Training Techniques | RMSProp, Weight Decay |
---|---|
Architecture | 1x1 Convolution, Batch Normalization, Convolution, Depthwise Separable Convolution, Dropout, Global Average Pooling, Inverted Residual Block, Residual Connection, ReLU, Max Pooling, Softmax |
ID | mnasnet_100 |
SHOW MORE |
Architecture | 1x1 Convolution, Batch Normalization, Convolution, Depthwise Separable Convolution, Dropout, Global Average Pooling, Inverted Residual Block, Residual Connection, ReLU, Max Pooling, Softmax, Squeeze-and-Excitation Block |
---|---|
ID | semnasnet_100 |
SHOW MORE |
MnasNet is a type of convolutional neural network optimized for mobile devices that is discovered through mobile neural architecture search, which explicitly incorporates model latency into the main objective so that the search can identify a model that achieves a good trade-off between accuracy and latency. The main building block is an inverted residual block (from MobileNetV2).
To load a pretrained model:
import timm
m = timm.create_model('mnasnet_100', pretrained=True)
m.eval()
Replace the model name with the variant you want to use, e.g. mnasnet_100
. You can find the IDs in the model summaries at the top of this page.
You can follow the timm recipe scripts for training a new model afresh.
@misc{tan2019mnasnet,
title={MnasNet: Platform-Aware Neural Architecture Search for Mobile},
author={Mingxing Tan and Bo Chen and Ruoming Pang and Vijay Vasudevan and Mark Sandler and Andrew Howard and Quoc V. Le},
year={2019},
eprint={1807.11626},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
BENCHMARK | MODEL | METRIC NAME | METRIC VALUE | GLOBAL RANK |
---|---|---|---|---|
ImageNet | semnasnet_100 | Top 1 Accuracy | 75.45% | # 223 |
Top 5 Accuracy | 92.61% | # 223 | ||
ImageNet | mnasnet_100 | Top 1 Accuracy | 74.67% | # 235 |
Top 5 Accuracy | 92.1% | # 235 |