Comments (6)
It worked after I modified the Makefile.
Makefile:
pre:
python -m pip install -r requirements.txt -f https://download.pytorch.org/whl/torch_stable.html
mkdir -p thirdparty
git clone https://github.com/open-mmlab/mmdetection.git thirdparty/mmdetection
cd thirdparty/mmdetection && git checkout v2.28.1 #switch to old version
cd thirdparty/mmdetection && python -m pip install -e .
install:
make pre
python -m pip install -e .
clean:
rm -rf thirdparty
rm -r ssod.egg-info
Aditionally, I installed mmcv-full==1.3.17 (pip install mmcv-full==1.3.17 -f https://download.openmmlab.com/mmcv/dist/cu111/torch1.9.0/index.html)
from consistentteacher.
The installed mmcv & mmdet version may not match the requirements listed here.
from consistentteacher.
@Johnson-Wang, Thank you.
The installed mmcv & mmdet version may not match the requirements listed here.
I followed the listed versions. However, mmdet 2.25.0 is not compatible with mmcv1.3.9 here. So, I used within mmcv-full>=1.3.17, <1.6.0 range. All did not work out for me. It generates the same error. I also tried the second option:mmdetection=2.28.1
mmcv=1.7.1, but the same scenario.
Can it be the dataset structure? I followed the configs file data structure. data/coco/annotations/annotation_files and data/coco_semi/semi_supervised/*semi_supervised_files
Additionally, what is the need for cloning "git clone https://github.com/open-mmlab/mmdetection.git"? The make install command download and install mmdetction.
from consistentteacher.
Dear @Simeon340703 ,
The reason for git clone https://github.com/open-mmlab/mmdetection.git
is to reuse some configs from mmdet repo such as dataset definition and training scheduler. I would suggest you
git clone https://github.com/open-mmlab/mmdetection.git
cd mmdetection
git checkout 2.x
which leads to the old version config for model 2.x.
Best
from consistentteacher.
@Johnson-Wang Thank you,
I did not try that way. I used git checkout in the Makefile file (cd thirdparty/mmdetection && git checkout v2.x. ). I will try this way and let you know.
Thanks again.
from consistentteacher.
It worked after I modified the Makefile. Makefile: pre: python -m pip install -r requirements.txt -f https://download.pytorch.org/whl/torch_stable.html mkdir -p thirdparty git clone https://github.com/open-mmlab/mmdetection.git thirdparty/mmdetection cd thirdparty/mmdetection && git checkout v2.28.1 #switch to old version cd thirdparty/mmdetection && python -m pip install -e . install: make pre python -m pip install -e . clean: rm -rf thirdparty rm -r ssod.egg-info Aditionally, I installed mmcv-full==1.3.17 (pip install mmcv-full==1.3.17 -f https://download.openmmlab.com/mmcv/dist/cu111/torch1.9.0/index.html)
Hello,
I encountered the same problem as you, and according to the solution you provided, the following problem occurred: AssertionError: MMCV=1.3.17 is used but incompatible Please install mmcv>=2.0.0rc4,<2.2.0
I am not sure what is the cause of the problem.
Thanks,
from consistentteacher.
Related Issues (20)
- Heatmap of predicted bboxes confidence and its IoU score with GTs HOT 5
- Question about the missing ASA in consistent_teacher.py.
- Q&A and clarification HOT 2
- how i can use this in dota datasets HOT 1
- if the batch is bigger, the training time is longer? HOT 1
- Environmental configuration HOT 4
- ok HOT 18
- Multi card parallel HOT 17
- thanks HOT 2
- This code is the worst HOT 3
- How to match vision of mmdet with mmcv or mmcv-full?
- 关于GMM模块的疑问
- FCOS backbone
- Pretraining weights HOT 4
- training method HOT 12
- 8 * v100 parallel training HOT 3
- TypeError: ConsistentTeacher: __init__() got an unexpected keyword argument 'pretrained'
- AttributeError: ConsistentTeacher: 'NoneType' object has no attribute 'num_scores' HOT 2
- AssertionError: train_cfg specified in both outer field and model field HOT 2
- RuntimeError: DeformConv is not compiled with GPU support
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 consistentteacher.