Comments (4)
GoGoDuck912
commented on July 2, 2024
Please check your settings.
I just re-run the evaluation and list the detailed logs as below.
python evaluate.py --model-restore ./pretrain_model/exp-schp-201908261155-lip.pth
image mean: [0.406, 0.456, 0.485]
image std: [0.225, 0.224, 0.229]
input space:BGR
BGR Transformation
Totoal testing sample numbers: 10000
Pixel accuracy: 88.087233
Mean accuracy: 72.721079
Mean IU: 58.589516
OrderedDict([('Background', 88.3442361878685), ('Hat', 69.96947809134201), ('Hair', 73.54309620912845), ('Glove', 50.30418305554127), ('Sunglasses', 40.64712540171674), ('Upper-clothes', 69.88614578224038│·························
), ('Dress', 39.12394794248705), ('Coat', 57.426223799119505), ('Socks', 54.19401400183561), ('Pants', 75.97924333900869), ('Jumpsuits', 32.68272162400498), ('Scarf', 26.34789636820655), ('Skirt', 31.5923│·························
39627695193), ('Face', 76.19456391929387), ('Left-arm', 68.6327781746178), ('Right-arm', 70.9168761750908), ('Left-leg', 67.22625658157922), ('Right-leg', 66.52695495012838), ('Left-shoe', 55.736763623158│·························
82), ('Right-shoe', 56.51546955975324), ('Pixel accuracy', 88.08723287463441), ('Mean accuracy', 72.72107948118848), ('Mean IU', 58.58951572069085)])
Note this is the plain single scale evaluation result.
from self-correction-human-parsing.
YaoooLiang
commented on July 2, 2024
I have used torch.nn.BatchNorm2d instead of InPlaceABNSync, that's may be the reason. Thank you for quickly reply
from self-correction-human-parsing.
GoGoDuck912
commented on July 2, 2024
Actually, InPlaceABNSync = torch.nn.BatchNorm2d + torch.nn.LeakyRelu
from self-correction-human-parsing.
YaoooLiang
commented on July 2, 2024
Thank you, I will try this
from self-correction-human-parsing.
Related Issues (20)
- LIP dataset not available HOT 3
- How to visualize edge prediction? HOT 1
- Where is the pre-trained model for Multiple Human Parsing?
- Training Time is huge
- CIHP dataset Training
- Testing on CPU HOT 12
- MulitHuman Parsing Resutls
- How to trained on custom dataset？ HOT 2
- LIP dataset - crashed link HOT 1
- CUDA problem on training HOT 1
- generating human-parsing HOT 1
- Hi! I want to convert model to onnx but i have some error with InPlaceABNSync. can you give me some solutions. My error "RuntimeError: ONNX export failed: Couldn't export Python operator InPlaceABNSyn"
- Pascal-Person-Part求助
- Training with others backbone
- Excuse me, why is it stuck here all the time?I need help.
- Real-time test problem
- Cannot reproduce the results from the paper - Problem with InPlaceABNSync?
- 我下载到本地后，在pycharm上运行出的结果和colab上边运行的结果不一样是为啥呀？？都是按照colab的步骤做的呀！ HOT 1
- Which torch and torchvision version are appropriate? HOT 1
- Wrong Inference Results of demo with pretrained HOT 1
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 self-correction-human-parsing.