Comments (4)
Hi, before training and testing, the Kinetics400 training and validation datasets we use are preprocessed by making height of video 256, this may cause little difference. You can contact us by [email protected] for discussing more details.
from video-swin-transformer.
Facing the same problem. I use the annotation files in this repo and test the provided pretrain models without any extra modification. Swin-T and Swin-B achieved 78.4% and 80.1% top-1 accuracy on K400, which seemed slightly worse than the reported results. @zehzhang Have you resolved this issue? If so, could you please share your solution?
Thanks for confirming the problem. I got similar decrease with SwinT (-0.4% top1 acc) and SwinB pretrained on ImageNet21k (-0.5% top1 acc). I'm reaching out to the other first co-author (referred to by @hust-nj ) and hopefully will figure out what is going on soon. I will keep this thread updated.
from video-swin-transformer.
After a careful comparison, we find out that the performance gap is due to the slight difference on data. Our kinetics-400 data with 256 resolution was obtained (with broken video removed) from nonlocal networks which was also used in many other series of works.
More details and data download link can be found here https://github.com/youngwanLEE/VoV3D/blob/main/DATA.md#kinetics-400, facebookresearch/video-nonlocal-net#67
from video-swin-transformer.
@zehzhang Hi, Thanks for your issues. Do you train video-swin from scratch without imagenet-21k, did it drop severely? thank you.
from video-swin-transformer.
Related Issues (20)
- Can't export ONNX transformer HOT 3
- is there a priority between drop out and patch norm?
- Error found in the code about shift_size calculation
- How much video memory does a single GPU need to run SSV2? I use the 16G ,then it prompt CUDA out of memory
- Pretraining for SSv2
- Code for duplicate the weights pretrained on Imagenet
- Welcome update to OpenMMLab 2.0
- Pretrained Swin-L on Kinects 600 dataset
- How to resolve this error when training with swin_base_patch244_window877_kinetics400_1k.py
- Missing keys in source state_dict: backbone.patch_embed.proj.weight, backbone.patch_embed.proj.bias, and so on.
- size mismatch for cls_head.fc_cls.weight: copying a param with shape torch.Size([400, 1024]) from checkpoint, the shape in current model is torch.Size([101, 1024]).
- How to solve this: AttributeError: 'Recognizer2D' object has no attribute 'demo/label_map_k400'
- TypeError: 'str' object is not callable HOT 2
- IncompatibleKeys appear when I loaded the pretrained 2d model
- Steps to convert mmaction's video-swin-transformer to ONNX successfully HOT 4
- Model weights for Swin-L
- inconsistent patch size between swin_tiny.py and pretrained model
- swin_transformer.py compute_mask doesn't have expected torch dtype
- how to use pretrained model in simplest way? HOT 1
- Failed to run demo. 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 video-swin-transformer.