search

fabro66 / GAST-Net-3DPoseEstimation

A Graph Attention Spatio-temporal Convolutional Networks for 3D Human Pose Estimation in Video (GAST-Net)
MIT License
312 stars 70 forks source link

pretrained model error #23

Closed shenbiren closed 3 years ago

shenbiren commented 3 years ago

Hello!Thanks for your excellent work. I have some problem in testing with the pretrained model I downloaded the 81 frame model and use the command: python trainval.py -k cpn_ft_h36m_dbb -arc 3,3,3,3 -c checkpoint/gastnet --evaluate 81_frame_model.bin -ch 64 however I got a error: Traceback (most recent call last): File "trainval.py", line 55, in model_pos_train, model_pos, checkpoint = load_weight(args, model_pos_train, model_pos) File "/home/leiq1995/25years/GAST-Net-3DPoseEstimation-master/main.py", line 237, in load_weight model_pos_train.load_state_dict(new_state_dict) File "/root/anaconda3/envs/gast/lib/python3.6/site-packages/torch/nn/modules/module.py", line 1045, in load_state_dict self.class.name, "\n\t".join(error_msgs))) RuntimeError: Error(s) in loading state_dict for SpatioTemporalModelOptimized1f: Missing key(s) in state_dict: "init_bn.weight", "init_bn.bias", "init_bn.running_mean", "init_bn.running_var", "expand_bn.weight", "expand_bn.bias", "expand_bn.running_mean", "expand_bn.running_var", "shrink.weight", "expand_conv.weight", "layers_conv.0.weight", "layers_conv.1.weight", "layers_conv.2.weight", "layers_conv.3.weight",......

The error information is very long.

When I run the train command, there is no error.

fabro66 commented 3 years ago

Hi ~ Thank you for your interest in our work. I have fixed the bug about evaluation on CPU. Please replace trainval.py and main.py files to evaluate the pretrained model. best wishes!