laura-ham Goto Github PK

active-qa

bert

TensorFlow code and pre-trained models for BERT

chainlit-weaviate-example

chainreaction

chaos-game

Playing around with Chaos Game in JS

classical-piano-composer

contextionary

The contextionary which allows for semantic context-based searches in weaviate

contributor_covenant

Pledge your respect and appreciation for contributors of all kinds to your open source project.

darkflow

Translate darknet to tensorflow. Load trained weights, retrain/fine-tune using tensorflow, export constant graph def to mobile devices

demo-datasets

Weaviate Demo Docker Compose files

face_recognition_with_siamese_network

A face recognition model is trained by using Transfer Learning and applying Siamese Network and Triplet Loss principles.

facial-similarity-with-siamese-networks-in-pytorch

Implementing Siamese networks with a contrastive loss for similarity learning

haystack

:mag: Transformers at scale for question answering & neural search. Using NLP via a modular Retriever-Reader-Pipeline. Supporting DPR, Elasticsearch, HuggingFace's Modelhub...

hm-fashion-image-neural-search

H&M Fashion Image similarity search with Weaviate and DocArray

intex

INteractive Topic Explorer

laura-ham.github.io

lifestyle-informatics-lab

The design of an ambient agent supporting humans in intensive visual tasks.

ln2sql

A tool to query a database in natural language

pepper-emotions

Emotion recognition, generation and expression with a Pepper robot

quizbutton

recepten-website

Recipe website from Laura Ham

seenonme

semi-website

SeMI Website

swagger-codegen-generators

t2v-spacy-models

text-mining

Project files for the Text Mining course 2017 at VU Amsterdam. Song lyrics of 380.000+ songs has been scraped and used for sentiment analysis and geolocation visualisation. Natural Language Processing tools such as tokenization, Named Entity Recognition and sentiment analysis has been used.

tilingproject

For the course 'Heuristics' at the VU Amsterdam we designed three algorithms with different heuristics, in order to determine a better constructive algorithm to solve the tiling problem. The tiling problem consists of a given set of fields, with each their own specific set of tiles, where each field needs to be solved by placing each tile on the field in such way that the whole field is covered and there is no overlap amongst the tiles.

tutorials

Code for all my tutorials

weaviate

Decentralised Knowledge Graph

weaviate-examples

Weaviate vector search engine – examples