Over the past few months we've been building up a library of easy to use computer vision models. We've now given them a home: the Roboflow Model Library.
There, you can access information about each model (we will even be adding pros/cons, and benchmarks of each model soon) and quickly access the code on github, via a downloadable Jupyter notebook, or on Google Colab (which provides free GPUs for training).
So far we have 5 object detection models and 1 classification model. This will grow over time. Our goal is to have easy to use, up to date models right at your fingertips.
Importantly, these notebooks all work seamlessly with Roboflow. So if you've uploaded your images and annotations, you can seamlessly export them for use with any of these models in a single click. That means you can quickly experiment across a wide variety of models and frameworks to get solid baselines (or ablation tests) and find a good starting point for tuning.
Roboflow accelerates your computer vision workflow through automated annotation
quality assurance, universal annotation format conversion (like
PASCAL VOC XML to COCO JSON and
), team sharing and versioning, and easy integration with popular
open source computer vision models.
Getting started with your first 1GB of data is completely free.