Coderx7

Here's my experience of installing the NVIDIA CUDA kit 8.0 on a fresh install of Ubuntu Desktop 16.04.3 LTS.

Table of Contents

• Install NVIDIA Graphics Driver via apt-get
• Install CUDA
• Install cuDNN

Table of contents generated with markdown-toc

1. Install NVIDIA Graphics Driver via apt-get

Coderx7

In this article, I will share some of my experience on installing NVIDIA driver and CUDA on Linux OS. Here I mainly use Ubuntu as example. Comments for CentOS/Fedora are also provided as much as I can.

Table of Contents

• Install NVIDIA Graphics Driver via apt-get
• Install NVIDIA Graphics Driver via runfile
  • Remove Previous Installations (Important)
  • Download the Driver
  • Install Dependencies
  • Creat Blacklist for Nouveau Driver

• Stop lightdm/gdm/kdm

Coderx7

#برای نمایش بلادرنگ نمودار ترینینگ و تست ما 
import numpy as np
from matplotlib import pyplot as plt

class LivePlotNotebook(object):
    """
    Live plot using %matplotlib notebook in jupyter notebook
    original url : https://gist.github.com/wassname/04e77eb821447705b399e8e7a6d082ce
    """

Coderx7

# Dinesh Jayaraman

# Based on code by
# Authors: Fabian Pedregosa <[email protected]>
#          Olivier Grisel <[email protected]>
#          Mathieu Blondel <[email protected]>
#          Gael Varoquaux
# License: BSD 3 clause (C) INRIA 2011

print(__doc__)

Coderx7

import argparse
import time

start_time=time.time();
################## Argument Parsing #####################################
parser=argparse.ArgumentParser();
parser.add_argument('-s','--solver', default='', type=str); # if empty, solver is created, else read
parser.add_argument('-res', '--resume_from', default='', type=str); #if not empty, resumes training from given file
parser.add_argument('-ft', '--finetune_from', default='', type=str);

Coderx7

#in the name of God the most compassionate the most merciful
# added mean subtraction so that, the accuracy can be reported accurately just like caffe when doing a mean subtraction
# Seyyed Hossein Hasan Pour
# [email protected]
# 7/3/2016 
# Added Recall/Precision/F1-Score as well 
# 01/03/2017
# added classification using caffe.classifier, which lets you do oversampling (allows for multiple preprocessing options and crops) + the manual center crop
# method. its sloppy at the moment. for some reason, caffe.classifier's predict method, provides less accuracy compared to the caffes test accuracy
# the manuall center crop (i.e. the one which uses Caffe.Net, works just well and achieves the same accuracy as Caffe's)

Coderx7

#in the name of God the most compassionate the most merciful
# added mean subtraction so that, the accuracy can be reported accurately just like caffe when doing a mean subtraction
# Seyyed Hossein Hasan Pour
# [email protected]
# 7/3/2016 
# Added Recall/Precision/F1-Score as well 
# 01/03/2017
# Added batch processing, not what used to take a minute or so, takes only several seconds!
# 07/25/2017
#info: 

Coderx7

# Seyyed Hossein Hasan Pour
# [email protected]
# 7/3/2016 
# Added Recall/Precision/F1-Score as well 
# 01/03/2017
#info: 
#if on windows, one can use these command in a batch file and ease him/her self 
#REM Calculating Confusing Matrix 
#python confusionMatrix_convnet_test.py --proto cifar10_deploy.prototxt --model cifar10_.caffemodel --mean mean.binaryproto --lmdb cifar10_test_lmdb  
#pause

Coderx7

#In the name of God
#Use this script to convert a binaryproto mean file to an equivalent python npy mean file.
#simply use the script like this : 
#python convert_protomean.py mean.binaryproto mean.npy"
import caffe
import numpy as np
import sys

if len(sys.argv) != 3:
    print "Usage: python convert_protomean.py mean.binaryproto mean.npy"

Coderx7

name: "wrn_128_v8"
layer {
  name: "Data1"
  type: "Data"
  top: "Data1"
  top: "Data2"
  include {
    phase: TRAIN
  }
  transform_param {