songh11

ImportError: cannot import name '_C'

13

作者你好，当我运行 cd siamreppoints/tools python test.py \ --snapshot ./snapshot/siamreppoints.model \ #model path --dataset VOT2018 \ #dataset name --config ./experiments/siamreppoints/config_vot2018_offline.yaml #config file 这里时，一直会报这个错，请问这个是哪里没有配置好吗还是其他原因，谢谢

您好，我已经改成了这个，Cuda compilation tools, release 9.0, V9.0.176 请问为什么还是会报错，有什么解决办法吗

DiT 在 1.1 版本推理出错

3

报错信息： RuntimeError: Error(s) in loading state_dict for DiT: size mismatch for x_embedder.proj.weight: copying a param with shape torch.Size([1152, 4, 2, 2]) from checkpoint, the shape in current model is torch.Size([1152,...

使用静态图推理时，当输入数量改变，静态图必须要重编吗，请问如何可以不用重编呢

2

### Description 就比如 SD 的 controlnet，我们修改 pipeline 使当 conditioning_scale 为 0 时，使 down_block_res_samples 和 mid_block_res_sample 为 None，这样会导致 unet 的输入数量改变而重新编译静态图，请问有办法不重编么

It doesn't accelerate very well at L4

1

I'm glad the torch.compile is speeding up very quickly. On A5000 it can speed up 60%, but there's no acceleration at l4. I want to know why is it happen?...

数据集特征点

2

作者您好，请问x1, y1, x2, y2, x3, y3, x4, y4需要和图片宽高做归一化吗，另外这四个点的标注要有顺序么

I found openmp is slow than gcc. When I read the code, I saw that there was a data access conflict. I wonder if you have compared the speed, hope...

I'm trying to convert sdxl unet to onnx, but it's been 30 hours and no success.Have you ever tried sdxl or how do I estimate export onnx-model time.

Great job❤️! Does this method works for controlnet?