valeoai / beef Goto Github PK
View Code? Open in Web Editor NEWLicense: Other
License: Other
Hi, thanks for sharing the code. Is it possible for you to share how to obtain code for I3D and I3D+GCN implementations by [21] you used in your paper?
Hi, we downloaded HDD data from https://usa.honda-ri.com/HDD after submitting the required documents. We are trying to replicate the results of Table1 in your paper.
We could not find train.txt and val.txt in downloaded HDD Data as shared in your readme.
It would be great if you can help us with the files or their sample contents/explaination on how to create them.
Thanks
Rohit
Hi, I am able to successfully run the code for 3 fps and horizon = 2. when I process the videos again for horizon = 5, the processing gets completed nicely but during training I get following error:
^[[1m^[[90m[I 2022-02-24 11:46:26]^[[0m .../criterions/factory.py.16: Creating criterion from bootstrap for eval mode...
^[[1m^[[90m[I 2022-02-24 11:46:26]^[[0m ...els/metrics/factory.py.14: Creating metric from bootstrap for train mode...
^[[1m^[[90m[I 2022-02-24 11:46:34]^[[0m ...els/metrics/factory.py.14: Creating metric from bootstrap for eval mode...
^[[1m^[[90m[I 2022-02-24 11:46:39]^[[0m .../optimizers/factory.py.31: Creating optimizer adam ...
^[[1m^[[90m[I 2022-02-24 11:46:39]^[[0m ...strap/views/factory.py.7: Creating views...
^[[1m^[[90m[I 2022-02-24 11:46:45]^[[0m ...trap/engines/engine.py.113: Launching training procedures
^[[1m^[[90m[I 2022-02-24 11:46:45]^[[0m ...trap/engines/engine.py.151: Training model on trainset for epoch 0
^[[1m^[[90m[I 2022-02-24 11:46:49]^[[0m ...kages/bootstrap/run.py.120: Traceback (most recent call last):
File "/home/rohit.saluja/miniconda3/envs/beef/lib/python3.7/site-packages/bootstrap/run.py", line 113, in main
run(path_opts=path_opts)
File "/home/rohit.saluja/miniconda3/envs/beef/lib/python3.7/site-packages/bootstrap/run.py", line 104, in run
engine.train()
File "/home/rohit.saluja/miniconda3/envs/beef/lib/python3.7/site-packages/bootstrap/engines/engine.py", line 117, in train
self.train_epoch(self.model, self.dataset['train'], self.optimizer, self.epoch)
File "/home/rohit.saluja/miniconda3/envs/beef/lib/python3.7/site-packages/bootstrap/engines/engine.py", line 170, in train_epoch
out = model(batch)
File "/home/rohit.saluja/miniconda3/envs/beef/lib/python3.7/site-packages/torch/nn/modules/module.py", line 541, in __call__
result = self.forward(*input, **kwargs)
File "/home/rohit.saluja/miniconda3/envs/beef/lib/python3.7/site-packages/bootstrap/models/model.py", line 77, in forward
cri_tmp = self.criterions[self.mode](net_out, batch)
File "/home/rohit.saluja/miniconda3/envs/beef/lib/python3.7/site-packages/torch/nn/modules/module.py", line 541, in __call__
result = self.forward(*input, **kwargs)
File "/home/rohit.saluja/BEEF5sec/models/criterions/multitask.py", line 23, in forward
out_traj = self.traj_loss(net_out, batch)
File "/home/rohit.saluja/miniconda3/envs/beef/lib/python3.7/site-packages/torch/nn/modules/module.py", line 541, in __call__
result = self.forward(*input, **kwargs)
File "/home/rohit.saluja/BEEF5sec/models/criterions/l2_points.py", line 19, in forward
diff = pred_xy - gt_xy
RuntimeError: The size of tensor a (13) must match the size of tensor b (31) at non-singleton dimension 1
3
Please suggest what could have gone wrong. Do I need to change anything else in the code which I am missing?
Thanks
Hey, I request your help with something slightly off-topic, i.e. obtaining the video clips used in the BDD-X paper.
The BDDX-Project Page doesn't have the video dataset available, only the explanation text CSV.
Could you possibly help with how to access those videos?
Thanks!
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ 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. ๐๐๐
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 โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.