← all repositories

pytorch/text

A PyTorch library providing data loaders, datasets, vocabularies, and text-processing transforms for NLP.

3.6k stars Python Data ToolingML Frameworks
text
Velocity · 7d
+1.0
★ / day
Trend
steady
star history

Torchtext is a PyTorch ecosystem library for natural language processing. It provides raw text iterators for common NLP datasets, vocabulary and embedding classes, text-processing transformations, and pre-trained model utilities. The library enables loading, preprocessing, and tokenizing text data for training deep learning models in PyTorch.

heatdrop uses Google Analytics to see which pages get read — nothing else. Your call. How we handle data.