Comments (8)
Thanks for your attention.
Sorry, we do not deploy our project to the Windows platform.
Our environment is Linux(Ubuntu 16.04/18.04). I think you'd better experiment in Linux platform.
from sstnet.
Thank you for your reply. I also managed to get all the dependencies configured under Ubuntu 20.04, and when I input
CUDA_VISIBLE_DEVICES=0 python train.py --config config/default.yaml
I get the following situation, is this a normal situation? Because I am always in this state for a few hours without any changes. If there is something wrong with it, could you please tell me where the problem is?
(pytorch171) fan@fan-Ubuntu:~/KI_Model/1_Instance_segmetation/SSTNet$ CUDA_VISIBLE_DEVICES=0 python train.py --config config/default.yaml
[10/23 22:27:55 gorilla.config.backup] INFO: gorilla-core version backup.py:39
is 0.2.7.6
[10/23 22:27:55 gorilla.config.backup] INFO: gorilla3d version is backup.py:47
0.1.0
[10/23 22:27:55 gorilla.config.backup] INFO: backup files at backup.py:51
./log/default/backup
[10/23 22:27:55 gorilla.core.env] INFO: set random seed: 123 env.py:161
[10/23 22:27:55 gorilla] INFO: => creating model ... train.py:281
[10/23 22:27:57 gorilla] INFO: #classifier parameters new: train.py:296
113232677
Traceback (most recent call last):
File "/home/fan/anaconda3/envs/pytorch171/lib/python3.8/site-packages/gorilla/
core/registry.py", line 344, in build_from_cfg
return obj_cls(**args)
File
"/home/fan/KI_Model/1_Instance_segmetation/SSTNet/sstnet/data/scannetv2.py",
line 62, in init
self.load_files()
File
"/home/fan/KI_Model/1_Instance_segmetation/SSTNet/sstnet/data/scannetv2.py",
line 66, in load_files
self.files = [torch.load(i) for i in gorilla.track(file_names)]
File
"/home/fan/KI_Model/1_Instance_segmetation/SSTNet/sstnet/data/scannetv2.py",
line 66, in
self.files = [torch.load(i) for i in gorilla.track(file_names)]
File "/home/fan/anaconda3/envs/pytorch171/lib/python3.8/site-packages/torch/se
rialization.py", line 595, in load
return _legacy_load(opened_file, map_location, pickle_module,
**pickle_load_args)
File "/home/fan/anaconda3/envs/pytorch171/lib/python3.8/site-packages/torch/se
rialization.py", line 764, in _legacy_load
magic_number = pickle_module.load(f, **pickle_load_args)
EOFError: Ran out of input
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "train.py", line 309, in
gorilla.launch(
File "/home/fan/anaconda3/envs/pytorch171/lib/python3.8/site-packages/gorilla/
core/launch.py", line 72, in launch
main_func(*args)
File "train.py", line 299, in main
do_train(model, cfg, logger)
File "train.py", line 54, in do_train
train_dataset = gorilla.build_dataset(cfg.dataset)
File "/home/fan/anaconda3/envs/pytorch171/lib/python3.8/site-packages/gorilla/
solver/build.py", line 125, in build_dataset
dataset = build_from_cfg(cfg, DATASETS, default_args)
File "/home/fan/anaconda3/envs/pytorch171/lib/python3.8/site-packages/gorilla/
core/registry.py", line 347, in build_from_cfg
raise type(e)(f'{obj_cls.name}: {e}')
EOFError: ScanNetV2Inst: Ran out of input
Working... ━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7% 0:00:27
from sstnet.
My graphics card is an RTX 3060 laptop 6GB
from sstnet.
Ok, I suggest you format your question in suitable markdown syntax, especially for the terminal outputs/Traceback. Such formating can provide a comfortable highlighting effect and mentioning effect.
After scanning your description, I think that your problem may occur in the dataset? Have you generated the files following the README.md
, which will generate some *.pth
files to load.
And RTX 3060 may not be able to withstand this project. I suggest that the minimum GPU-memory is 12G. 6G memory may not be enough.
Good luck.
from sstnet.
Hello, thank you for your quick reply, I have read the README and have completed the steps in the README. My data file is as follows
Thanks for the reminder that my GPU should probably try the S3DIS dataset. Is there a step-by-step guide like README for processing S3DIS datasets?
from sstnet.
I was got that error after I have read the README and completed the steps in the README, there has *.pth files
from sstnet.
I have not met this problem. I think that you can try to use the ipdb
python-lib(pip install ipdb
) to add the breakpoint(ipdb.set_trace()
) and debug in the SSTNet/sstnet/data/scannetv2.py
's __init__
or load_files
functions.
Tip: if you want to put your console outputs in the issue, please format it in suitable markdown syntax. It makes your issue content seem more comfortable. Thanks.
Good luck.
from sstnet.
I close this issue. If you have any problem with it, please feel free to reopen it.
from sstnet.
Related Issues (14)
- Spconv Installation HOT 1
- KeyError: 'S3DISInst' not in the datasets registry'
- Can the original scenes be removed from the results of the test? HOT 1
- Config file for S3DIS
- Test.py for S3DIS dataset
- SparseConvTensor attribute is missing for spconv at runtime HOT 2
- scene_name.txt_sem_label.npy HOT 4
- Would you show how SSTNet work on the S3DIS? HOT 11
- empty distancetrix bug HOT 2
- about the training time HOT 3
- The number of observations cannot be determined on an empty distance matrix.
- Is this project can run in windows system? Looking forward to your reply. HOT 1
- Inference with own scene data?
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from sstnet.