Lobe is an easy-to-use visual tool that lets you build custom deep learning models, quickly train them, and ship them directly in your app without writing any code. Start by dragging in a folder of training examples from your desktop. Lobe automatically builds you a custom deep learning model and begins training. When you’re done, you can export a trained model and ship it directly in your app.

Connect together smart building blocks called lobes to quickly create custom deep learning models. For example, connect the Hand & Face lobe to find the most prominent hand in the image. Then connect the Detect Features lobe to find important features in the hand. Finally connect the Generate Labels lobe to predict what emoji is in the image. Refine your model by adjusting each lobes unique settings or go under the hood and edit any lobe’s sub-layers.

Your entire dataset is displayed visually so you can easily browse and sort through all your examples. Select any icon and see how that example is performing in your model. Your dataset is also automatically split into a Lesson, used to teach your model during training, and a Test used to evaluate how your model performs in the real world on new examples it has never seen before.

Super fast cloud training gives you real-time results without bogging down your computer. Interactive charts let you monitor your model’s accuracy and understand how it is improving over time. The best accuracy is automatically selected and saved so you don’t need to worry about overfitting.

Built on top of the deep learning frameworks TensorFlow and Keras, Lobe lets you go under the hood and control every layer of your model. Tune hyperparameters, add layers, and design completely new architectures using hundreds of advanced building block lobes. Watch your model take shape as you visually edit your graph and see your changes take effect right away.