Coder Social home page Coder Social logo

generative_ia_with_large_leangue_models's Introduction

Generative_IA_With_Large_Leangue_Models1.

segameaker de aws como usar sagemaker (es pago)

Esta es la guia del curos: https://labs.vocareum.com/main/main.php?m=clabide&mode=s&asnid=1843537&stepid=1843538&hideNavBar=1

  1. https://github.com/veiro/Generative_IA_With_Large_Leangue_Models/blob/main/Lab_1_summarize_dialogue.ipynb

Resumen: Carga el model T5 y hacer fewshots traingng para ver como mejora los resultados en resumen de dialogo https://www.coursera.org/learn/generative-ai-with-llms/gradedLti/loNJu/lab-1-generative-ai-use-case-summarize-dialogue.

Explicacion del laboratorio https://www.coursera.org/learn/generative-ai-with-llms/lecture/wno7h/lab-1-walkthrough

  1. https://github.com/veiro/Generative_IA_With_Large_Leangue_Models/blob/main/Lab_2_fine_tune_generative_ai_model.ipynb Fine-tuning

Resumen: usando soft propmts de un modelo T5. Luego Fine-tuning con la tecnica PEFT y compara con flan-T5 que es uno finetuneado por ellos. Comapra con la metrica ROGUE como funciona el resumen. En resumen el de ellos es mejor, pero llevo mucho mas computo que las tecnicas utilizadas en este laboratorio.

Explicacion del laboratorio https://www.coursera.org/learn/generative-ai-with-llms/lecture/A6TDx/lab-2-walkthrough

  1. https://github.com/veiro/Generative_IA_With_Large_Leangue_Models/blob/main/Lab_3_fine_tune_model_to_detoxify_summaries.ipynb

Resumen: En este cuaderno, ajustará un modelo FLAN-T5 para generar contenido menos tóxico mediante el modelo de recompensa por incitación al odio de Facebook. El modelo de recompensa es un clasificador binario que predice "no odiar" u "odiar" por el texto dado. Utilizará la optimización de políticas próximas (PPO) para ajustar y desintoxicar el modelo.

Explicacion del laboratorio https://www.coursera.org/learn/generative-ai-with-llms/lecture/yPzI4/lab-3-walkthrough

generative_ia_with_large_leangue_models's People

Contributors

veiro avatar

Watchers

 avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo 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.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.