Comments (5)
In my case,
self.data_dir
└── Nukki
└── baidu_V1
├── input
├── target
├── train.txt
└── val.txt
└── baidu_V2
├── input
├── target
├── train.txt
└── val.txt
└── Portrait
this_additional is " '/Nukki/baidu_V1/'"
And code call additional dataset by
img_file = ((self.data_dir).strip() + this_additoinal+'input/' + line.strip())
label_file = ((self.data_dir).strip() + this_additoinal+'target/' + line.strip())
Finally
self.data_dir/Nukki/baidu_V1/input/ and self.data_dir/Nukki/baidu_V1/target/
The train.txt is also
self.data_dir/Nukki/baidu_V1/train.txt
Could you tell me your definition of additional_data variable???
from ext_portrait_segmentation.
I only want to train baidu_V1 & baidu_V2, so from loadData.py I comment-out from line 57 to line 112.
see loadData.py.zip
SINet.json
"data_dir": "./dataset/",
Dataset folder tree
home@home-desktop:~/p10/ext_portrait_segmentation/dataset$ tree -L 3
.
├── Nukki
│ ├── baidu_V1
│ │ ├── input
│ │ ├── target
│ │ ├── train.txt
│ │ └── val.txt
│ └── baidu_V2
│ ├── input
│ ├── target
│ ├── train.txt
│ └── val.txt
└── Portrait
dataloader.py
additional_data = []
Now I am getting error:
see full log
log.txt
Traceback (most recent call last):
File "main.py", line 251, in <module>
lossVal, ElossVal, mIOU_val = val_edge(num_gpu, valLoader, model, criteria, Lovasz)
File "/home/home/p10/ext_portrait_segmentation/etc/help_function.py", line 25, in val_edge
rand_pick = np.random.randint(0, len(val_loader))
File "mtrand.pyx", line 992, in mtrand.RandomState.randint
ValueError: Range cannot be empty (low >= high) unless no samples are taken
from ext_portrait_segmentation.
I am sorry for late answer for you.
These days, I leave from my home country for an internship in another country with a long trip.
Still, I need time more to set new experiment environment.
I will check your code this week.
Thanks for your waiting.
from ext_portrait_segmentation.
I understand, wishing you all the best.
from ext_portrait_segmentation.
I think your problem of "rand_pick = np.random.randint(0, len(val_loader))" is that
you don't use any validation dataset.
if you comment-out everything.
If you want to train with Baidu dataset
please just remove this 2 lines in loadData.py line 89, 90
self.trainImList.append(img_file)
self.trainAnnotList.append(label_file)
You can get this result
Load public Baidu train dataset
This num of data is 10160
Load public val dataset
This num of data is 270
Originally,
Load public Baidu train dataset
This num of data is 11469
Load public val dataset
This num of data is 270
This means that you train with only Baidu dataset and validate with EG1800 validation set
from ext_portrait_segmentation.
Related Issues (20)
- the size of input_image HOT 1
- Accuracy issue, implementation difference compared to paper. HOT 4
- visualizing results HOT 4
- The speed about ExtremeC3Net. HOT 1
- Rough edges in mask? HOT 2
- Training new dataset
- main.py HOT 1
- Testing new Image HOT 2
- How to use EG1800
- Error in Visualize_video.py when using result/SINet/SINet.pth HOT 5
- Why is the label of eg1800 data set inconsistent with the number of images? HOT 3
- about test!!! HOT 4
- Converting .pth model to onnx HOT 7
- about inference time
- why ignore_idx==255? in lovasz loss? HOT 2
- Pretrained model output bad segmentation result HOT 4
- trainning new data and test error, help please HOT 3
- Model does not perform well on real world portrait images/videos HOT 1
- groups config setting
- custom dataset training 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 ext_portrait_segmentation.