Skip to content

The author has trained the model to predict the generated code in Efficient-Code-Generation-with-E-Code

License

Notifications You must be signed in to change notification settings

CodeGeneration2/Generated-code-has-been-predicted

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

9 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Generated-code-has-been-predicted

In the Efficient-Code-Generation-with-E-Code work, the authors use a fine-tuned pre-trained model to predict a range of codes to be generated. Due to the need for comparative experiments, three code generation models are available. The three code generation models are E-code 350M, GPT-Neo 125M, and No expert group E-code 350M. We use each of the three fine-tuned code generation models to generate codes. Below we have the code generated by the three fine-tuned code generation models.

E-code 350M

We give the results of 3 times code generation in the E-code 350M model.

GPT-Neo 125M

We give the case results of one code generation for the GPT-Neo 125M model.

No expert group E-code 350M

We give the case results of one code generation for the no expert group E-code 350M model.

About

The author has trained the model to predict the generated code in Efficient-Code-Generation-with-E-Code

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published