daniel-furman Goto Github PK

followers: 41.0 following: 36.0 repos: 50.0 gists: 0.0

Name: Daniel Furman

Type: User

Bio: I like interesting datasets and large neural networks.

Twitter: dryanfurman

Location: San Francisco

╔═══════════════ Hi 👋 I'm Daniel ═══════════════╗ Daniel Furman                                              
║ I'm a Data Scientist living in San Francisco.  ║ ┣━━ 🦾 Featured Code                                       
║ In my free time, I enjoy surfing, golfing, and ║ ┃   ┣━━ Are foundation LMs polyglots? (EMNLP 2023)         
║ spending time with friends.                    ║ ┃   ┣━━ Demos for sft of instruction-tuned LLMs            
╚════════════════════════════════════════════════╝ ┃   ┣━━ Lightweight backend server via flask, mysql, and rq
                                                   ┃   ┣━━ Text clustering: HDBSCAN is all you need           
                                                   ┃   ┣━━ Bayesian MCMC parameter space sampling class       
                                                   ┃   ┗━━ Computational and numerical mathematics algorithms 
                                                   ┗━━ 👨‍💻 Featured Articles                                 
                                                       ┣━━ Why Startup fundamentals are key to AI strategy    
                                                       ┣━━ Default HF models are probably all you need        
                                                       ┗━━ Species distribution modeling tutorial in Python

Daniel Furman's Projects

.github

algorithm-archive

A collaborative book on algorithms

audio-classification-lesson

An audio deep learning mini-lesson from the 2021 DAT/Artathon fellowship.

awesome-chatgpt-prompts-clustering

Text clustering: HDBSCAN is probably all you need.

birdclef21_stft_transformer

Code for STFT Transformer used in BirdCLEF 2021 competition.

blockcv

The blockCV package creates spatially or environmentally separated training and testing folds for cross-validation to provide a robust error estimation in spatially structured environments. See