bbidong
bbidong
When I run python train/train.py , it shows me as the follow , I guess there is something wrong , but the process next don't appear error , I don't...
just in terms of segmentation , what's the diffrence KittiSeg from the FCN ? Are they same ??
The pose weight --ckpt_file models/model-100280 is trained on KITTI odometry , could I directly use it to predict the pose on the other dataset such as KITTI tracking data???
Have the train code for pytorch released ? i have not find it .....
Thank you for your efforts. And I note that `datasets/vid_dataset.py` is used in `evaluate.py`, `datasets/vid_dataset_new.py` is used in `datasets/vid_dataset_new.py`. So what's the difference between `datasets/vid_dataset.py` and `datasets/vid_dataset_new.py` ?
When i specify the GPUs in `confing.py` as following:  And the memory occupation of gpu 2,3 is shown as following  Why gpu3 is not used ? What should...
how should I train my own weight rather than download from the link??