Comments (4)
Besides,What is the ratio between training datasets and validation datasets?
from mbrs.
Hello, thanks for your issue. It's just because I forget to change the loading function in the code.
In fact I load the whole dataset into the memory before training. It takes long time to load them at once, but will speed up the later training, just like a future step for pin_memory (you can find I store the loaded dataset in train_dataloader in train.py).
But it seems not the normal usage of torch.utils.data.dataset, so I will change it back to use getitem function, before next Monday.
For the training time, it takes about 12 hours to train the model MBRS_256_256 which embeds 256bits into a 256*256 RGB image, with 4 2080Ti GPUs. And it will be much quicker to train for a 128*128 image.
from mbrs.
In my implementation, I used 5000 images for validation because it didn't take much time. You can change it as you like.
from mbrs.
Code updated. Closed.
from mbrs.
Related Issues (16)
- Training Error HOT 5
- Two-stage training strategy HOT 7
- unpack error HOT 2
- decode the message HOT 2
- The extraction of watermarks has a high error rate HOT 6
- 关于message diffusion后的图像 HOT 2
- BUG in SENet.py HOT 1
- Crop attack HOT 7
- About Dataset HOT 1
- The calculation formula of PSNR HOT 3
- An error in ssim? HOT 2
- 水印图像视觉质量 HOT 6
- run-time error HOT 2
- what(): NCCL Error 1: unhandled cuda error HOT 2
- The message length of the diffusion model HOT 4
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 mbrs.