Stories by isusmelj

Show HN: LightlyTrain – Pretrain YOLO/ResNet on unlabeled data beats ImageNet

Nvidia B200 vs. H100: Independent Training Performance Tests

Clip and Friends: How Vision-Language Models Evolved

Albumentations vs. PIL: 2x Speedup for Model Training Pipelines

Show HN: RustyNum – Fast Rust-Powered NumPy Alternative

Image Retrieval using your fine-tuned DINO model

Data Curation Demystified for Stable Video Diffusion

Show HN: Lightly Insights – open-source dataset analysis

Show HN: Lightly – A Python library for self-supervised learning on images

Show HN: Labelformat now supports all major vision labeling formats

Show HN: Labelformat – Python package for converting computer vision labels

Counting “Generative AI” Appearances on Google Cloud Next Page

Apple M1 and M2 Performance for Training SSL Models

Improve Your Large Language Models (LLMs) with Active Learning

Boosting YOLOv8 Accuracy and Saving 77% in Labeling Costs with Active Learning

A Python library for self-supervised learning on images

Self-supervised learning trends and what to expect in 2023

Guide for active learning in computer vision

Can we paint using ChatGPT?

1