Comments (2)
Thanks for your comments.
I'll check it and training it again.
Thank you :)
from stand-alone-self-attention.
I think you're right, the relative embedding are added to the keys and not the values. As detailed in eq. 3 of the paper. But given that we unfold the keys, we need to compute the shifted keys before that. So we first shift the keys based on the relative positions and then unfold them.
k_out_h, k_out_w = k_out.split(self.out_channels // 2, dim=1)
k_out = torch.cat((k_out_h + self.rel_h, k_out_w + self.rel_w), dim=1)
k_out = F.unfold(k_out, kernel_size=(self.kernel_size, self.kernel_size), stride=self.stride)
from stand-alone-self-attention.
Related Issues (20)
- The wrong imp of the inner-product operation HOT 3
- problem with unfold HOT 1
- Does not work when out_channels is not even
- Question about einsum.
- how about replacing einsum with normal multiplication
- Train with IMAGENET HOT 3
- matrix multiplication instead of scalar dot product HOT 4
- Problems about groups HOT 4
- Error loading pretrained model HOT 1
- Excessive Memory Usage HOT 14
- Can anyone train resnet50 successfully without NaN HOT 3
- Has anyone tried changing the batch size HOT 1
- Add `sum(dim=2)` for dot-product HOT 1
- A question about relative position embeddings HOT 1
- Stand alone self attention combine with CycleGAN
- Loss is NaN HOT 2
- Large memory consumption HOT 2
- How to calculate the relative positional embeddings from a row offset and column offset? HOT 3
- Is the 89% acc reported in the readme consisitent with the paper? 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 stand-alone-self-attention.