Error occurs when run the test.py

[xbxdghx]

Hi，i use the following instructions to run the test.py:
python test.py --cfg_file /home/lbh/HSSDA/tools/cfgs/kitti_models/pv_rcnn_ssl.yaml \ --batch_size 8 \ --ckpt /home/lbh/HSSDA/output/kitti_models/pv_rcnn_ssl/pv_rcnn_002_1/ckpt/checkpoint_epoch_80.pth
then the error message display：
File "/home/lbh/HSSDA/tools/../pcdet/models/detectors/pv_rcnn_ssl.py", line 27, in init
self.label_idx = np.loadtxt(model_cfg['LABELED_FRAME_IDX'])

But i think that pv_rcnn.yaml may also fit the logic of running, because semi-supervised training can be seen as two fully supervised trainings with parameter passes，so，i use the instructions as followed:
'
python test.py --cfg_file /home/lbh/HSSDA/tools/cfgs/kitti_models/pv_rcnn.yaml
--batch_size 8
--ckpt /home/lbh/HSSDA/output/kitti_models/pv_rcnn_ssl/pv_rcnn_002_1/ckpt/checkpoint_epoch_80.pth
'
but the result is not ideal:
2023-12-21 13:37:03,532 INFO Car [email protected], 0.70, 0.70:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Car [email protected], 0.70, 0.70:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Car [email protected], 0.50, 0.50:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Car [email protected], 0.50, 0.50:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Pedestrian [email protected], 0.50, 0.50:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Pedestrian [email protected], 0.50, 0.50:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Pedestrian [email protected], 0.25, 0.25:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Pedestrian [email protected], 0.25, 0.25:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Cyclist [email protected], 0.50, 0.50:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Cyclist [email protected], 0.50, 0.50:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Cyclist [email protected], 0.25, 0.25:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00
Cyclist [email protected], 0.25, 0.25:
bbox AP:0.0000, 0.0000, 0.0000
bev AP:0.0000, 0.0000, 0.0000
3d AP:0.0000, 0.0000, 0.0000
aos AP:0.00, 0.00, 0.00

Plz tell me how to use test.py or your Reasoning process.

[azhuantou]

Hi @xbxdghx

You cannot test the model trained with HSSDA in this way because the weight parameters of PVRCNN and HSSDA do not match. The occurrence of zero results is due to the model not being loaded successfully.

[xbxdghx]

Thanks a lot！So could you plz tell me what is the command you run test.py with HSSDA？

[azhuantou]

Hi, @xbxdghx
You can try the following command to run test.py
python test.py --cfg_file /home/lbh/HSSDA/tools/cfgs/kitti_models/pv_rcnn_ssl.yaml \ --batch_size 8 \ --ckpt /home/lbh/HSSDA/output/kitti_models/pv_rcnn_ssl/pv_rcnn_002_1/ckpt/checkpoint_epoch_80.pth --labeled_frame_idx <path_to_labeled_idx_txt>