algorithms...
It's possible to distill between one-stage and two-stage networks, ref to https://arxiv.org/abs/2207.02039. For example, with a MaskRCNN-Swin detector as the teacher, ResNet-50 based RetinaNet and FCOS achieve 41.5% and 43.9%...
Hi! Sorry for the inconvenience to you. We split the checkpoint based on `algorithm.pruner.deploy_subnet` (refer to [here](https://github.com/open-mmlab/mmrazor/blob/3cc359e364601b2be52342f7ab030a8a76601968/tools/model_converters/split_checkpoint.py#L58)). When overwriting the original parameter with the sliced parameter of a `nn.Module`, it...
Hi! Thank you for your issue. Could you provide your pytorch version?
I'm sorry for the inconvenience caused to you. The auto-trace in pruner goes wrong with pytorch 1.11.0 and we will fix the errors in codes as soon as possible. In...
I'm sorry for the late reply and the inconvenience to you. You are right, and `in_channels`, `out_channels` and `groups` should be all the same in depth-wise conv. Are you interested...
Pr Submission Guidelines: For English: https://mmcv.readthedocs.io/zh_CN/latest/community/pr.html and https://mmcv.readthedocs.io/zh_CN/latest/community/contributing.html For Chinese: https://github.com/open-mmlab/mmcv/blob/master/CONTRIBUTING.md PR welcome！
Hi! I print the node2parents, as shown below. Parents of the concat node (``concat_2430076418760_2430076418888``) are ``backbone.stage4.2.blocks.3.conv2.conv`` and ``backbone.stage4.2.short_conv.conv``, and perhaps no modules are mistakenly ignored. 
Yes, I use the latest code. If there are still problems with 'concat' node, pls let me know.
Sorry for the inconvenience. Two new backward ops are added in torch 1.11 and 1.12. Please add the following two lines to `https://github.com/open-mmlab/mmrazor/blob/master/mmrazor/models/pruners/structure_pruning.py#L750` and try again: ``` @register_parser(BACKWARD_PARSER_DICT, 'ConvolutionBackward') @register_parser(BACKWARD_PARSER_DICT,...