Comments (1)
Yes, for single point precision there is a limit to how accurate things can be. I wouldn't call this an obvious difference however, you are summing up the error over lots of locations. If I follow your code with a slight modification:
import torch
import pytorch_wavelets
DWTForward = pytorch_wavelets.DWTForward()
DWTInverse = pytorch_wavelets.DWTInverse()
x = torch.abs(torch.randn(17, 7, 32, 32))
yl, yh = DWTForward(x)
y = DWTInverse((yl, yh))
e = y - x
then the standard deviation of e
is around 1e-7
, giving a SNR of 137dB, which by all accounts is a very very small error. If you do need more precision however, then you can move to double precision with torch.set_default_dtype(torch.float64)
before creating the DWT objects. This reduces the std of e
to 1e-16
from pytorch_wavelets.
Related Issues (20)
- Padding for powers of 2 HOT 1
- Reconstructed wavelet HOT 1
- Export DWTForward to onnx HOT 1
- Not preserving spatial dimensions HOT 1
- Make pypi release with 1D DWT
- using dtcwt as a high freq feature extractor HOT 1
- DWTInverse is not the inverse of DWTForward for small image sizes.
- ValueError: step must be greater than zero HOT 1
- Compatiable with torch AMP?
- How does output yl reflect the two approximate subbands?
- How to change default tensor dtype in pytorch_wavelets HOT 2
- Why must kernels be fixed?
- Hints on avoiding the scaled ouput
- Installation error HOT 1
- How to implement DTCWT for 1-D signals? Looking forward to your answer.
- py37_cu102/fused/fused.so
- How to calculate the params of DWT and IDWT?
- something wrong when we use encoder to process the high_frequency component
- 3D wavelet transform HOT 1
- maximum recursion depth exceeded in comparison
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 pytorch_wavelets.