reza-zereh / modelfactory

Where Computer Vision Models Are Made

ModelFactory

ModelFactory is an open-source library for automating computer vision model development from end-to-end. With ModelFactory, you can train highly accurate computer vision models with just a few lines of code.

ModelFactory handles all the tedious CV/ML tasks - data loading/preprocessing, augmentation, neural architecture search, hyperparameter tuning, and model evaluation/selection - enabling you to go straight from images to deployable CV models.

Key Features:

• Simple API for training CV models like classifiers, detectors, segmenters
• Automated data normalization, cleaning, augmentation
• Neural architecture search for optimal model topology
• Robust hyperparameter tuning for computer vision
• Clear model diagnostics and evaluation metrics
• Smooth model export for immediate deployment
• Modular codebase designed for customization

ModelFactory lets you skip past the drudgery and focus on innovating new ways to extract insights from images and video. Spend less time on ML ops, and more time focused on your vision tasks and data.

Whether you're an experienced CV engineer looking to accelerate development, or new to ML and want an easy starting point, ModelFactory provides the tools you need to go from images to production-ready models fast.

The only limit is your data. Get started building better computer vision models faster with ModelFactory!