search

Chiaraplizz / ST-TR

Spatial Temporal Transformer Network for Skeleton-Based Activity Recognition
MIT License
294 stars 57 forks source link

Can not find these weights #13

Open Hiawons opened 3 years ago

Hiawons commented 3 years ago

when I run without bones on NTU-RGB-D, this problem has occurred as the follow picture. Looking forward to your early reply.

image

Chiaraplizz commented 3 years ago

when I run without bones on NTU-RGB-D, this problem has occurred as the follow picture. Looking forward to your early reply.

image

Hi! Can you send me your train.yml? It should be some error in that.

Chiara

wangpitao commented 3 years ago

I also meet this problem when the last epoch:

Validation: Epoch [119/120], Samples [14206.0/16487], Loss: 0.305077463388443, Validation Accuracy: 86.16485716018681 [ Fri Apr 9 03:30:50 2021 ] Load weights from ./prova20/epoch119_model.pt. Can not find these weights: module.gcn0.bn.weight module.backbone.0.tcn1.bn.num_batches_tracked module.gcn0.conv_list.1.bias module.backbone.1.tcn1.conv.bias module.tcn0.bn.running_mean module.backbone.6.down1.conv.weight module.backbone.5.gcn1.bn.weight module.gcn0.bn.bias module.backbone.0.gcn1.bn.running_mean module.backbone.0.gcn1.conv_list.0.weight module.backbone.3.gcn1.bn.running_mean ......

Hiawons commented 3 years ago

This is my train.yaml, can you find anything wrong?

feeder

feeder: st_gcn.feeder.Feeder feeder_augmented: st_gcn.feeder.FeederAugmented train_feeder_args: data_path: /home/lenovo/hxx/actionDetection/data/NTU-RGB-D-twoperson/xview/train_data.npy label_path: /home/lenovo/hxx/actionDetection/data/NTU-RGB-D-twoperson/xview/train_label.pkl

data_path: ./Output_skeletons_without_missing_skeletons/xsub/train_data_joint_bones.npy

label_path: ./Output_skeletons_without_missing_skeletons/xsub/train_label_filtered.pkl

random_choose: False random_shift: False random_move: False window_size: -1 normalization: False mirroring: False

test_feeder_args: data_path: /home/lenovo/hxx/actionDetection/data/NTU-RGB-D-twoperson/xview/val_data.npy label_path: /home/lenovo/hxx/actionDetection/data/NTU-RGB-D-twoperson/xview/val_label.pkl

data_path: ./Output_skeletons_without_missing_skeletons/xsub/val_data_joint_bones.npy

label_path: ./Output_skeletons_without_missing_skeletons/xsub/val_label_filtered.pkl

model

model: st_gcn.net.ST_GCN training: True

model_args: num_class: 26 channel: 3 window_size: 300 num_point: 25 num_person: 2 mask_learning: True use_data_bn: True attention: True only_attention: True tcn_attention: False data_normalization: True skip_conn: True weight_matrix: 2 only_temporal_attention: True bn_flag: True attention_3: False kernel_temporal: 9 more_channels: False double_channel: True drop_connect: True concat_original: True all_layers: False adjacency: False agcn: False dv: 0.25 dk: 0.25 Nh: 8 n: 4 dim_block1: 10 dim_block2: 30 dim_block3: 75 relative: False graph: st_gcn.graph.NTU_RGB_D visualization: False graph_args: labeling_mode: 'spatial'

optical_flow: True

optim

0: old one, 1: new one

scheduler: 1 weight_decay: 0.0001 base_lr: 0.1 step: [60,90]

training

device: [0,1] batch_size: 2 test_batch_size: 8 num_epoch: 120 nesterov: True

zhangwuji1998 commented 3 years ago

I have meet the same question

zhangwuji1998 commented 3 years ago

This is my train.yaml, can you find anything wrong?

feeder

feeder: st_gcn.feeder.Feeder feeder_augmented: st_gcn.feeder.FeederAugmented train_feeder_args: data_path: /home/lenovo/hxx/actionDetection/data/NTU-RGB-D-twoperson/xview/train_data.npy label_path: /home/lenovo/hxx/actionDetection/data/NTU-RGB-D-twoperson/xview/train_label.pkl

data_path: ./Output_skeletons_without_missing_skeletons/xsub/train_data_joint_bones.npy

label_path: ./Output_skeletons_without_missing_skeletons/xsub/train_label_filtered.pkl

random_choose: False random_shift: False random_move: False window_size: -1 normalization: False mirroring: False

test_feeder_args: data_path: /home/lenovo/hxx/actionDetection/data/NTU-RGB-D-twoperson/xview/val_data.npy label_path: /home/lenovo/hxx/actionDetection/data/NTU-RGB-D-twoperson/xview/val_label.pkl

data_path: ./Output_skeletons_without_missing_skeletons/xsub/val_data_joint_bones.npy

label_path: ./Output_skeletons_without_missing_skeletons/xsub/val_label_filtered.pkl

model

model: st_gcn.net.ST_GCN training: True

model_args: num_class: 26 channel: 3 window_size: 300 num_point: 25 num_person: 2 mask_learning: True use_data_bn: True attention: True only_attention: True tcn_attention: False data_normalization: True skip_conn: True weight_matrix: 2 only_temporal_attention: True bn_flag: True attention_3: False kernel_temporal: 9 more_channels: False double_channel: True drop_connect: True concat_original: True all_layers: False adjacency: False agcn: False dv: 0.25 dk: 0.25 Nh: 8 n: 4 dim_block1: 10 dim_block2: 30 dim_block3: 75 relative: False graph: st_gcn.graph.NTU_RGB_D visualization: False graph_args: labeling_mode: 'spatial'

optical_flow: True

optim

0: old one, 1: new one

scheduler: 1 weight_decay: 0.0001 base_lr: 0.1 step: [60,90]

training

device: [0,1] batch_size: 2 test_batch_size: 8 num_epoch: 120 nesterov: True

have you save the problem？

Hiawons commented 3 years ago

This is my train.yaml, can you find anything wrong?

feeder

feeder: st_gcn.feeder.Feeder feeder_augmented: st_gcn.feeder.FeederAugmented train_feeder_args: data_path: /home/lenovo/hxx/actionDetection/data/NTU-RGB-D-twoperson/xview/train_data.npy label_path: /home/lenovo/hxx/actionDetection/data/NTU-RGB-D-twoperson/xview/train_label.pkl

data_path: ./Output_skeletons_without_missing_skeletons/xsub/train_data_joint_bones.npy

label_path: ./Output_skeletons_without_missing_skeletons/xsub/train_label_filtered.pkl

random_choose: False random_shift: False random_move: False window_size: -1 normalization: False mirroring: False test_feeder_args: data_path: /home/lenovo/hxx/actionDetection/data/NTU-RGB-D-twoperson/xview/val_data.npy label_path: /home/lenovo/hxx/actionDetection/data/NTU-RGB-D-twoperson/xview/val_label.pkl

data_path: ./Output_skeletons_without_missing_skeletons/xsub/val_data_joint_bones.npy

label_path: ./Output_skeletons_without_missing_skeletons/xsub/val_label_filtered.pkl

model

model: st_gcn.net.ST_GCN training: True model_args: num_class: 26 channel: 3 window_size: 300 num_point: 25 num_person: 2 mask_learning: True use_data_bn: True attention: True only_attention: True tcn_attention: False data_normalization: True skip_conn: True weight_matrix: 2 only_temporal_attention: True bn_flag: True attention_3: False kernel_temporal: 9 more_channels: False double_channel: True drop_connect: True concat_original: True all_layers: False adjacency: False agcn: False dv: 0.25 dk: 0.25 Nh: 8 n: 4 dim_block1: 10 dim_block2: 30 dim_block3: 75 relative: False graph: st_gcn.graph.NTU_RGB_D visualization: False graph_args: labeling_mode: 'spatial'

optical_flow: True

optim

0: old one, 1: new one

scheduler: 1 weight_decay: 0.0001 base_lr: 0.1 step: [60,90]

training

device: [0,1] batch_size: 2 test_batch_size: 8 num_epoch: 120 nesterov: True

have you save the problem？

Hiawons commented 3 years ago

This is my train.yaml, can you find anything wrong?

feeder

feeder: st_gcn.feeder.Feeder feeder_augmented: st_gcn.feeder.FeederAugmented train_feeder_args: data_path: /home/lenovo/hxx/actionDetection/data/NTU-RGB-D-twoperson/xview/train_data.npy label_path: /home/lenovo/hxx/actionDetection/data/NTU-RGB-D-twoperson/xview/train_label.pkl

data_path: ./Output_skeletons_without_missing_skeletons/xsub/train_data_joint_bones.npy

label_path: ./Output_skeletons_without_missing_skeletons/xsub/train_label_filtered.pkl

random_choose: False random_shift: False random_move: False window_size: -1 normalization: False mirroring: False test_feeder_args: data_path: /home/lenovo/hxx/actionDetection/data/NTU-RGB-D-twoperson/xview/val_data.npy label_path: /home/lenovo/hxx/actionDetection/data/NTU-RGB-D-twoperson/xview/val_label.pkl

data_path: ./Output_skeletons_without_missing_skeletons/xsub/val_data_joint_bones.npy

label_path: ./Output_skeletons_without_missing_skeletons/xsub/val_label_filtered.pkl

model

model: st_gcn.net.ST_GCN training: True model_args: num_class: 26 channel: 3 window_size: 300 num_point: 25 num_person: 2 mask_learning: True use_data_bn: True attention: True only_attention: True tcn_attention: False data_normalization: True skip_conn: True weight_matrix: 2 only_temporal_attention: True bn_flag: True attention_3: False kernel_temporal: 9 more_channels: False double_channel: True drop_connect: True concat_original: True all_layers: False adjacency: False agcn: False dv: 0.25 dk: 0.25 Nh: 8 n: 4 dim_block1: 10 dim_block2: 30 dim_block3: 75 relative: False graph: st_gcn.graph.NTU_RGB_D visualization: False graph_args: labeling_mode: 'spatial'

optical_flow: True

optim

0: old one, 1: new one

scheduler: 1 weight_decay: 0.0001 base_lr: 0.1 step: [60,90]

training

device: [0,1] batch_size: 2 test_batch_size: 8 num_epoch: 120 nesterov: True

have you save the problem？

Chiaraplizz commented 3 years ago

Hi 😊 Which model are you loading? I noticed that you set double_channels: True. Please notice that models using joint information only have been trained with double_channels: False.

Chiara

kartheekkotha commented 7 months ago

Hi Chiaraplizz, I ran into the same issue, even after changing the double_channels to True, the error persists. Looking forward to your early reply. Thank you.

lailiang2333 commented 4 months ago

Hi Chiaraplizz, I ran into the same issue, even after changing the double_channels to True, the error persists. Looking forward to your early reply. Thank you.

Hi,have you solved the problem?