Vis4SciML/Landscaper
Landscaper is a comprehensive Python framework designed for exploring the loss landscapes of deep learning models.
This tool helps machine learning engineers and researchers understand how their deep learning models learn and behave. You input a trained deep learning model, and it produces detailed visualizations and quantifiable metrics about its 'loss landscape'. This allows you to go beyond simple accuracy scores and gain deeper insights into model stability and performance.
Available on PyPI.
Use this if you need to deeply analyze the training dynamics and generalization capabilities of your deep learning models.
Not ideal if you only need basic performance metrics for your deep learning models, such as accuracy or F1 score, without needing to explore their internal learning mechanisms.
Stars
41
Forks
—
Language
Python
License
—
Category
Last pushed
Jan 27, 2026
Commits (30d)
0
Dependencies
8
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/Vis4SciML/Landscaper"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
DPIRD-DMA/OmniWaterMask
Python library for high-accuracy water segmentation in satellite and aerial imagery, combining...
prs-eth/Popcorn
[RSE 2024] 🍿POPCORN: High-resolution Population Maps Derived from Sentinel-1 and Sentinel-2 🌍🛰️
xinluo2018/WatNetv2
A new WatNet version which further extends the data applicability from Sentinel-2 image to...
sei-latam/WETSAT_v2
Wetlands flooding extent and trends using SATellite data and Machine Learning v2.0
sidgan/ETCI-2021-Competition-on-Flood-Detection
Experiments on Flood Segmentation on Sentinel-1 SAR Imagery with Cyclical Pseudo Labeling and...