mmurray / cvdn Goto Github PK
View Code? Open in Web Editor NEWCooperative Vision-and-Dialog Navigation
License: Other
Cooperative Vision-and-Dialog Navigation
License: Other
Sorry to bother but there is a little problem while we try this project.
The Matterport_Data is too big to download.
Did the project uses the whole Matterport_Data,or just part of it.
Can i just download the specific Matterport_Data part to run the code?
Looking forward to your reply.
Hi,
I have a question about the report of the experiment results. In your 'summarize_perf.py' file, you use the max function to choose the performance. Does it mean that different metrics ('success_rate', 'dist_to_end_reduction', 'oracle path_success_rate' and so on) are reported by the optimal results, regardless of whether they come from the same iteration? Is this the way you report the performance in the paper?
Thank you for your attention :)
Hello, in train.json downloaded by tasks/CVDN/data/download.sh, the navigator's ' elevation' and 'heading' values feel strange.
The images we extracted from train.json do not correspond to the dialogue.
As shown in the image above, the excessive 'elevation' value causes the robot to look upwards all the time, which does not help the robot to navigate.
We don't know if 'elevation' is handled in the code, and we look forward to your help. Thank you!
When I try to execute the train.py file with given example parametres, I get the following error:
/usr/local/lib/python2.7/dist-packages/torch/nn/modules/rnn.py:46: UserWarning: dropout option adds dropout after all but last recurrent layer, so non-zero dropout expects num_layers greater than 1, but got dropout=0.5 and num_layers=1
"num_layers={}".format(dropout, num_layers))
Traceback (most recent call last):
File "tasks/NDH/train.py", line 253, in
train_val(path_type, max_episode_len, history, MAX_INPUT_LENGTH, feedback_method, n_iters, model_prefix, blind)
File "tasks/NDH/train.py", line 191, in train_val
dropout_ratio, bidirectional=bidirectional).cuda()
File "/usr/local/lib/python2.7/dist-packages/torch/nn/modules/module.py", line 260, in cuda
return self._apply(lambda t: t.cuda(device))
File "/usr/local/lib/python2.7/dist-packages/torch/nn/modules/module.py", line 187, in _apply
module._apply(fn)
File "/usr/local/lib/python2.7/dist-packages/torch/nn/modules/rnn.py", line 117, in _apply
self.flatten_parameters()
File "/usr/local/lib/python2.7/dist-packages/torch/nn/modules/rnn.py", line 113, in flatten_parameters
self.batch_first, bool(self.bidirectional))
RuntimeError: cuDNN error: CUDNN_STATUS_EXECUTION_FAILED
I would be grateful if someone could tell me a solution to this.
Can't load 'all.json' file when run './docker/run_server.sh'. Did you miss this file in the image of the docker?
Thx!
Firstly, thank you for your project!
How to visualize the Interactive Demo?
Is the following instructions suitable for CVDN?
To run an interactive demo, build the docker image as described above (docker build -t mattersim .
), then run the docker container while sharing the host's X server and DISPLAY environment variable with the container:
xhost +
nvidia-docker run -it -e DISPLAY -v /tmp/.X11-unix:/tmp/.X11-unix --mount type=bind,source=$MATTERPORT_DATA_DIR,target=/root/mount/Matterport3DSimulator/data/v1/scans,readonly --volume `pwd`:/root/mount/Matterport3DSimulator mattersim
cd /root/mount/Matterport3DSimulator
If you get an error like Error: BadShmSeg (invalid shared segment parameter) 128
you may also need to include -e="QT_X11_NO_MITSHM=1"
in the docker run command above.
Build the simulator using any rendering option. Commands for running both python and C++ demos are provided below. These are very simple demos designed to illustrate the use of the simulator in python and C++. By default, these demos have depth rendering off. Check the code and turn it on if you have preprocessed the depth outputs and want to see depth as well (see Depth Outputs above).
Python demo:
python3 src/driver/driver.py
C++ demo:
build/mattersim_main
Thank you
hi,sorry to bother. I notice that the Ubuntu vesrsion required is Ubuntu16.04, could we use some new version such us Ubuntu20.04? thanks~
Hi,
Firstly, thank you for your project!
When I run python task/NDH/eval.py
, an error occurs:
I've already read https://evalai.cloudcv.org/web/challenges/challenge-page/463/submission. There is no test
split provided by https://github.com/mmurray/cvdn/blob/master/tasks/NDH/data/download.sh but test_cleaned
split. Could you tell me the details about how to evaluate and submit the result of test
set?
Thank you
Hi, I noticed that the cvdn.dev website is currently down. Since all the dataset download scripts are getting data from that website, none of them work now. I wonder if the authors would be willing to restore the website or provide an alternative way of downloading the data. Thanks a lot!
The upstream Matterport3DSim codebase has moved on to Python 3 (and updated some other dependencies). We should pull those changes down, but it will require combing through the CVDN/NDH python files and updating old Python 2 syntax.
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.