Comments (8)
Thanks, @jonahduncan . I revised the codes and uploaded the repo again.
If you have any other problems, please let me know.
from deepsegmentor.
Hi, did you change anything in test_roadnet.sh
? It seems that there was an unrecognized arguments 0
, which I used for setting the gpu id. @jonahduncan
from deepsegmentor.
The only modification that I made was to change python3 to python as that's what points to the python version for my set-up.
from deepsegmentor.
Do you solve the problem? I close this issue first, if you still have problem, I can reopen it later.
from deepsegmentor.
@yhlleo I have not yet been able to solve this issue
from deepsegmentor.
If you revised GPU_IDS=$1
to something like GPU_IDS=0
, you don't need to call the script in this way: sh ./scripts/test_roadnet.sh 0
. In this case, please call sh ./scripts/test_roadnet.sh
directly.
from deepsegmentor.
If you revised
GPU_IDS=$1
to something likeGPU_IDS=0
, you don't need to call the script in this way:sh ./scripts/test_roadnet.sh 0
. In this case, please callsh ./scripts/test_roadnet.sh
directly.
I've tried this before and didn't have any success. I still get the same error. I get the error even if I remove the gpu_id argument when calling test.py and relying on the default. It appears to be related to the the use_augment argument, which isn't recognizing 0, which make sense when I look at the argument in base_options.py:
parser.add_argument('--use_augment', action='store_true', help='using data augmentation for training') #parser.add_argument('--use_augment', type=int, default=1, help='using data augmentation for training')
From my understanding of your code, this argument is used in both deepcrack_dataset.py and roadnet_dataset.py to apply affine transforms with if self.opt.use_augment: ...
. I assume that your intent is to set this to false. I was able to run test_roadnet.sh by removing --use augment 0
from the script, which I saw from the resulting options list sets it to False.
The only other thing I had to do was to modify pretrained_net_G.pth to be latest_net_G.pth
from deepsegmentor.
Hi @jonahduncan ,
I checked again, it's an error of the released codes. I revised the use_augment
type from int
to action
(see details here). So, you're right to simply delete the line --use augment 0
from the script to keep the parameter value to false.
To test the pre-trained model, you're right to rename the file.
from deepsegmentor.
Related Issues (20)
- Abnormal html saved output HOT 2
- [Q&A] What is the <path_to_segmentation> HOT 2
- weights balancing for multilabels data HOT 1
- Train on my dataset
- "All CUDA-capable devices are busy or unavailable" when running on Google Colab HOT 6
- Target size error during train HOT 1
- About training
- 当我在运行train.py遇到了ValueError 好像是 张量不匹配 HOT 4
- 在运行这个代码时并得到可视化结果需要修改哪些参数,和手动创建哪些文件夹呢 HOT 2
- Error 803 HOT 3
- Is the evaluation process omitted?
- labels HOT 1
- ValueError: signal number 32 out of range HOT 1
- Training issue due to environment.yml file is not updated HOT 10
- BATCH_SIZE too small?
- How to evaluate the edge,centerline?
- roadnet评估问题 HOT 1
- Issue when loading weights for the deepcrack model HOT 1
- Can I get the crack of a picture that is taken by myself?
- How to set up datasets/ folder for "RoadNet" 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 deepsegmentor.