Comments (7)
Hi, according to the images the model is not training, it's strange. Your result looks like they are just in the first two epochs. When we train with BIPED, we augment the data 200 by 280. This mean that in one epoch you used 56000 images. If you just used 1000 images this means that your 40 epochs used just 40000 images.
from dexined.
Ok, my original Image is 1024*768
, I will do random cropping and run it for 56*20
epochs, which should be equivalent to your 20 epochs. I've uploaded the code to the server I will let you know once I get the results
from dexined.
This is my tensorboard result, I will just cancel the training since it plateaued for 2 hours right now(I do added a lr.scheduler) . And the provided images are very similar to the images provided last time
from dexined.
Ok, my original Image is
1024*768
, I will do random cropping and run it for56*20
epochs, which should be equivalent to your 20 epochs. I've uploaded the code to the server I will let you know once I get the results
I think you need to use the augmentation methods of BIPED, check the BIPED repo and let me know. Maybe we can have a chat by zoom, I'm interested how ends your proyect.
from dexined.
This is my tensorboard result, I will just cancel the training since it plateaued for 2 hours right now(I do added a lr.scheduler) . And the provided images are very similar to the images provided last time
Can you tell me which kind of machine are you using for training? With dexined 2 hours is a short time.
from dexined.
Hey, I am using GPU equivalent to Nvidia V100 with 16GB Vram but it is neither Nvidia nor Amd. I will check the augmentation some time. I would be happy to have a zoom call some day with you once the new computation result is out. I guess you are in Spain? so we can probably do around 18:00 Beijing Time, basically +7 for EU time zone.
from dexined.
Hey, I am using GPU equivalent to Nvidia V100 with 16GB Vram but it is neither Nvidia nor Amd. I will check the augmentation some time. I would be happy to have a zoom call some day with you once the new computation result is out. I guess you are in Spain? so we can probably do around 18:00 Beijing Time, basically +7 for EU time zone.
Ok, let me know when you get the result. Actually, I'm in Ecuador, You may write me by my email in the github profile.
Have a good day.
from dexined.
Related Issues (20)
- about evaluation HOT 2
- Works once but stops working after testing own image HOT 3
- Can I change the resize value in dataset transform function? HOT 3
- About Bsds and Biped HOT 2
- Size of the CLASSIC images
- About args.double_img HOT 5
- Can't parse 'dsize'. Sequence item with index 0 has a wrong type HOT 2
- Should the l_weight value change according to the dataset? HOT 3
- Inference mode is available? HOT 2
- GT HOT 1
- loss does not converge HOT 25
- How can I calculate ODS, OIS, AP metrics? HOT 5
- does DexiNed support multiple GPUs .. if Yes, how to implement it for DexiNed TF2 HOT 4
- high train loss
- About Precision-Recall Curves on Biped
- Getting an NoneType error while testing with my images HOT 1
- Interpretation of result.png HOT 2
- Hello! Why does the following problem occur after running main.py
- Can you share the results of DexiNed's testing on top of the BSDS500?
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 dexined.