Jisoo

I has sloved this problem!when i run bash compile_op.sh,it still occur error，you should to check if It generate file that neareast_neighbors.cpython-37m-x86_64-linux-gnu.so in utils/nearest_neighbors/lib/python .The file is the same as your...

@huixiancheng thanks,has solved!

> At first , grid_sample in torch seems can't be converted to onnx, So I use belinear_grid_sample in mmcv to replace it. I succeed creating a .onnx file ,but when...

> Thanks for your great work and code release. I have a question about training process. In [this ](https://github.com/liuquande/SAML/blob/263764b9fcdb566864a31eedb347ac4674ca77ea/train.py#L41) why num_meta_train is 2 ? **Totally we have 5 source domains...

> ``` > 您好，我自己用完整的nuScenes数据集训练，显卡是两张3090， BATCH_SIZE_PER_GPU=8，用的是‘cbgs_voxel0075_voxelnext.yaml’配置文件。训练时间稳定的话大概是200个小时，这是正常的吗。 > ``` 你好，请问这个问题解决了吗，我也遇到了同样的问题，训练时间非常长！

> 你好，我有个问题，特征动态聚合模块的代码在哪里？ 我在 LoGoHead_kitti.py 中看到了代码 > > pooled_features = pooled_features + localgrid_densityfeat_fuse.permute(0, 2, 1) 和tention_output = pooled_features +tention_output > > 如果此代码表示三个特征 F p B + F l B + F...