huixiancheng / no-cpgnet Goto Github PK

你好，我在google drive中下载了训练好的模型（wce 和 wce_two_stage）并分别在kitti数据集中进行测评，测评代码是工程中的evaluate.py 使用kitti中的valid集（08），但是测试wce 和 wce_two_stage的结果分别为：
Epoch 44; car iou: 0.92425615; bicycle iou: 0.34408772; motorcycle iou: 0.327422; truck iou: 0.51052713; other-vehicle iou: 0.19114032; person iou: 0.5734159; bicyclist iou: 0.84467614; motorcyclist iou: 0.0; road iou: 0.81974715; parking iou: 0.31249848; sidewalk iou: 0.6598457; other-ground iou: 0.006617649; building iou: 0.8843032; fence iou: 0.48770225; vegetation iou: 0.86043215; trunk iou: 0.60325515; terrain iou: 0.6995845; pole iou: 0.59409064; traffic-sign iou: 0.46759114; mean iou: 0.53216815

Epoch 39; car iou: 0.94990486; bicycle iou: 0.5258709; motorcycle iou: 0.7058376; truck iou: 0.89650655; other-vehicle iou: 0.47812566; person iou: 0.64498734; bicyclist iou: 0.9207215; motorcyclist iou: 0.01826144; road iou: 0.890772; parking iou: 0.31654277; sidewalk iou: 0.71569484; other-ground iou: 0.0074295653; building iou: 0.88886327; fence iou: 0.5177564; vegetation iou: 0.86766565; trunk iou: 0.6517532; terrain iou: 0.74467117; pole iou: 0.6320016; traffic-sign iou: 0.4729658; mean iou: 0.62349117

命令分别为：CUDA_VISIBLE_DEVICES=0 python -m torch.distributed.launch --nproc_per_node=1 evaluate.py --config config/wce.py --start_epoch 44 --end_epoch 44

CUDA_VISIBLE_DEVICES=0 python -m torch.distributed.launch --nproc_per_node=1 evaluate.py --config config/wce_two_stage.py --start_epoch 39 --end_epoch 39
不知道这里面出现了什么问题

Hi Huixian,

Thank you for opening this repo.

When reading the code, in the file datasets/data.py,

    def form_seq(self, meta_list):
        fname_pcd, fname_label, pose_diff, _, _ = meta_list
        # load pcd
        pcds_tmp = np.fromfile(fname_pcd, dtype=np.float32).reshape((-1, 4))
        pcds_ht = utils.Trans(pcds_tmp, pose_diff)

        # load label
        pcds_label = np.fromfile(fname_label, dtype=np.uint32)
        pcds_label = pcds_label.reshape((-1))
        sem_label = pcds_label & 0xFFFF
        inst_label = pcds_label >> 16

Could you tell me what is the function of pcds_ht = utils.Trans(pcds_tmp, pose_diff), cus I visualize the pcds_tmp and pcds_ht , but it seems like they are identical,

Traceback (most recent call last):
File "evaluate.py", line 155, in
main(args, config)
File "evaluate.py", line 125, in main
model.load_state_dict(checkpoint['model_state_dict'], strict=True)
File "/home/liuk/anaconda3/envs/cpgnet/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1482, in load_state_dict
raise RuntimeError('Error(s) in loading state_dict for {}:\n\t{}'.format(
RuntimeError: Error(s) in loading state_dict for AttNet:
Missing key(s) in state_dict: "bev_net.up0.att_layer.0.weight", "bev_net.up0.att_layer.1.weight", "bev_net.up0.att_layer.1.bias", "bev_net.up0.att_layer.1.running_mean", "bev_net.up0.att_layer.1.running_var", "bev_net.up0.att_layer.3.weight", "bev_net.up0.att_layer.3.bias", "bev_net.up0.conv_high.0.weight", "bev_net.up0.conv_high.1.weight", "bev_net.up0.conv_high.1.bias", "bev_net.up0.conv_high.1.running_mean", "bev_net.up0.conv_high.1.running_var", "bev_net.up0.conv_low.0.weight", "bev_net.up0.conv_low.1.weight", "bev_net.up0.conv_low.1.bias", "bev_net.up0.conv_low.1.running_mean", "bev_net.up0.conv_low.1.running_var", "rv_net.up0.att_layer.0.weight", "rv_net.up0.att_layer.1.weight", "rv_net.up0.att_layer.1.bias", "rv_net.up0.att_layer.1.running_mean", "rv_net.up0.att_layer.1.running_var", "rv_net.up0.att_layer.3.weight", "rv_net.up0.att_layer.3.bias", "rv_net.up0.conv_high.0.weight", "rv_net.up0.conv_high.1.weight", "rv_net.up0.conv_high.1.bias", "rv_net.up0.conv_high.1.running_mean", "rv_net.up0.conv_high.1.running_var", "rv_net.up0.conv_low.0.weight", "rv_net.up0.conv_low.1.weight", "rv_net.up0.conv_low.1.bias", "rv_net.up0.conv_low.1.running_mean", "rv_net.up0.conv_low.1.running_var", "bev_net_2.up0.att_layer.0.weight", "bev_net_2.up0.att_layer.1.weight", "bev_net_2.up0.att_layer.1.bias", "bev_net_2.up0.att_layer.1.running_mean", "bev_net_2.up0.att_layer.1.running_var", "bev_net_2.up0.att_layer.3.weight", "bev_net_2.up0.att_layer.3.bias", "bev_net_2.up0.conv_high.0.weight", "bev_net_2.up0.conv_high.1.weight", "bev_net_2.up0.conv_high.1.bias", "bev_net_2.up0.conv_high.1.running_mean", "bev_net_2.up0.conv_high.1.running_var", "bev_net_2.up0.conv_low.0.weight", "bev_net_2.up0.conv_low.1.weight", "bev_net_2.up0.conv_low.1.bias", "bev_net_2.up0.conv_low.1.running_mean", "bev_net_2.up0.conv_low.1.running_var", "rv_net_2.up0.att_layer.0.weight", "rv_net_2.up0.att_layer.1.weight", "rv_net_2.up0.att_layer.1.bias", "rv_net_2.up0.att_layer.1.running_mean", "rv_net_2.up0.att_layer.1.running_var", "rv_net_2.up0.att_layer.3.weight", "rv_net_2.up0.att_layer.3.bias", "rv_net_2.up0.conv_high.0.weight", "rv_net_2.up0.conv_high.1.weight", "rv_net_2.up0.conv_high.1.bias", "rv_net_2.up0.conv_high.1.running_mean", "rv_net_2.up0.conv_high.1.running_var", "rv_net_2.up0.conv_low.0.weight", "rv_net_2.up0.conv_low.1.weight", "rv_net_2.up0.conv_low.1.bias", "rv_net_2.up0.conv_low.1.running_mean", "rv_net_2.up0.conv_low.1.running_var".
ERROR:torch.distributed.elastic.multiprocessing.api:failed (exitcode: 1) local_rank: 0 (pid: 349586) of binary: /home/liuk/anaconda3/envs/cpgnet/bin/python