Computing (and experimenting with) sentence embeddings in Python

Three different methods namely TFIDF, word average embedding method and inverse document frequency method were used to build a text matching system. The systems were tested on the first 100 questions which were duplicate. A maximum accuracy score of 77% and 67% in top5 and top 2 matches was obtained using average word model.

State-of-the-Art Ember embedding model for retrieval augmented generation

RecurrentGPT (unofficial)

Sentence Embeddings with BERT & XLNet

Amazon Reviews Sentiment Classifications and Textual Similarities with Universal Sentence Encoder.

In this demo, we illustrate the the possibility of using Semantic Search + Recognising Textual Entailment with Gradio to build an automated fact checking tool

Playground/toolkit para probar distintos modelos de embeddings, hacer búsquedas de similitud entre queries y datasets que subas.

A tool to map FrameNet Lexical Units to BabelNet synsets using the distance between sentence embeddings of corresponding definitions

In this project, I use sentence embeddings to identify question pairs

Implementation of different Sentence Encoders trained on Natural Language Inference and evaluated on unseen tasks through Meta's SentEval framework.

Embedding based semantic search app for poetry [App and EDA notebooks]

A FastAPI application with Redis, Celery, and Flower

Using different technique to deal with Text Classification for Sentimental Analsysi from the a base TF-IDF encoding to pre-trained sentence embeddings.

Using Local & Sagemaker Provisioned sentence transformer in Python

Code Repository for AKRaNLU @ CLEF JOKER 2023: Using Sentence Embeddings and Multilingual Models to Detect and Interpret Wordplay

NLP Project - Sentence Classification - Toxicity- Approx 20,000 comments - ranging from 2 to 30 words. Balanced Data Set. 1. Traditional, pre-2010 NLP and ML techniques used. 2. Dense Word Vectors - w2v & Glove, sentence vector created from averaged word vectors, ANN. 3. Glove combined with bi-LSTMs and 2D Convs.

Streamlit web application for tc*, a tool that can be used to flag non-standard clauses in Terms & Conditions agreements.

City-Data.com Corpus Python Recipes