mim-solutions/bert_for_longer_texts
BERT classification model for processing texts longer than 512 tokens. Text is first divided into smaller chunks and after feeding them to BERT, intermediate results are pooled. The implementation allows fine-tuning.
This tool helps data scientists and NLP practitioners classify or predict values from very long text documents, such as full articles, detailed reviews, or extensive reports. It takes your raw text and assigns it to categories (like sentiment or author) or predicts a numerical value (like a product rating). This is ideal for anyone working with text that exceeds typical length limits for common language models.
146 stars. No commits in the last 6 months.
Use this if you need to fine-tune a BERT or RoBERTa model for text classification or regression on documents longer than 512 words.
Not ideal if your texts are consistently short or if you prefer using models like BigBird or Longformer, which have different internal architectures.
Stars
146
Forks
33
Language
Python
License
—
Category
Last pushed
Jun 19, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/mim-solutions/bert_for_longer_texts"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
lonePatient/Bert-Multi-Label-Text-Classification
This repo contains a PyTorch implementation of a pretrained BERT model for multi-label text...
OctoberChang/X-Transformer
X-Transformer: Taming Pretrained Transformers for eXtreme Multi-label Text Classification
QData/LaMP
ECML 2019: Graph Neural Networks for Multi-Label Classification
illiterate/BertClassifier
基于PyTorch的BERT中文文本分类模型（BERT Chinese text classification model implemented by PyTorch）
a-tokyo/ai-zero-shot-classifier
🧠 leverage advanced AI embeddings to perform multilingual zero-shot text classification. Whether...