Comments (7)
65.748 is not the single-view result, it is the average precision of all clips (when testing, each video has 5x3 views). In validation, we always take the middle clip of the video, so the accuracy(67.0) is higher.
from videomaev2.
The inconsistent results when reloading ckpt are due to the run-time parameter, such as running_mean
& running_var
in normalization layers.
During training, each model on every GPU has its own run-time parameters. However, we only save the checkpoint on GPU0. When reloading the model, the run-time parameters of the model on other GPUs are loaded from the model on GPU0, leading to slight differences in the results.
from videomaev2.
consistent results when reloading ckpt are due to the run-time parameter, such as
running_mean
&running_var
in normalization layers.During training, each model on every GPU has its own run-time parameters. However, we only save the checkpoint on GPU0. When reloading the model, the run-time parameters of the model on other GPUs are loaded from the model on GPU0, leading to slight differences in
As far as I know, VideoMAE has no BatchNorm so we needn't synchronize the running_mean
& running_var
. Are there any other parameters that need to be synchronized? And can we avoid this problem?
from videomaev2.
Another possibility is inconsistent batch sizes. When the testing data cannot be evenly divided by the batch size, the last batch will randomly select some videos to fill in.
from videomaev2.
I can't find that in your code. Can you show me the location of the corresponding implementation?
from videomaev2.
VideoMAEv2/run_class_finetuning.py
Lines 439 to 442 in 9492db0
This code is modified from DeiT, and I haven't looked closely at how it's handled, but it should be related to the sampler.
from videomaev2.
I see, thank you!
from videomaev2.
Related Issues (20)
- Error when running runclass_finetuning.py HOT 3
- Turning VideoMAEv2 into a next-frame prediction model HOT 1
- Extracting Features from Frame Level Data HOT 1
- VideoMAEv2-L Weights/Checkpoints HOT 1
- Pretrained smaller models availability HOT 1
- Initialize student model's weights HOT 3
- Apply VideoMAEV2 to other directions. HOT 2
- Finetuned smaller models HOT 1
- Could you provide features for ActivityNet 1.2 and ActivityNet 1.3 features extracted by videomaev2 ?
- Could you provide ActivityNet 1.2 and ActivityNet 1.3 features extracted by videomaev2 ? HOT 1
- Where to find the script of finetuning on 'Temporal action detection' task? HOT 6
- What should I do if I want to get the features of ActivityNet-1.3? HOT 7
- The parameter grad_norm appears to be inf and then nan when input resolution is 112*112 during the pre-training on VIT-Small backbone HOT 1
- 2333
- fine-tuning AVA dataset for spatiotemporal detection
- Impact of Something Something and Kinetics during Unlabeled Pre-training
- Code implementation of model inference
- Finetuning with more than 16 frames
- CLS token
- Zero-shot evaluations on downstream datasets
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 videomaev2.