Deeplite Neutrino: An End-to-End Framework for Constrained Deep Learning Model Optimization

Abstract

Designing deep learning-based solutions is becoming a race for training deeper models with a greater number of layers. While a large-size deeper model could provide competitive accuracy, it creates a lot of logistical challenges and unreasonable resource requirements during development and deployment. This has been one of the key reasons for deep learning models not being excessively used in various production environments, especially in edge devices. There is an immediate requirement for optimizing and compressing these deep learning models, to enable on-device intelligence. In this research, we introduce a black-box framework, Deeplite Neutrino for production-ready optimization of deep learning models. The framework provides an easy mechanism for the end-users to provide constraints such as a tolerable drop in accuracy or target size of the optimized models, to guide the whole optimization process. The framework is easy to include in an existing production pipeline and is available as a Python Package, supporting PyTorch and Tensorflow libraries. The optimization performance of the framework is shown across multiple benchmark datasets and popular deep learning models. Further, the framework is currently used in production and the results and testimonials from several clients are summarized.

arXiv comment"IAAI Deployed Application Award", IAAI 2021 @ AAAI 2021

References in corpus (16)

• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• Language Models are Few-Shot Learners
• Exploiting Linear Structure Within Convolutional Networks for Efficient Evaluation
• The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks
• AMC: AutoML for Model Compression and Acceleration on Mobile Devices
• Rethinking the Value of Network Pruning
• A Survey of Model Compression and Acceleration for Deep Neural Networks
• Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism
• Speeding up Convolutional Neural Networks with Low Rank Expansions
• Horovod: fast and easy distributed deep learning in TensorFlow
• Speeding-up Convolutional Neural Networks Using Fine-tuned CP-Decomposition
• Towards Understanding Knowledge Distillation
• Pruning Algorithms to Accelerate Convolutional Neural Networks for Edge Applications: A Survey
• Few Shot Network Compression via Cross Distillation
• Knowledge Squeezed Adversarial Network Compression
• DARB: A Density-Aware Regular-Block Pruning for Deep Neural Networks