Comments (1)
vztu
commented on August 24, 2024
That's a great question. However, our main experiment on imagenet-1k at 224x224 resolution didn't allow us to do this experiment. Why? Because our backbone has four stages: (224) -- 56 -- 28 -- 14 -- 7. We don't quite have many choices on the patch-size/grid-size parameter. This is the reason why Swin Transformer only uses window-size=7 as well.
However, if we run experiment on imagenet-1k at 256x256 resolution, we may have several options - {2,4,8}. We haven't done experiment on this bit, unfortunately.
from maxvit.
Related Issues (19)
- Extract the features
- How to fit for different input size? HOT 1
- How to use your checkpoint file in the pytorch ? HOT 1
- convert your tensorflow pretrained weights to the pythoch format HOT 1
- Some questions about grid_partition
- Is there a pytorch implementation? HOT 2
- COCO Training Details
- How to load imagenet21k pretrained weights
- Add preprocessing method that takes the raw image tensor
- Get The process cannot access the file because it is being used by another process error on Windows
- Maxvit local global
- About Classification Code in ImageNet HOT 1
- Where are the pretrained models? No open? HOT 2
- maxvit-gan? HOT 1
- Bug: get_config() not in MaxViT HOT 1
- Gradients do not exist for variables 'maxvit/block_00_00/attention/relative_bias:0' HOT 8
- Exact Training Details to Reproduce Tiny Model Results HOT 3
- No module named 'maxvit.models’ HOT 3
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 maxvit.