Comments (8)
Hi, have you solved this issue? I encountered this problem when training on Sceneflow.
from crestereo.
Hi, have you solved this issue? I encountered this problem when training on Sceneflow.
yes,I solved it by https://github.com/ibaiGorordo/CREStereo-Pytorch
and,there is not much difference in their effectiveness after 500 epoch
from crestereo.
Hi, have you solved this issue? I encountered this problem when training on Sceneflow.
besides,I think it's the lack of memory that causes it NAN.You can try to solve this problem
from crestereo.
Hi, have you solved this issue? I encountered this problem when training on Sceneflow.
yes,I solved it by https://github.com/ibaiGorordo/CREStereo-Pytorch and,there is not much difference in their effectiveness after 500 epoch
Thank you for your reply! I will try in this way.
from crestereo.
Hi, have you solved this issue? I encountered this problem when training on Sceneflow.
besides,I think it's the lack of memory that causes it NAN.You can try to solve this problem
Were you able to reproduce their performance with the pytorch implementation? I tried that repo you mentioned, but I'm still suffering from Nan loss after some epochs. If you were able to reproduce, what datasets did you use? Please specify the sub-datasets such as "monkaa", and "clean" or "final" versions you used. Thanks!
from crestereo.
@deephog
I used https://github.com/ibaiGorordo/CREStereo-Pytorch and solved NAN. I use the datasets of Baidu Web disk provided by the author (Download from BaiduCloud here(Extraction code: aa3g) and extract the tar files manually).This is the result after 200 epochs
from crestereo.
@deephog I used https://github.com/ibaiGorordo/CREStereo-Pytorch and solved NAN. I use the datasets of Baidu Web disk provided by the author (Download from BaiduCloud here(Extraction code: aa3g) and extract the tar files manually).This is the result after 200 epochs
Did you compare the final results to the pre-trained model they provide? I can get similar results, but I can never get as good as theirs
from crestereo.
@deephog I used https://github.com/ibaiGorordo/CREStereo-Pytorch and solved NAN. I use the datasets of Baidu Web disk provided by the author (Download from BaiduCloud here(Extraction code: aa3g) and extract the tar files manually).This is the result after 200 epochs
Did you compare the final results to the pre-trained model they provide? I can get similar results, but I can never get as good as theirs
I'm sorry,This is what I did a few months ago,I only remember that after 500 epochs the results were more or less adequate for my needs.However, in general, the author's pre-trained model is the best, and it is normal that you cannot achieve the author's results.
from crestereo.
Related Issues (20)
- the GPU memory is too large HOT 1
- WRN Not FormattedTensorValue input for AttachGrad op: AttachGradValue{key=grad_1}
- Datasets in training and schedule HOT 1
- Predicting disparity for the right image
- Is there is a way to speed up model training, like using the Torch DDP method?
- Running CREStereo with the latest version of MegEngine
- Assistance Requested: Issues Encountered with train.py Script in CREStereo Repository
- The stereo algorithm is implemented using TensorRT
- Is there any email in this paper?
- CREStereo Dataset
- A Problem About Code
- corr.py syntax error
- The issue of disparity to depth conversion HOT 1
- TypeError: load() got an unexpected keyword argument 'map_location'
- 想问一下你的数据集的视差图是从blender里面导出的,有推荐的学习教程吗
- Crestereo direct model conversion to onnx using MgeConvert
- Stereo Algorithms (Include:CREStereo,RAFT-Stereo,Hitnet,FastACVNet_plus,Stereo Transformers,RealtimeStereo,DistDepth) with TensorRT,ORT,OpenVINO
- Stereo Algorithms (Include:CREStereo,RAFT-Stereo,Hitnet,FastACVNet_plus,Stereo Transformers,RealtimeStereo,DistDepth) with TensorRT,ORT,OpenVINO
- confidence map for disparity map
- TypeError: __init__() got an unexpected keyword argument 'divide'
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 crestereo.