Semantic segmentation of aerial imagery into 5 land-cover classes — 0.72 mIoU on LandCover.ai with a pretrained EfficientNet-B0 U-Net and Gradio demo.
-
Updated
Aug 13, 2026 - Jupyter Notebook
Semantic segmentation of aerial imagery into 5 land-cover classes — 0.72 mIoU on LandCover.ai with a pretrained EfficientNet-B0 U-Net and Gradio demo.
Maximum-quality open-source LLMs on molab (Blackwell B6000, 96 GB) — trains Pink Elephant GPT-2 1B from scratch with auto HF checkpoint uploads, plus high-quality Llama 3.1 70B inference.
Dashboard for NHSE RTT monthly data using marimo and molab
Shadow-price-guided SDF level-set compression for frozen driving perception, expressed as an interactive marimo notebook
Add a description, image, and links to the molab topic page so that developers can more easily learn about it.
To associate your repository with the molab topic, visit your repo's landing page and select "manage topics."