Closed aiai84 closed 3 years ago
/home/.conda/envs/pvn3d/lib/python3.6/site-packages/torch/nn/parallel/_functions.py:61: UserWarning: Was asked to gather along dimension 0, but all input tensors were scalars; will instead unsqueeze and return a vector.
warnings.warn('Was asked to gather along dimension 0, but all '
obj_id: 1 0.1 diameter: 0.010209865663
1 ape
add auc: 0
adds auc: 0
add(-s) auc: 0
add < 0.1 diameter: 0.0
***adds < 0.1 diameter: 0.0
=== Training Progress ===
acc_rgbd --- val: 0.9926
loss --- val: 954.2041
loss_rgbd_seg --- val: 1.3864
loss_kp_of --- val: 912.4642
loss_ctr_of --- val: 40.3535
loss_target --- val: 954.2041
For testing, we set the batch size to 1 link. I think the error is not about batch size.
Is the pre-trained model loaded successfully? Check that you put the checkpoint to the corresponding folder and set the tst_mdl
path accordingly.
cls='ape'
tst_mdl=train_log/linemod/checkpoints/${cls}/${cls}_pvn3d_best.pth.tar
python3 -m train.train_linemod_pvn3d -checkpoint $tst_mdl -eval_net --test --cls $cls
Check the log whether the pre-trained model is successfully loaded when you run the scripts. If it's done successfully, it will print something like " loading from xxx. Done."
Thank you very much, I tested it once again and it worked!
Hello, thank you for the nice work of you. For the lackness of enough GPU memory, I set the batchsize to 10, and evaluate the trained model's add and adds, they are all zero, who can tell me what is the reason. I changed the batchsize only. Thank you very much!