Comments (4)
I have prepared a pull request that should fix the issue (#10). I think that this should not be added to the global loss.
from vicregl.
I see that you have used detach. The learned representation was thus not affected by the gradients coming from the classification loss. Still, this is ugly and the actual loss value does thus not represent the actual VICRegL loss.
from vicregl.
@Adrien987k thank you for the work and the implementation.
I have noticed that you are also using classification loss, in addition to the VICReg and VICRegL losses. The classification loss is not mentioned in the paper and was also not used in the original VICReg.
On the first hand, it seems a bit off, given that you won't have labels available when pretraining in the wild and it destroys the purpose of "self-supervised" pre-training ?
Hi! have you ever tried to reproduce the models without the classification loss?
from vicregl.
Hi,
Thanks for pointing this out, as you mentioned, there is a detach operation preventing the gradient for propagating from the classification head to the pretrained backbone.
There is absolutely no difference between having this head and not having it. The VICRegL loss is just the total loss minus the classification loss, it will not change the learning dynamics.
from vicregl.
Related Issues (17)
- README typo HOT 2
- Linear segmentation evaluation HOT 4
- where were they from to views and locations? were they from f"{prefix}_images.npy"? HOT 2
- Missing maps_projection in the checkpoint and batch_norm in the code HOT 5
- High local loss while finetuning resnet50 HOT 4
- Unable to reproduce object detection results on Pascal VOC
- Plots of losses for different configurations HOT 1
- Request for Change in Licensing Terms HOT 1
- Dimension of global and local
- Does this have a requirements file? HOT 1
- fine tune vicregl to coco object detection
- Missing keys in ConvNeXt state dict HOT 2
- Need seg pretrained model
- Do you really use the top-k filtering on feature-based matching? HOT 3
- Symmetry on local and feature-based loss functions HOT 2
- No 'InterpolationMode' in torchvision 0.8.2 HOT 2
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 vicregl.