Skip to content

This project generate MCU script using GPT-2 model.

Notifications You must be signed in to change notification settings

fpem123/GPT2-MCU

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

17 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

GPT2 MCU

Run on Ainize

This project generate MCU script using GPT-2 model.

Fine tuning data: Kaggle

Model download: Google drive

Model information

Base model: gpt-2 large
Epoch: 30
Train runtime: 2025.88 secs
Loss: 0.0221

How to use

* First, Choose MCU character name.
* Second, Fill what the character will say in text. This will be base of script.
* And then, Fill number in length. Text is created as long as "length". I recommend between 100 and 300.
* If length is so big, generate time will be long.

Post parameter

name: The MCU character name.
text: The base of script.
length: The size of generated text. (min: 50)

Output foramt

{"0": [[character name, dialog], [character name, dialog], ...]}

Image reference

static/README.md

* With CLI *

Input example

curl -X POST "https://master-gpt2-mcu-fpem123.endpoint.ainize.ai/mcu" -H "accept: application/json" -H "Content-Type: multipart/form-data" -F "name=TONY STARK" -F "text=Today, we" -F "length=100"

Output example

{
  "0": [
    [
      "TONY STARK",
      " Today, we're gonna talk about humility. But I'm sure if you were to wrap this up, you'd say, “I know I didn't do that.”"
    ],
    [
      "HELMUT ZEMO",
      " That you did."
    ],
    [
      "TONY STARK",
      " I know. I’m a bit of a diva. I can’t control the stuff that happens around me, but I can control how I react to it. And that’s where you come in."
    ]
  ]
}

* With swagger *

API page: In Ainize

* With a Demo *

Demo page: End-point

About

This project generate MCU script using GPT-2 model.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published