DeepSeek/DualPipe
1
0
Fork 0
mirror of https://github.com/deepseek-ai/DualPipe synced 2025-05-31 18:38:10 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

update readme

Browse Source
This commit is contained in:
ljss 2025-03-04 17:54:21 +08:00
parent 8ec5883b30
commit 194bfb3995
1 changed files with 3 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
README.md
View File

@ -23,10 +23,12 @@ have mutually overlapped computation and communication
full backward chunk, 𝑊 denotes the execution time of a "backward for weights" chunk, and 𝐹&𝐵
denotes the execution time of two mutually overlapped forward and backward chunks.
# DualPipeV
## DualPipeV
DualPipeV is a concise V-shape schedule derived from DualPipe using a "cut-in-half" procedure, introduced by Sea AI Lab as "Cut-in-half" in their [blog post](https://hackmd.io/@ufotalent/r1lVXsa9Jg). Thanks to them for this efficient schedule!
### Schedules
![dualpipev](images/dualpipev.png)
Example DualPipeV scheduling for 4 PP ranks and 10 micro-batches.