limaosen0 / dmgnn Goto Github PK
View Code? Open in Web Editor NEWThe implementation of DMGNN
License: MIT License
The implementation of DMGNN
License: MIT License
Can you tell me how to process the skeleton sequence coordinates of 15 key points in several frames and then train the network for prediction?
Can you help me how can I call REC_Processor class in jupyter notebook? I got this error:
usage: ipykernel_launcher.py [-w WORK_DIR] [-c CONFIG] [--phase PHASE]
[--save_result SAVE_RESULT] [--iter_num ITER_NUM]
[--use_gpu USE_GPU]
[--device DEVICE [DEVICE ...]]
[--log_interval LOG_INTERVAL]
[--save_interval SAVE_INTERVAL]
[--eval_interval EVAL_INTERVAL]
[--savemotion_interval SAVEMOTION_INTERVAL]
[--save_log SAVE_LOG] [--print_log PRINT_LOG]
[--pavi_log PAVI_LOG] [--actions ACTIONS]
[--data_dir DATA_DIR] [--subtrain SUBTRAIN]
[--subtest SUBTEST] [--batch_size BATCH_SIZE]
[--source_seq_len SOURCE_SEQ_LEN]
[--target_seq_len TARGET_SEQ_LEN] [--model MODEL]
[--model_args MODEL_ARGS]
[--edge_weighting EDGE_WEIGHTING]
[--weights WEIGHTS]
[--ignore_weights IGNORE_WEIGHTS [IGNORE_WEIGHTS ...]]
[--base_lr BASE_LR] [--step STEP [STEP ...]]
[--optimizer OPTIMIZER] [--nesterov NESTEROV]
[--weight_decay WEIGHT_DECAY] [--lamda LAMDA]
[--fusion_layer_dir FUSION_LAYER_DIR]
[--learning_rate_dir LEARNING_RATE_DIR]
[--lamda_dir LAMDA_DIR] [--crossw_dir CROSSW_DIR]
[--note NOTE] [--debug DEBUG]
ipykernel_launcher.py: error: unrecognized arguments: -f /home/elham/.local/share/jupyter/runtime/kernel-f4a67c30-c0fb-4808-808d-8763f16ab43e.json
An exception has occurred, use %tb to see the full traceback.
SystemExit: 2
c12_1 = self.j2p_1(x_s1_1, x_s2_1, relrec_s1, relsend_s1, relrec_s2, relsend_s2)
r12_1 = self.p2j_1(x_s2_1, x_s1_1, relrec_s2, relsend_s2, relrec_s1, relsend_s1)
c23_1 = self.p2b_1(x_s2_1, x_s3_1, relrec_s2, relsend_s2, relrec_s3, relsend_s3)
r23_1 = self.b2p_1(x_s3_1, x_s2_1, relrec_s3, relsend_s3, relrec_s2, relsend_s2)
传入的relrec_s1, relsend_s1, relrec_s2, relsend_s2,relrec_s3, relsend_s3分别表示什么啊?
Hello, are there some more instructions for the code implementation.
您好,在您的paper中,讲到human3.6有32个关节点,但是您代码中的数据并不是表示关节点的三维坐标,而是表示旋转向量;其次,您的代码中,您训练后的模型并没有完整的提供99个数据,训练模型仅仅输出了30个数据,请问您是如何可视化最后的预测结果的?得到您paper中的图的呢?
Thanks for your perfect work. Could you please provide more setting details (e.g., how many batches you used) about the average time cost comparison.
Thanks again.
Hello:
I want to run the code.First, I prepare environment according to " cd torchlight, python setup.py install, cd .." while I want to train model, the error "cannot deom torchlight import import_class" occurred. How to solve it?
The files are missing in:
Long term
Can I place files from Short term in the long term folder?
What are other changes ?
I have this problem when I run it processors['prediction'] = import_class('processor.recognition.REC_Processor')
and ImportError: cannot import name 'Processor' from 'processor'
, how to solve it?
Traceback (most recent call last):
File "main.py", line 23, in
p.start()
File "/media/ubuntu/data2/DMGNN-master/h36m-short/processor/processor.py", line 126, in start
self.test()
File "/media/ubuntu/data2/DMGNN-master/h36m-short/processor/recognition.py", line 252, in test
self.mean_MAE_select[iter_time] = np.mean((self.MAE_tensor[iter_time, :, 1]+
AttributeError: 'REC_Processor' object has no attribute 'mean_MAE_select'
Could you please provide a visualization code which is missing in the repository?
I tried to convert the expmap output to euler and then to a skeleton by using the code given in data_tools.py.
I'm confused on hyper-parameter selection. It seems author select hyper-parameter on test dataset and I also wonder which checkpoint should be chosen in the test phase
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.