EncoDeep
| Autor: | Farinaz Koushanfar, Mohammad Samragh, Mojan Javaheripi |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
010302 applied physics
Artificial neural network Computer science 02 engineering and technology Python (programming language) computer.software_genre 01 natural sciences 020202 computer hardware & architecture Computer architecture Hardware and Architecture Encoding (memory) 0103 physical sciences 0202 electrical engineering electronic engineering information engineering Compiler Field-programmable gate array Throughput (business) computer Software Dram MNIST database computer.programming_language |
| Zdroj: | ACM Transactions on Embedded Computing Systems. 19:1-29 |
| ISSN: | 1558-3465 1539-9087 |
| DOI: | 10.1145/3391901 |
| Popis: | This article proposes EncoDeep, an end-to-end framework that facilitates encoding, bitwidth customization, fine-tuning, and implementation of neural networks on FPGA platforms. EncoDeep incorporates nonlinear encoding to the computation flow of neural networks to save memory. The encoded features demand significantly lower storage compared to the raw full-precision activation values; therefore, the execution flow of EncoDeep hardware engine is completely performed within the FPGA using on-chip streaming buffers with no access to the off-chip DRAM. We further propose a fully automated optimization algorithm that determines the flexible encoding bitwidths across network layers. EncoDeep full-stack framework comprises a compiler that takes a high-level Python description of an arbitrary neural network. The compiler then instantiates the corresponding elements from EncoDeep Hardware library for FPGA implementation. Our evaluations on MNIST, SVHN, and CIFAR-10 datasets demonstrate an average of 4.65× throughput improvement compared to stand-alone weight encoding. We further compare EncoDeep with six FPGA accelerators on ImageNet, showing an average of 3.6× and 2.54× improvement in throughput and performance-per-watt, respectively. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|