Join the conversation

Join the community of Machine Learners and AI enthusiasts.

Sign Up
awacke1Β 
posted an update 3 days ago
Post
1977
I introduce MIT license

ML Model Specialize Fine Tuner app "SFT Tiny Titans" πŸš€

Demo video with source.

Download, train, SFT, and test your models, easy as 1-2-3!
URL: awacke1/TorchTransformers-NLP-CV-SFT

This morning I completed the agentic demonstration with the SFT fine tuned model. Now the demo does the following:

  1. Allows user to download an open model they can specialize fine tune easily.
  2. Provide a CSV file with fine tune QA dataset.
  3. Using torch and transformers create an SFT output model with the new data trained to the model with 3 epochs.
  4. Ability to Test new model with list of QA questions and compare answers to expected.
  5. Demonstrate use of agents with smolagents tools.

Try it here:

https://huggingface.co/spaces/awacke1/TorchTransformers-NLP-CV-SFT

Download ML Models:

image.png

Build ML Models with Specialized Fine Tuning to a CSV:

image.png

Test your SFT Version with your changes.

image.png

Have an Agentic RAG Party with agents and your SFT model

image.png

Hope you enjoy and this demo is useful to you on your learning journey in ML!!

--Aaron Wacker
https://huggingface.co/awacke1
https://x.com/Aaron_Wacker

Β·

I have tried with a small data set and it never finished the training. It took really, really long time. I don't know how long. Maybe one hour. I don't know. I was just watching how it is in the stage one, but it was very, very long time.