Error during training tgnet_fps

[FengZhongLiuDong]

Traceback (most recent call last): File "start_train.py", line 59, in <module> runner(config, model) File "/root/ToothGroupNetwork/runner.py", line 57, in runner trainner.run() File "/root/ToothGroupNetwork/trainer.py", line 100, in run self.train(epoch, train_data_loader) File "/root/ToothGroupNetwork/trainer.py", line 31, in train loss = self.model.step(batch_idx, batch_item, "train") File "/root/ToothGroupNetwork/models/fps_grouping_network_model.py", line 37, in step output = self.module(inputs) File "/root/miniconda3/lib/python3.8/site-packages/torch/nn/modules/module.py", line 727, in _call_impl result = self.forward(*input, **kwargs) File "/root/ToothGroupNetwork/models/modules/grouping_network_module.py", line 72, in forward nn_crop_indexes = ou.get_nearest_neighbor_idx(org_xyz_cpu, cluster_centroids, self.config["model_parameter"]["crop_sample_size"]) File "/root/ToothGroupNetwork/ops_utils.py", line 159, in get_nearest_neighbor_idx indexs = tree.query(sampled_clusters[batch_idx], k=crop_num, return_distance = False) File "sklearn/neighbors/_binary_tree.pxi", line 1106, in sklearn.neighbors._kd_tree.BinaryTree.query File "/root/miniconda3/lib/python3.8/site-packages/sklearn/utils/validation.py", line 938, in check_array raise ValueError( ValueError: Expected 2D array, got 1D array instead: array=[]. Reshape your data either using array.reshape(-1, 1) if your data has a single feature or array.reshape(1, -1) if it contains a single sample.
Configuration information：
NVIDIA GeForce RTX 4060 Ti
cuda 11.8 pytorch 2.1
At first, I thought it was due to the environment, but I switched to the same environment as the author.But the same error occurred。
NVIDIA GeForce RTX 2080 Ti
cuda 11.0 pytorch 1.7
After reviewing others' issues, I recompiled pointops (https://github.com/POSTECH-CVLab/point-transformer), but the problem still persists
Help!

[FengZhongLiuDong]

[FengZhongLiuDong]

[rebecca0307ss]

Hi, I'm having the same problem as you, have you solved the problem now?