electra
ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators
AI summary
Language model training
A method for pre-training transformer networks to learn language representations from text data without labeled supervision
- stars
- 2.3K
- forks
- 354
- watching
- 59
Similar projects
Found by comparing what the projects do, not just their names.
Chinese Language Model
Trains and evaluates a Chinese language model using adversarial training on a large corpus.
Chinese NLP library
Provides pre-trained Chinese language models based on the ELECTRA framework for natural language processing tasks
Language model trainer
This project provides code and model for improving language understanding through generative pre-training using a transformer-based architecture.
BERT model
An implementation of Google's 2018 BERT model in PyTorch, allowing pre-training and fine-tuning for natural language processing tasks
Communications system
An implementation of a machine learning-based communications system using deep learning techniques.
Text Generator
A Python module for creating character-level or word-level neural networks for text generation and training on various datasets
Knowledge analyzer
Analyzing knowledge development and evolution in large language models during training
Model trainer
Enables distributed deep learning with Keras and Spark for scalable model training
Text-to-Image Generator
Implements Google's Text-to-Image Neural Network in PyTorch using a cascading DDPM architecture with dynamic clipping and noise level conditioning.
ML explanations
An explanation of key concepts and advancements in the field of Machine Learning
Language model trainer
Provides a framework for training large-scale language models on GPUs with advanced features and optimizations.
Music Generation Model
Implementation of Google's MusicLM model for music generation using attention networks and text-conditioning.
Deep learning implementations
Implementations of various deep learning algorithms and techniques with accompanying documentation
huggingface/transformers136.4K
Model repository
A collection of pre-trained machine learning models for various natural language and computer vision tasks, enabling developers to fine-tune and deploy these models on their own projects.
Synthetic data generator
Creates synthetic high-resolution spatiotemporal data for renewable energy resources using generative adversarial networks.
