Comments (11)
Confirmed fixed.
from lion-pytorch.
@nbardy do you get this behaviour without triton?
one thing i noticed here is that the triton code uses auto-tune + in-place updates which may cause issues. on the first step multiple differnt kernels will be launched which all do the same thing to see what is fastest. this is unique to the first step. usually this is not a problem when training from scratch as warmup is used but it may be here
from lion-pytorch.
Hi, thanks for the datapoint.
Do you have a comparison of the commands used for running with Lion and AdamW?
from lion-pytorch.
@xiangning-chen same command besides the lr_opt value
from lion-pytorch.
Oh I meant the learning_rate, lr_end, and weight decay comparison for Lion and AdamW.
from lion-pytorch.
They are in the main post.
‘Relevant Code’ is for lion
‘Relevant parameters’ is for Adam
from lion-pytorch.
@nbardy Sorry I'm a bit confused, in Relevant parameters
you set the --lion_opt
flag, but this is for Adam?
Can you please just tell me the learning_rate, lr_end, and weight decay for Lion and AdamW respectively, thanks!
from lion-pytorch.
@mitchellnw thanks for bringing this to my attention Mitchell!
@nbardy do you want to see if 6ab873a addresses the issue?
from lion-pytorch.
I have finally got back to training more diffusion models.
Tried upgrading to lion-pytorch==0.1.2
and still getting a reset it seems on first step
https://wandb.ai/nbardy-facet/sd_xl_train_t2iadapter/runs/eey3bj1n?workspace=user-nbardy-facet
from lion-pytorch.
lion-pytorch==0.1.2
pytorch-triton==2.1.0+e650d3708b
triton==2.0.0
torch==2.0.1
from lion-pytorch.
Turned off lion and it’s still there. This is probably something else from my changes. Will test more next week.
from lion-pytorch.
Related Issues (20)
- Instability when resuming trains HOT 22
- any new update? HOT 4
- What is the best learning rate you have found? for lora and dreambooth ty
- KeyError in update_fn_kernel when use_triton=True HOT 7
- add an 8-bit version with bitsandbytes HOT 5
- Loss explodes when resuming using trion implementation. HOT 3
- Add the implementation to official pytorch repo HOT 3
- Same amount of VRAM is taken as in AdamW HOT 6
- Strange Results on first step HOT 1
- Does the Lion optimizer work with grad accumulation? HOT 1
- Performance experiments over AdamW HOT 1
- Do you have the actual weights trained from the paper?
- Convergence guarantees for Lion
- AMD ROCM versions
- Always getting NaNs in long training HOT 5
- This issue actually still persists. My python environment:
- Using Triton with PyTorch 2.0 for AMP training results in tensors containing inf values.
- Learning rate scaling for distributed training? HOT 3
- Adaptive learning-rate optimization 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 lion-pytorch.