Comments (4)
Hi @XuMengyaAmy
Thanks for your interest in our work.
(1) The training set should be in webdataset format with image text pairs. The val
is just for zero-shot image classification evaluation, so it should be in [image, cls] pairs with webdataset format.
If you would like to evaluate on a custom segmentation dataset, please follow MMSegmentation dataset prepare tutorial
(2) It follows CLIP style here. You may read this paper for more details.
(3) Please follow data convert section to prepare your dataset, for example, gcc3m.
from groupvit.
Thank you for the answer. I have read the CLIP paper, but could not find any information that would address the following question. Regarding the point (2), is this definition of loss based on the assumption that the labels corresponding to samples within each batch are different? Would this loss still be valid if I have a batch size of 128 with only two classes for training?
from groupvit.
Thank you for the answer. I have read the CLIP paper, but could not find any information that would address the following question. Regarding the point (2), is this definition of loss based on the assumption that the labels corresponding to samples within each batch are different? Would this loss still be valid if I have a batch size of 128 with only two classes for training?
Have you resolved the question (2)? I also have a same confusion.
from groupvit.
My understanding is that the CLIP-based loss is effective if the labels (entities) within the training batch are different. And that seems to be the cases with the training datasets used in groupViT. For example the gcc3m has 16K entities (classes, or labels). I had to find it the hard way by running many experiments with limited number of entities. Let me know if you had a different understanding/observation.
from groupvit.
Related Issues (20)
- Inconsistent Results from the paper
- Parallel training hangs in reduce_tensor method HOT 1
- Learnable group tokens fine-tuning on out-of-domain datasets HOT 1
- Checkpoint for the model trained just with 1 grouping stage HOT 1
- Inquiry about the Grouping Blocks HOT 2
- error when use single node and multi gpu for training HOT 5
- background threshold issues HOT 1
- How fast can this model run?
- YFCC subset taking very long to process HOT 1
- Visualization of Concepts Learnt by Group Tokens
- output masks consistently showing the same structure HOT 5
- How to generate the mask in Figure 1 in Appendix?
- Questions about input pred label file HOT 3
- question about training setup [#GPUs].
- use gumbel softmax only for training
- How to generate image-text pairs when i want to use other datasets such as ADE20K or medical segmentation dataset? HOT 1
- Command 'apex' not found HOT 1
- AttributeError: module 'torch.nn' has no attribute 'backends'
- Open-vocabulary semantic segmentation HOT 1
- Are there any plans to retrain a version with larger data and parameter scale?
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 groupvit.