some errors when train,Can someone help me?please ：)

[my-hello-world]

(becasue i get the error #20 ,and i cannot modified it ) , i changed CONV_BODY: FPN3D.add_fpn_ResNet101_conv5_body to CONV_BODY: FPN.add_fpn_ResNet101_conv5_body
and NUM_GPUS: 2
when I run this:
python launch.py -c configs/video/2d_best/my_01_R101_best_hungarian-2GPU.yaml

i have same errors like #35

WARNING workspace.py: 192: File "/home/vivian/HelloWorld/tracker/DetectAndTrack/lib/modeling/model_builder.py", line 193, in _single_gpu_build_func Traceback (most recent call last): File "tools/train_net.py", line 258, in <module> checkpoints = net_trainer() File "tools/train_net.py", line 175, in net_trainer workspace.RunNet(model.net.Proto().name) File "/home/vivian/HelloWorld/tracker/tools/pytorch/build/caffe2/python/workspace.py", line 219, in RunNet StringifyNetName(name), num_iter, allow_fail, File "/home/vivian/HelloWorld/tracker/tools/pytorch/build/caffe2/python/workspace.py", line 180, in CallWithExceptionIntercept return func(*args, **kwargs) RuntimeError: [enforce fail at pybind_state.cc:1141] success. Error running net keypoint_rcnn frame #0: c10::ThrowEnforceNotMet(char const*, int, char const*, std::string const&, void const*) + 0x59 (0x7f83f96f8a89 in /home/vivian/HelloWorld/tracker/tools/pytorch/build/lib/libc10.so) frame #1: <unknown function> + 0x492af (0x7f83f994f2af in /home/vivian/HelloWorld/tracker/tools/pytorch/build/caffe2/python/caffe2_pybind11_state_gpu.so) frame #2: <unknown function> + 0x949a8 (0x7f83f999a9a8 in /home/vivian/HelloWorld/tracker/tools/pytorch/build/caffe2/python/caffe2_pybind11_state_gpu.so) <omitting python frames> frame #16: __libc_start_main + 0xf5 (0x7f8405347f45 in /lib/x86_64-linux-gnu/libc.so.6)

my json_datasets.py is

IM_DIR = 'image_directory' ANN_FN = 'annotation_file' SPLIT_INTO_FRAMES = Flase FRAMES_FROM_VIDEO = Flase COMPUTED_ANNOTATIONS_INFO = 'computed_annotations_info' ANN_DN = 'annotation_directory' DATASETS = { 'posetrack_v1.0_train': { IM_DIR: 'lib/datasets/data/PoseTrack/images/', ANN_FN: 'lib/datasets/lists/PoseTrack/v1.0/posetrack_train.json', #ANN_DN: 'lib/datasets/data/PoseTrackV1.0_Annots_train_json/', }, 'posetrack_v1.0_val': { IM_DIR: 'lib/datasets/data/PoseTrack/images/', ANN_FN: 'lib/datasets/lists/PoseTrack/v1.0/posetrack_val.json', ANN_DN: 'lib/datasets/data/PoseTrackV1.0_Annots_val_json', }, 'posetrack_v1.0_test': { IM_DIR: 'lib/datasets/data/PoseTrack/images/', ANN_FN: 'lib/datasets/lists/PoseTrack/v1.0/posetrack_test.json', # ANN_DN: 'lib/datasets/data/PoseTrackV1.0_Annots_test_json', }, } ,
any incorrect configs? and i downloaded the files from here, and extract them as lib/datasets/data/PoseTrackV1.0_Annots_val_json.
i think this folder is only required for evaluation. It can have any arbitrary value while training. if not，where can i get the PoseTrackV1.0_Annots_train_json and PoseTrackV1.0_Annots_test_json.

thanks！
：)