leoluopy

i am not familiar with detectron2 , how can i use the centermask2 in pure pytorch network defines and weights ?

2

hi @youngwanLEE as the title has described , i am not familiar with the detectron framework . how can i port the centermask network definition into the pytorch pure implementations...

xy and wh calculation

1

i notice that ， while in loss calculation and inference decoding : for x ,y and w,h calculation: (yolo.py) 59: y = y + torch.cat((x[i][:, :, :, :, 0:5].sigmoid(), torch.cat((x[i][:,...

masked Face Recognition Dataset available for download ? or any details of generating ?

3

thanks for the effort to the project . in the report the words below are mentioned: ``` we synthesize the masked fa- cial datasets based on MegaFace by using FMA-3D,...

hi @zouchuhang will this method work well outdoors ?

excuse me , i got a need for faster detection , could you please afford a darknet21 pretrained model in this pytorch implementation ?

How to inference with weights finish_converted.hdf5 ?

5

Hi @DingXiaoH : thanks to your great masterpiece . and as i wanna check the finished_converted.hdf5 that if the acc has dropped after folding conv and fusing bn ? so...

[BUG/Help] <title>dataclasses.FrozenInstanceError: cannot assign to field generation_max_length

9

### Is there an existing issue for this? - [X] I have searched the existing issues ### Current Behavior transformers/training_args.py", line 1712, in __setattr__ raise FrozenInstanceError(f"cannot assign to field {name}")...

differences between v2 and v1 ?

1

what are the main differences between v2 and v1 ? and is the pretrained model afforded ? are they training scales or some thing else ?

除了lora还有没有其他办法可以降低训练时显存的占用量

2

除了lora以及减小模型，还有没有其他办法可以降低训练时显存的占用量？ 哪些实践方法是比较有效的？

有什么方法可以把模型切成多份，放到不同显卡，然后训练，达到增大layer数量的方法吗？

4

如题： 目前训练，需要把模型完整加载在一张显卡里面，显卡资源有限，显存大小不够是个问题。 有什么方法可以把模型切成多份，放到不同显卡，然后训练，达到增大layer数量的方法吗？