0% found this document useful (0 votes)
21 views1 page

Readme

This document outlines the steps to illustrate the function of a convolutional neural network using a simple example of classifying images of O and X into two categories. It trains the network on these images, tests the network performance, and runs a Python script to classify new single images using the trained network.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as TXT, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
21 views1 page

Readme

This document outlines the steps to illustrate the function of a convolutional neural network using a simple example of classifying images of O and X into two categories. It trains the network on these images, tests the network performance, and runs a Python script to classify new single images using the trained network.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as TXT, PDF, TXT or read online on Scribd
You are on page 1/ 1

#

# Illustration of convolutional neural network (CNN) function.


# Two simple image categories: O and X images
#
# Inspired by Brandon Rohrer's 'Data Science and Robots' blog
# http://brohrer.github.io/how_convolutional_neural_networks_work.html
#
# Caffe http://caffe.berkeleyvision.org/ implemenation
#
# AH received a hardware grant from NVIDIA Corporation.
# We gratefully acknowledges the support of NVIDIA Corporation for our research.
# (c) A. Hanuschkin 2016

# 0) check for GPU


nvidia-smi
-> luckily we have a TitanX in this PC ;)

# 1) train the network


~/caffe_FAST/build/tools/caffe train --solver=ANet_3conv_solver.prototxt --gpu 0

# 2) test the network performance


~/caffe_FAST/build/tools/caffe test --model=ANet_3conv.prototxt --
weights=ANet_3conv_1strun_snapshot_iter_5000.caffemodel --gpu 0 --iterations 10

# 3) run a Python script to pass single images to the trained network


python ClassifySingleImagesWithPython.py

You might also like