Skip to content

Latest commit

 

History

History
28 lines (19 loc) · 896 Bytes

README.md

File metadata and controls

28 lines (19 loc) · 896 Bytes

Inception.torch

th

googlenet = dofile('googlenet.lua')
model = googlenet({cudnn.SpatialConvolution, cudnn.SpatialMaxPooling, cudnn.ReLU, 'BDHW', 'cudnn'})

print(model)

This repository contains a reference pre-trained network for the Inception model, complementing the Google publication.

Going Deeper with Convolutions, CVPR 2015. Christian Szegedy, Wei Liu, Yangqing Jia, Pierre Sermanet, Scott Reed, Dragomir Anguelov, Dumitru Erhan, Vincent Vanhoucke, Andrew Rabinovich.

You can view "inception.ipynb" directly on GitHub, or clone the repository, install dependencies listed in the notebook and play with code locally.

You may also be interested in the Multibox approach that uses the Inception architecture for object detection, also available on GitHub.

Disclaimer: this is not an official Google product (experimental or otherwise).