Comments (3)
I would like to contribute
from deepspeed.
Hi @Tejaswgupta, thanks for reaching out and wanting to contribute. Is there something specific you have in mind you'd like to contribute? If it was this issue I do have a PR being reviewed currently that fixes this issue. We're in the process of trying to create some new issues that would be good to take care of as a new user. Keep an eye out for those :)
@dapurv5, Thanks for bringing this bug to our attention. We previously fixed this type of bug in a different code path but now have added unit tests for it and fixed it in your code path and one other. The PR should be merged soon.
from deepspeed.
Sure @jeffra :)
from deepspeed.
Related Issues (20)
- [BUG] Does deepspeed MoE support dynamic network (with unused parameters)? HOT 5
- [BUG] Pipeline Parallel: Only few trainable parameters, Only GPU0 has parameters whose requires_grad is "True". Cause ValueError: optimizer got an empty parameter list when deepspeed.initialize
- [BUG]Loss decreases cyclically with epoch as the cycle HOT 1
- [BUG] 'type:transformer' partitioning doesn't ensure non-zero parameters on each pipeline rank. HOT 5
- [BUG] Profiler records values that differ from transformers
- [BUG] Problems with MiCS training HOT 1
- [BUG]Pipeline Parallel: engine.train_batch(data_iter=training_dataloader) HOT 1
- [BUG] deepspeed_aio swap file name too long. HOT 3
- [BUG] Error: Sizes of tensors must match except in dimension 1. Expected size 64 but got size 512 for tensor number 2 in the list. HOT 4
- A possible solution to resolve the issue of deepspeed.initialize() hanging.
- nv-nightly CI test failure HOT 1
- {{ env.GITHUB_WORKFLOW }} CI test failure
- [BUG] Error: Preparing DeepSpeed ZeRO stage 2 optimizer HIP error on AMD MI200 GPUs
- Confusing memory usage estimations
- [BUG] Unexpected GPU memory consumption when using transformers PEFT in Zero3
- [REQUEST] ZeRO - introduce replicas to keep GBS from getting too large on hundres of gpus HOT 4
- How to offload/move model weights manually ?
- Changing offload to NVMe instead of CPU causes error: TypeError: unsupported operand type(s) for *: 'NoneType' and 'int' HOT 2
- [BUG] TypeError: Can't instantiate abstract class MPS_Accelerator with abstract methods HOT 1
- [BUG] main_process_ip not working HOT 1
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 deepspeed.