Comments (6)
Hi @mlpen,
Yes, this seems a bit low for epoch 9.
Does your dataset is full and properly encoded?
See this page for Dataset Preparation
from slowfast.
Thanks for replying. I downloaded the dataset according to this repo and then follow the instructions from slowfast to prepare the dataset.
from slowfast.
How many videos do you have in your train and validation sets?
from slowfast.
240214 in train set and 19879 in val set.
from slowfast.
OK, those are not exactly the same numbers we have but close enough.
You can try and use our pre-trained model weights and only run evaluation on the valid set. It can help understand if the gap is related to data or something else.
from slowfast.
Ok, I will try the pre-trained weights. After the model training completed, the final val accuracy is 76.3 and train accuracy is 88.5.
from slowfast.
Related Issues (17)
- how to prepare kinetics400 dataset to run your code? HOT 1
- Thermal diagram of key frame HOT 1
- Unable to load weights of the 'VIT_B_VTN' model HOT 1
- VTN not converging on UCF101 HOT 2
- Error with Detectron2 when using run_net.py HOT 1
- VTN input example HOT 4
- How to use Resnet as backbone? HOT 2
- How to visualize the cls token in VTN ? HOT 2
- Visualization: how effective is a PART of frames on decision making HOT 7
- size of checkpoint file HOT 1
- Have you preprocessed the kinetics dataset? HOT 1
- Where is the code?
- VTN attention window HOT 2
- Position IDs in VTN HOT 1
- How to reproduce result in live camera feed without saving as video HOT 2
- Model weights for ViT-B-VTN (1 layer) 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 slowfast.