Skip to content
Proyectos

Whiax/BERT-Transformer-Pytorch

Ver en GitHub

Whiax/BERT-Transformer-Pytorch has 46 GitHub stars and 7 forks. Star changes are not measured yet. Last checked 2026-11-14 08:03 UTC. Last push 2026-06-27.

Cómo se miden estas cifras

Crecimiento de estrellas medido

in 24h
Recopilando historial
in 7 days
Recopilando historial
in 15 days
Recopilando historial
in 1 month
Recopilando historial

Cambios netos en períodos que se solapan, no totales diarios. Los datos no disponibles se dejan en blanco. Las estrellas indican interés, no calidad.

Acerca de

Basic implementation of BERT and Transformer in Pytorch in one short python file (also includes "predict next word" GPT task)

Aprendizaje automático

README.md

Original

BERT-Transformer-Pytorch

Basic implementation of BERT and Transformer in Pytorch in one python file of ~300 lines of code (train.py).

This project aims to provide an easy-to-run easy-to-understand code for NLP beginners and people who want to know how Transformers work.
The project uses a simplified implementation of BERT (no labels are required for training).
The original implementation of Transformer uses an encoder and a decoder, here we only need the encoder.
The model can train in 30 minutes on 1 x RTX2070Super GPU.

Visualization of word embeddings: alt text

Implementation details: https://hyugen-ai.medium.com/transformers-in-pytorch-from-scratch-for-nlp-beginners-ff3b3d922ef7

"Predict next word" task

August 2023 update:

  • For experiment purposes, I also implemented the "predict next word" task which is used to train GPT.
  • The code can be found in "main_predictnextword.py"
  • This code is a slight modification of train.py

Explorar proyectos relacionados