About this document
Performance Evaluation of Deep Learning Compilers For Edge Inference by uchihakuroshitsuji23 is a document available to read on EtoBox.
This paper evaluates the performance of TensorFlow Lite and TensorFlow-TensorRT inference compilers for deep learning on edge devices, focusing on throughput, latency, and power consumption. The study highlights the challenges of utilizing limited computational resources in edge computing and the need for optimized models for efficient inference. It also discusses the importance of standardized benchmarks for assessing compiler performance in edge environments.
- Author
- uchihakuroshitsuji23
- Language
- EN