DLPU (Deep Learning Processing Unit) - General name for the hardware that accelerate deep learning tasks.
Model - A neural network that has been trained to perform a specific task.
Model Zoo - A collection of pre-trained models.
Inference - The process of using a trained model to make predictions on new data.
Quantization - Process of converting weights of a model from float format to integer.
Backbone - The part of the model that is responsible for extracting features from the input image. Different task can use the same network architecture as backbone (e.g. classification and object detection).
Head - In the context of a Neural Network, the head is the part of the model that is responsible for the specific task. For example, in an object detection model, the head is responsible for predicting the bounding boxes and the class of the objects in the image.
Larod - It is the name of our Machine Learning API. It is used to run deep learning models in the TensorFlow Lite format.