gsarti/t5-flax-gcp

Tutorial to pretrain & fine-tune a 🤗 Flax T5 model on a TPUv3-8 with GCP

35
/ 100
Emerging

This project helps machine learning engineers or researchers train custom text generation models from scratch or adapt existing ones. You provide a large corpus of text data, and it outputs a highly optimized, domain-specific text generation model (like T5) that can then be used for tasks such as summarization or translation. It’s designed for those who need to build cutting-edge language models using Google Cloud's powerful TPU infrastructure efficiently.

No commits in the last 6 months.

Use this if you need to pretrain or fine-tune a T5-like large language model on your own extensive text dataset using Google Cloud TPUs to achieve high performance and speed.

Not ideal if you primarily work with NVIDIA GPUs, prefer PyTorch or TensorFlow natively without JAX/Flax, or if your text generation needs can be met by off-the-shelf models without custom training.

large-language-models natural-language-processing model-pretraining cloud-ml-ops text-generation
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 8 / 25
Maturity 16 / 25
Community 11 / 25

How are scores calculated?

Stars

58

Forks

6

Language

Python

License

Apache-2.0

Last pushed

Jul 28, 2022

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/transformers/gsarti/t5-flax-gcp"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.