A fast and flexible keyboard launcher

A project focused on building a multilabel classification model to detect hate speech on social media using BERT and ALBERT pretrained models. It includes data augmentation techniques and hyperparameter tuning for optimal performance in identifying abusive content and hate speech.

Social media platforms like Twitter are key sources of information during disasters, but sorting relevant tweets is challenging. In this project, I used machine learning to classify tweets as disaster-related or not, applying models like RNNs, BiLSTMs, GRUs, and ALBERT on a Kaggle dataset labeled for disaster content.

This repo includes all of the files pertaining to preparing, training/fine-tuning, using and evaluating a transformer architecture, Steam game-review classification model, based on the ALBERT-V2 model.

NLP (Natural Language Processing)

1. Use BERT, ALBERT and GPT2 as tensorflow2.0's layer. 2. Implement GCN, GAN, GIN and GraphSAGE based on message passing.

Tencent Pre-training framework in PyTorch & Pre-trained Model Zoo

A ULauncher/Albert extension that supports currency, units and date time conversion, as well as a calculator that supports complex numbers and functions.

Deployment a light and full OpenAI API for production with vLLM to support /v1/embeddings with all embeddings models.

中文命名实体识别

🧠💬 Articles I wrote about machine learning, archived from MachineCurve.com.

中文长文本分类、短句子分类、多标签分类、两句子相似度（Chinese Text Classification of Keras NLP, multi-label classify, or sentence classify, long or short），字词句向量嵌入层（embeddings）和网络层（graph）构建基类，FastText，TextCNN，CharCNN，TextRNN, RCNN, DCNN, DPCNN, VDCNN, CRNN, Bert, Xlnet, Albert, Attention, DeepMoji, HAN, 胶囊网络-CapsuleNet, Transformer-encode, Seq2seq, SWEM, LEAM, TextGCN

Simple State-of-the-Art BERT-Based Sentence Classification with Keras / TensorFlow 2. Built with HuggingFace's Transformers.

中文语言理解测评基准 Chinese Language Understanding Evaluation Benchmark: datasets, baselines, pre-trained models, corpus and leaderboard

Open Source Pre-training Model Framework in PyTorch & Pre-trained Model Zoo

This project aims to build a spam filter using a fine-tuned ALBERT (A Lite BERT) Transformer model. The ALBERT model, pre-trained on a large corpus of text, is fine-tuned on a spam detection dataset to create an efficient and accurate spam filter.

Comparative analysis of Bert Variants on MELD, Go_Emotions benchmark datasets.

🤖 A PyTorch library of curated Transformer models and their composable components

This repository contains the code used for training/fine-tuning the models introduced in the paper "MiniALBERT: Model Distillation via Parameter-Efficient Recursive Transformers".