Skip to content

simusid/byol

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

19 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Comparing scripts generated by OpenAI o3 and deepseek R1

R1 first output included two syntax errors. It left off two closing ")" in two function declarations. Then it had a data type issue. It treated the v1,v2 values as a list rather than tensor and was unable to execute .to(device). The first re-prompt fixed this.

o3 version worked fine out of the box with zero modifications.

training

Deepseek appears to consume more power than o3. will evaluate total runtime for each as well as backbone model used.

byol_deepseek took 104 minutes, byol_o3 took 115 minutes

About

LLM-generated implementations of BYOL method

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages