Comments (2)
hijkzzz
commented on May 29, 2024
This parameter is set by DeepSpeed command tools, i.e, deepspeed xxxx.py
We used the ZeRO parallel + Ray
from openrlhf.
ZiyiLiubird
commented on May 29, 2024
Does OpenRLFH support ZeRO parallel + DDP inference? For example, I have 8 GPUs and each GPU loads a llama 13B model. So the data is divided into 8 batches to send 8 GPUs every time, and I gather all inference results from all GPUs. Furthermore, ZeRO supports larger batch inference.
from openrlhf.
Related Issues (20)
- RLHF for classification tasks HOT 2
- 启用PPO Ray后无响应 HOT 3
- PPO采用zero 3 stage后产生time out error HOT 1
- maybe data bug with dpo trainer HOT 1
- QLORA model loading error HOT 5
- 我们正在对比DSchat跟OpenRLHF的性能以便完成选型工作，能否提供下修复后的DSChat代码，从而复现社区提供的性能对比数据 HOT 7
- Avoid monkey patching vLLM HOT 1
- Claim your paper on HF HOT 1
- [Question] EOS in reward model dataset HOT 2
- action_log_probs重复计算 HOT 2
- Support Llama-3 models HOT 1
- Incompatibility with Qwen HOT 2
- Suggestion on the configurations HOT 1
- Strange Kill of Critic Model HOT 5
- 使用Deepseek-lite训练DPO，显示expected mat1 and mat2 to have the same type, but got: float != c10: : BFLoat16 HOT 2
- Will 2 x GPU setups be supported HOT 1
- Dummy token for prompts in HH datasets HOT 2
- Does this codebase consider using "torch.compile"? HOT 1
- wrong action_log_probs returned? HOT 1
- 可以增加支持SimPO吗
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 openrlhf.