Comments (2)
Can you please first follow all default setting on a single GPU to check the performance?
I did not run on 8 GPUs. So the default learning rate 0.05 is perhaps mostly suitable for batch size of 64. Maybe you need to scale the learning rate linearly according to the batch size, see this paper
Even you scale up the learning rate, the default setting might still be slightly better. I have compared two settings: (1) batch size of 64, learning rate 0.05; (2) batch size of 128, learning rate 0.1. In general, training from scatch with (1) gives better vanilla models, so I stick to (1). That's also why If you compare my vanilla models with other papers, you could see our baseline is stronger.
from repdistiller.
I also found something I did not describe in the repo. It has been described in the paper, though.
When training the following 3 models with batch size of 64, you need a learning rate 0f 0.01 to achieve better accuracy (better than 0.05): 'MobileNetV2', 'ShuffleV1', 'ShuffleV2'
. I just added two lines here to hard-code it. Please pull it again and sorry for missing it :(
I guess part of the reason of underperformance when student and teacher are of different architectures, is that you used different learning rates. For example, when using above 3 models as student and distilling from teacher model, the learning rate is correctly set as 0.01 (which was hard-coded), but it is incorrect (0.05) for the vanilla model (as I forgot to hard-code it, now fixed). And if you divide the learning rate by # of gpus, e.g., 8, things get reversed. The second wrong one actually becomes better (0.008), and the first one (originally better on single GPU) becomes worse (0.00125).
from repdistiller.
Related Issues (20)
- Cross modal KD implementation release? HOT 1
- Question on memory consumption for CRD loss when the dataset is very large HOT 3
- Why "opt.nce_k" in dataset cifar100 is 16384? How can I get this ? HOT 2
- Has anyone implemented Wasserstein Contrastive Representation Distillation HOT 1
- Error while running the code
- Training scheme for linear probe on STL10 and TinyImagenet
- Problem of the order of the normalization in Similarity-Preserving loss.
- resnet structure seems to be a bit wrong HOT 3
- Failed to download the teacher models HOT 2
- about using the resnet models for cifar10 HOT 1
- crd used in image enhancement task like Denoise\SR\Deblur.
- Ensemble Task Implementation HOT 2
- Question about normalization constant Z_v1 and Z_v2 in the ContrastMemory
- Why using log_softmax instead of softmax? HOT 1
- Hyperparameter Settings for KD on Imagenet
- Is Ensemble distillation also included?
- the result is different in resnet56 HOT 1
- ERROR :run ./fetch_pretrained_teachers.sh HOT 1
- How to use myself datasets? HOT 1
- No dev set split
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 repdistiller.