Shivanandroy/simpleT5
simpleT5 is built on top of PyTorch-lightning⚡️ and Transformers🤗 that lets you quickly train your T5 models.
Abstracts away PyTorch-Lightning boilerplate for sequence-to-sequence tasks, supporting multiple T5 variants (mT5, byT5, CodeT5) through a unified API that handles tokenization, batching, and distributed training configuration. Expects input as pandas DataFrames with source/target text pairs and exposes hyperparameters like token length limits, batch size, and early stopping directly in the train method. Integrates with HuggingFace's model hub for pretrained checkpoints and includes ONNX export support for inference optimization.
400 stars and 309 monthly downloads. No commits in the last 6 months. Available on PyPI.
Stars
400
Forks
60
Language
Python
License
MIT
Category
Last pushed
May 19, 2023
Monthly downloads
309
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/Shivanandroy/simpleT5"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related models
chanind/frame-semantic-transformer
Frame Semantic Parser based on T5 and FrameNet
Shivanandroy/KeyPhraseTransformer
KeyPhraseTransformer lets you quickly extract key phrases, topics, themes from your text data ...
osainz59/t5-encoder
A extension of Transformers library to include T5ForSequenceClassification class.
conceptofmind/t5-pytorch
Implementation of Exploring the Limits of Transfer Learning with a Unified Text-to-Text...
0x7o/text2keywords
Trained T5 and T5-large model for creating keywords from text