johnypark / MnasNet

A TensorFlow 2.0 implementation of MnasNet: Platform-Aware Neural Architecture Search for Mobile.

MnasNet Tensorflow 2 Implementation

Mingxing Tan, Bo Chen, Ruoming Pang, Vijay Vasudevan, Mark Sandler, Andrew Howard, Quoc V. Le. MnasNet: Platform-Aware Neural Architecture Search for Mobile. CVPR 2019. Arxiv link: https://arxiv.org/abs/1807.11626

Usage

Available implementations: a1, b1, small, d1, d1_320

from MnasNet_models import Build_MnasNet

# Standard model
model = Build_MnasNet('a1')


# Change default parameters:
model = Build_MnasNet('a1', dict(input_shape=(128, 128, 3), dropout_rate=0.5))

Pretrained models

Model	Dataset	Input Size	Depth Multiplier	Top-1 Accuracy	Top-5 Accuracy	Pixel 1 latency (ms)	DownLoad Link
MnasNet-A1	ImageNet	224*224	1.0	75.2	95.2	78ms	Google Drive

Reference

MnasNet - Official implementation for Cloud TPU