Comments (2)
lessw2020
commented on July 17, 2024
2
Hi LifeIsStrange -
re: "I would have wished that state of the art in one Nlp task would be a combination of e.g 50 innovative and complementary ideas from researchers."
We did just that this week :) I'm working on a Medium article about it but by combining Ranger (Lookahead + RAdam) + Mish + new flat/cosine lr curve + simple self attention...
we just beat the FastAI leaderboard for 5 epochs by a very large margin - was 55.2%, we got it to 74.97%.
So yes, combining all the best ideas really can and does work. I'll try and get the Medium article about it out soon :)
from ranger-deep-learning-optimizer.
lessw2020
commented on July 17, 2024
Here's the article on our efforts:
https://medium.com/@lessw/how-we-beat-the-fastai-leaderboard-score-by-19-77-a-cbb2338fab5c
I've got a new repo setup with all the tools in there for anyone to try out our setup of combining Ranger+Mish + Flat-Cosine annealing.
Hopefully we'll make more progress integrating other new ideas!
from ranger-deep-learning-optimizer.
Related Issues (20)
- Is there a publication of Ranger? HOT 2
- It makes sense to use it on a batch of 1? HOT 3
- Benchmarck Adaptive Scheduling of Stochastic Gradients
- Gradient centralization was updated HOT 4
- [question] Why Ranger is not available as a pip package HOT 3
- What the "GC operations" mean?
- TypeError in GC operation for Conv layers and FC layers
- Not able to save the model_state_dict. HOT 1
- Does it works well for transformer?
- The results I tested on the cifar10 dataset are as follows. Ranger's results look strange HOT 1
- RangerVA with GC
- This overload of addcmul_ is deprecated: addcmul_(Number value, Tensor tensor1, Tensor tensor2) HOT 5
- How to use ranger in keras? Please help me. HOT 1
- Stochastic Weight Averaging support HOT 1
- Loss stuck after 1 epoch HOT 1
- Is adabelief the best optimizer? HOT 7
- best result : flat learning rate for 75% it means ranger optimizer is not sensitive to lr? HOT 1
- Please note in the documentation (or in the constructor) that closures must be enabled
- ranger and cosine annealing LR leads to different schedule than SGD optimizer? o_O
- Collate pip package so that it picks up from main repo. 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 ranger-deep-learning-optimizer.