Comments (5)
同样地,我在运行demo.ipynb的“Generate Results for the Whole Video”这一部分代码时,也遇到了同样的问题:
Output exceeds the size limit. Open the full output data in a text editor---------------------------------------------------------------------------
RuntimeError Traceback (most recent call last)
Cell In[15], line 52
50 segtracker.add_reference(frame, pred_mask)
51 else:
---> 52 pred_mask = segtracker.track(frame,update_memory=True)
53 torch.cuda.empty_cache()
54 gc.collect()
File /media/alienware/新加卷1/Research/Segment-and-Track-Anything-main/SegTracker.py:75, in SegTracker.track(self, frame, update_memory)
67 def track(self,frame,update_memory=False):
68 '''
69 Track all known objects.
70 Arguments:
(...)
73 merged_mask: numpy array (h,w)
74 '''
---> 75 pred_mask = self.tracker.track(frame)
76 if update_memory:
77 self.tracker.update_memory(pred_mask)
File ~/anaconda3/envs/SAMTrack/lib/python3.9/site-packages/torch/autograd/grad_mode.py:28, in _DecoratorContextManager.call..decorate_context(*args, **kwargs)
25 @functools.wraps(func)
26 def decorate_context(*args, **kwargs):
27 with self.class():
...
579 dim=1).clamp(1e-5, 1 - 1e-5)
580 merged_logit = torch.logit(merged_prob)
582 return merged_logit
RuntimeError: CUDA out of memory. Tried to allocate 1.09 GiB (GPU 0; 7.78 GiB total capacity; 5.02 GiB already allocated; 182.88 MiB free; 5.21 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
from segment-and-track-anything.
Hi, what pytorch version do you use? We recommend pytorch >= 1.10. If yes, please check how many objects are tracked when you use the notebook. If the object number is large, you may try using a smaller points_per_side, e.g. 8 and a larger sam_gap, e.g. 30 for your video to control the object number.
from segment-and-track-anything.
感谢,我试着调整了一下这两个参数,可以多处理一些帧,但是仍旧无法;处理整个视频。我想问一下你们使用的显卡配置,看看能不能向老师申请换个好点的,谢谢!
from segment-and-track-anything.
We use RTX 3090 24G, but according to our observation, 10G may be sufficient to run a video with upto 200 objects. There may be two solutions,
1, downscale your video to a smaller size, e.g. 640x320 and I guess it can be processed on a 8G GPU card. We are considering to add a parameter for users to control the size.
2, switch to a new enviroment with higher pytorch version and try again. The memory management mechanism may be different in different versions.
from segment-and-track-anything.
witch to a new e
非常感谢!
from segment-and-track-anything.
Related Issues (20)
- How to obtain the pixel coordinates of masks in the image?
- DeAOTB_PRE_YTB_DAV在哪里有下载 HOT 1
- 第100張影像後都會變全影像切割 HOT 3
- FPS problem?
- Tuning DeAOT to forget objects quickly? HOT 1
- latest installation notes win HOT 1
- How to train aot? Where is aot's train.py? HOT 1
- Error in demo_instseg.ipynb?
- FileNotFoundError: file "D:\Segment-and-Track-Anything-main\src\groundingdino\groundingdino\config\GroundingDINO_SwinT_OGC.py" does not exist.When i use Click-Tab ,I have this file but there is a error.I would be grateful if you could answer for me. HOT 1
- Potential error in find_new_objs
- OSError: We couldn't connect to 'https://huggingface.co' HOT 1
- 显卡要求什么配置What kind of graphics card can run this experiment? HOT 1
- how many pth files do the project need? only three ?
- How to select new objects/masks in intermediate frames with roll back functionality?
- File "/content/Segment-and-Track-Anything/app.py", line 29, in <module> from ast_master.prepare import ASTpredict ModuleNotFoundError: No module named 'ast_master HOT 3
- 视频跟踪,除了第一帧,masks都是黑的 HOT 3
- UserWarning: Failed to load custom C++ ops. Running on CPU mode Only! HOT 1
- The confidence score is reduce when it have two text prompt
- 会有基于摄像头的实时demo吗
- can not find the package ast_master ?
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 segment-and-track-anything.