Model Packaging

TensorFlow is an open-source machine learning library that supports multiple programming languages including Python and Javascript. TensorFlow provides various tools for building and bringing deep learning models to production. The library became more popular in its second version(TensorFlow 2.x) because of integrating Keras as its high-level API. Keras is an open-source library that provides a simple to use Python interface for neural networks.

Amazon SageMaker is a cloud machine-learning platform that enables developers to create, train, and deploy machine-learning models in the cloud. It also enables developers to deploy ML models on embedded systems and edge-devices.

MLflow is an open-source platform for managing the machine learning lifecycle, including experimentation, reproducibility, deployment, and a central model registry

Kubeflow is an open-source cloud-native machine learning platform for orchestrating complicated machine learning workflows on containerized environments using Kubernetes.

The Open Neural Network Exchange is an open-source artificial intelligence ecosystem used for representing traditional and modern deep learning models. It allows developers to interchange models between various ML frameworks and tools.

BentoML is an open-source platform for serving, managing, and deploying high-performance machine learning models. It supports all major ML frameworks and works with the most popular DevOps and infrastructure tools.

MMdnn (Model Management for deep neural networks) is an open-source model conversion tool. It helps to inter-operate among different deep learning frameworks, convert, visualize and diagnose deep learning models.

Neural Network Distiller by Intel AI Lab is a Python package for neural network compression research. Distiller provides a PyTorch environment for prototyping and analyzing compression algorithms.

OctoML is an acceleration platform that allows the deployment of machine learning models on any hardware. It’s built on top of the open-source Apache TVM compiler framework project. It supports a wide variety of machine learning frameworks like PyTorch,TensorFlow, and ONNX serialized models as well as hardware backends like NVIDIA/CUDA, x86, AMD, ARM, Intel, MIPS, and more.

Deeplite created Neutrino, an intelligent optimization engine for Deep Neural Networks (DNNs) deployed on cloud servers and edge devices. Neutrino helps to automatically optimize high-performance DNN models to satisfy target computation constraints. The engine inputs large, initial DNN models and constraints to produce smaller, highly efficient models that meet these constraints.

Core ML is an Apple framework to integrate machine learning models into your app. It provides a unified representation for all models. In its substance, Core ML supports Vision for analyzing images, Natural Language for processing text, Speech for converting audio to text, and Sound Analysis for identifying sounds in audio.