caffe将各种原始图片数据集转换为lmdb格式并训练网络

作者: Ericzhang922 | 来源:发表于2017-04-25 15:48 被阅读4434次

1.Caltech-UCSD Birds200 鸟类图像数据

Caltech-UCSD Birds200 是一个鸟类图片数据集，包含 200 不同种鸟类，共计 11788 张图片
此处下载该数据集Caltech-UCSD Birds200 鸟类图像数据

文件夹images内包含200个文件夹，其中每一个文件夹包含一个分类.（由于都是一一对应的关系，所以我们可以直接利用word中表格栏选项中的文本转换为表格的方法，将其暂时转换为表格形式，然后再copy到excel中做进一步处理）

README.txt是该数据集的解释文件
images.txt是该数据集内images内图片的目录文件每一行代表一个子文件夹内的一个图片文件

id	content
1	001.Black_footed_Albatross/Black_Footed_Albatross_0046_18.jpg
2	001.Black_footed_Albatross/Black_Footed_Albatross_0009_34.jpg
3	001.Black_footed_Albatross/Black_Footed_Albatross_0002_55.jpg
4	001.Black_footed_Albatross/Black_Footed_Albatross_0074_59.jpg
5	001.Black_footed_Albatross/Black_Footed_Albatross_0014_89.jpg
6	001.Black_footed_Albatross/Black_Footed_Albatross_0085_92.jpg
7	001.Black_footed_Albatross/Black_Footed_Albatross_0031_100.jpg
....	....

classes.txt是标签文件代表200个种类的200个标签

id	label
1	001.Black_footed_Albatross
2	002.Laysan_Albatross
3	003.Sooty_Albatross
4	004.Groove_billed_Ani
5	005.Crested_Auklet
6	006.Least_Auklet
7	007.Parakeet_Auklet
....	....

image_class_labels.txt对应于images.txt文件表示每一个图片的标签

id	label
1	1
2	1
3	1
4	1
5	1
6	1
7	1
....	....

train_test_split.txt对应于images.txt中分割train和test图片，其中1表示train图片而0表示test图片

id	train/test
1	0
2	1
3	0
4	1
5	1
6	0
7	1
....	....

通过以上的简单转换就可以变成excel表格形式，然后将其全部copy到一个excel文件中，再做简单的处理将重复的id栏去掉，可以得到如下表格：

id	image	lable	train/test
1	001.Black_footed_Albatross/Black_Footed_Albatross_0046_18.jpg	1	0
2	001.Black_footed_Albatross/Black_Footed_Albatross_0009_34.jpg	1	1
3	001.Black_footed_Albatross/Black_Footed_Albatross_0002_55.jpg	1	0
4	001.Black_footed_Albatross/Black_Footed_Albatross_0074_59.jpg	1	1
5	001.Black_footed_Albatross/Black_Footed_Albatross_0014_89.jpg	1	1
6	001.Black_footed_Albatross/Black_Footed_Albatross_0085_92.jpg	1	0
7	001.Black_footed_Albatross/Black_Footed_Albatross_0031_100.jpg	1	1
8	001.Black_footed_Albatross/Black_Footed_Albatross_0051_796103.jpg	1	1
9	001.Black_footed_Albatross/Black_Footed_Albatross_0010_796097.jpg	1	1
10	001.Black_footed_Albatross/Black_Footed_Albatross_0025_796057.jpg	1	0
....	....	....	....

所以通过train/test的选择，我们就可以将其分成训练集和测试集。再将其copy回word文档，就可以产生两个需要用到的文件train.txt和val.txt。这就是参考文档中的filelist文件，所以可以跳过参考文档中的做法。

train.txt
001.Black_footed_Albatross/Black_Footed_Albatross_0009_34.jpg 1
001.Black_footed_Albatross/Black_Footed_Albatross_0074_59.jpg 1
001.Black_footed_Albatross/Black_Footed_Albatross_0014_89.jpg 1
001.Black_footed_Albatross/Black_Footed_Albatross_0031_100.jpg 1
001.Black_footed_Albatross/Black_Footed_Albatross_0051_796103.jpg 1
001.Black_footed_Albatross/Black_Footed_Albatross_0010_796097.jpg 1
001.Black_footed_Albatross/Black_Footed_Albatross_0023_796059.jpg 1
001.Black_footed_Albatross/Black_Footed_Albatross_0040_796066.jpg 1
001.Black_footed_Albatross/Black_Footed_Albatross_0089_796069.jpg 1
001.Black_footed_Albatross/Black_Footed_Albatross_0067_170.jpg 1
001.Black_footed_Albatross/Black_Footed_Albatross_0060_796076.jpg 1
....
val.txt
001.Black_footed_Albatross/Black_Footed_Albatross_0046_18.jpg 1
001.Black_footed_Albatross/Black_Footed_Albatross_0002_55.jpg 1
001.Black_footed_Albatross/Black_Footed_Albatross_0085_92.jpg 1
001.Black_footed_Albatross/Black_Footed_Albatross_0025_796057.jpg 1
001.Black_footed_Albatross/Black_Footed_Albatross_0086_796062.jpg 1
001.Black_footed_Albatross/Black_Footed_Albatross_0049_796063.jpg 1
001.Black_footed_Albatross/Black_Footed_Albatross_0006_796065.jpg 1
001.Black_footed_Albatross/Black_Footed_Albatross_0016_796067.jpg 1
001.Black_footed_Albatross/Black_Footed_Albatross_0065_796068.jpg 1
001.Black_footed_Albatross/Black_Footed_Albatross_0042_796071.jpg 1
001.Black_footed_Albatross/Black_Footed_Albatross_0090_796077.jpg 1
....

接下来我们将图片和用到的文件放到一个文件夹mydata下

train.txt
val.txt
images

在caffe中，作者为我们提供了这样一个文件：convert_imageset.cpp，存放在根目录下的tools文件夹下。编译之后，生成对应的可执行文件放在 $cafferoot/tools/ 下面，这个文件的作用就是用于将图片文件转换成caffe框架中能直接使用的db文件。
error:

./include/caffe/util/cudnn.hpp:8:34: fatal error: caffe/proto/caffe.pb.h: No such file or directory
#include "caffe/proto/caffe.pb.h"

解决方法:fatal error: caffe/proto/caffe.pb.h: No such file or directory

$ protoc src/caffe/proto/caffe.proto --cpp_out=.
$ mkdir include/caffe/proto
$ mv src/caffe/proto/caffe.pb.h include/caffe/proto
$ cp -r include/caffe/proto ./

执行下面的命令，编译convert_imageset.cpp文件，执行不成功，显示proto的错误，此时我们改变方法，利用自带的example/imagenet下的文件来生成所需的文件。由于image的大小不一样，所以最好在开始的时候就将所有图片转换成为同样大小，利用以下命令转换：

find ./ -name '*.jpg' -exec convert -resize 600x480 {} {} \;

在image文件夹的目录内执行此命令后就将所有的图片都转换成为320x240大小的图片了。copy example/imagenet下的文件到mydata文件夹下，修改create_imagenet.sh文件(该文件内也有resize的选项)
在运行create_imagenet.sh时出现E0425 14:24:29.828167 5561 io.cpp:80] Could not open or find file /home/hypervision/work/caffe/mydata/images/val//home/hypervision/work/caffe/mydata/images/val/001.Black_footed_Albatross/Black_Footed_Albatross_0006_796065.jpg 找不到图片的现象是因为txt文件中image和对应的label之间只能有一个空格(只能是英文输入环境下的空格！！！)，这用excel转word的时候会在这里产生中文环境下的空格而产生错误！！！caffe训练自己的模型步骤要改正此错误也很好改：

#用查找和替换方式
.jpg (中文输入环境下的空格)
.jpg (英文输入环境下的空格)

这时运行该文件就可以生成在次数据集上的lmdb格式的数据文件了。create_imagenet.sh文件为：

#!/usr/bin/env sh
# Create the imagenet lmdb inputs
# N.B. set the path to the imagenet train + val data dirs
set -e
#全部的文件(包含数据)都放在/caffe/mydata文件夹内
EXAMPLE=../mydata   #存放输出lmdb文件的文件夹
DATA=../mydata      #存放train.txt和val.txt文件的文件夹
TOOLS=../build/tools #调用convert_imageset程序

TRAIN_DATA_ROOT=/home/hypervision/work/caffe/mydata/images/train/  #train数据存放目录(可包含子文件夹)
VAL_DATA_ROOT=/home/hypervision/work/caffe/mydata/images/val/      #val数据存放目录(可包含子文件夹)


# Set RESIZE=true to resize the images to 256x256. Leave as false if images have
# already been resized using another tool.
RESIZE=false
if $RESIZE; then
  RESIZE_HEIGHT=256
  RESIZE_WIDTH=256
else
  RESIZE_HEIGHT=0
  RESIZE_WIDTH=0
fi

if [ ! -d "$TRAIN_DATA_ROOT" ]; then
  echo "Error: TRAIN_DATA_ROOT is not a path to a directory: $TRAIN_DATA_ROOT"
  echo "Set the TRAIN_DATA_ROOT variable in create_imagenet.sh to the path" \
       "where the ImageNet training data is stored."
  exit 1
fi

if [ ! -d "$VAL_DATA_ROOT" ]; then
  echo "Error: VAL_DATA_ROOT is not a path to a directory: $VAL_DATA_ROOT"
  echo "Set the VAL_DATA_ROOT variable in create_imagenet.sh to the path" \
       "where the ImageNet validation data is stored."
  exit 1
fi

echo "Creating train lmdb..."

GLOG_logtostderr=1 $TOOLS/convert_imageset \
    --resize_height=$RESIZE_HEIGHT \
    --resize_width=$RESIZE_WIDTH \
    --shuffle \
    $TRAIN_DATA_ROOT \
    $DATA/train.txt \
    $EXAMPLE/ilsvrc12_train_lmdb

echo "Creating val lmdb..."

GLOG_logtostderr=1 $TOOLS/convert_imageset \
    --resize_height=$RESIZE_HEIGHT \
    --resize_width=$RESIZE_WIDTH \
    --shuffle \
    $VAL_DATA_ROOT \
    $DATA/val.txt \
    $EXAMPLE/ilsvrc12_val_lmdb

echo "Done."

在当前目录开启terminal，执行create_imagenet.sh:

hypervision@hypervision-700:~/work/caffe/mydata$ ./create_imagenet.sh 
Creating train lmdb...
I0425 14:33:11.755530  5941 convert_imageset.cpp:86] Shuffling data
I0425 14:33:11.995342  5941 convert_imageset.cpp:89] A total of 5994 images.
I0425 14:33:11.995568  5941 db_lmdb.cpp:35] Opened lmdb ../mydata/ilsvrc12_train_lmdb
I0425 14:33:16.814551  5941 convert_imageset.cpp:147] Processed 1000 files.
I0425 14:33:21.675709  5941 convert_imageset.cpp:147] Processed 2000 files.
I0425 14:33:26.373101  5941 convert_imageset.cpp:147] Processed 3000 files.
I0425 14:33:31.185261  5941 convert_imageset.cpp:147] Processed 4000 files.
I0425 14:33:36.626116  5941 convert_imageset.cpp:147] Processed 5000 files.
I0425 14:33:41.590888  5941 convert_imageset.cpp:153] Processed 5994 files.
Creating val lmdb...
I0425 14:33:42.242558  5971 convert_imageset.cpp:86] Shuffling data
I0425 14:33:42.513573  5971 convert_imageset.cpp:89] A total of 5794 images.
I0425 14:33:42.513809  5971 db_lmdb.cpp:35] Opened lmdb ../mydata/ilsvrc12_val_lmdb
I0425 14:33:47.492820  5971 convert_imageset.cpp:147] Processed 1000 files.
I0425 14:33:58.676864  5971 convert_imageset.cpp:147] Processed 2000 files.
I0425 14:34:16.156783  5971 convert_imageset.cpp:147] Processed 3000 files.
I0425 14:34:36.846071  5971 convert_imageset.cpp:147] Processed 4000 files.
I0425 14:34:56.638617  5971 convert_imageset.cpp:147] Processed 5000 files.
I0425 14:35:11.903795  5971 convert_imageset.cpp:153] Processed 5794 files.
Done.

在当前目录可以看到生成了两个文件夹ilsvrc12_train_lmdb和ilsvrc12_val_lmdb分别存放训练和验证所需的数据。
然后利用 make_imagenet_mean.sh 生成所需要的 mean file，和create_imagenet.sh同样的设置：

#!/usr/bin/env sh
# Compute the mean image from the imagenet training lmdb
# N.B. this is available in data/ilsvrc12

EXAMPLE=../mydata
DATA=../mydata
TOOLS=../build/tools

$TOOLS/compute_image_mean $EXAMPLE/ilsvrc12_train_lmdb \
  $DATA/imagenet_mean.binaryproto

echo "Done."

在运行过程中如果出现如下错误：

I0425 14:46:16.075706  6398 db_lmdb.cpp:35] Opened lmdb ../mydata/ilsvrc12_train_lmdb
I0425 14:46:16.076617  6398 compute_image_mean.cpp:70] Starting iteration
F0425 14:46:16.076819  6398 compute_image_mean.cpp:79] Check failed: size_in_datum == data_size (230400 vs. 144720) Incorrect data field size 230400
*** Check failure stack trace: ***
    @     0x7fb05aca25cd  google::LogMessage::Fail()
    @     0x7fb05aca4433  google::LogMessage::SendToLog()
    @     0x7fb05aca215b  google::LogMessage::Flush()
    @     0x7fb05aca4e1e  google::LogMessageFatal::~LogMessageFatal()
    @           0x4025d8  main
    @     0x7fb059c13830  __libc_start_main
    @           0x402bb9  _start
    @              (nil)  (unknown)
Aborted (core dumped)
Done.

检查图片大小在之前的resize过程中是否都设置一样了，如果存在不一样则会出现上诉错误，利用create_imagenet.sh中的resize再次生成一下lmdb文件，并再次运行此mean文件就可以得到imagenet_mean.binaryproto文件：

hypervision@hypervision-700:~/work/caffe/mydata$ sh ./make_imagenet_mean.sh 
I0425 14:51:12.212188  6553 db_lmdb.cpp:35] Opened lmdb ../mydata/ilsvrc12_train_lmdb
I0425 14:51:12.213258  6553 compute_image_mean.cpp:70] Starting iteration
I0425 14:51:13.346879  6553 compute_image_mean.cpp:101] Processed 5994 files.
I0425 14:51:13.347925  6553 compute_image_mean.cpp:108] Write to ../mydata/imagenet_mean.binaryproto
I0425 14:51:13.349079  6553 compute_image_mean.cpp:114] Number of channels: 3
I0425 14:51:13.349189  6553 compute_image_mean.cpp:119] mean_value channel [0]: 110.145
I0425 14:51:13.349315  6553 compute_image_mean.cpp:119] mean_value channel [1]: 127.242
I0425 14:51:13.349419  6553 compute_image_mean.cpp:119] mean_value channel [2]: 123.707
Done.

我们利用caffe官方给出的文本定义网络结构和solver文件来训练一个神经网络，选择/caffe/models/bvlc_alexnet,查看solver.prototxt，可以不用修改该文件：

net: "../models/bvlc_alexnet/train_val.prototxt"
test_iter: 1000
test_interval: 1000
base_lr: 0.01
lr_policy: "step"
gamma: 0.1
stepsize: 100000
display: 20
max_iter: 450000
momentum: 0.9
weight_decay: 0.0005
snapshot: 10000
snapshot_prefix: "../models/bvlc_alexnet/caffe_alexnet_train"
solver_mode: GPU

查看train_val.prototxt文件，我们需要修改的是输入端的各种数据及mean file，如下：

name: "AlexNet"
layer {
  name: "data"
  type: "Data"
  top: "data"
  top: "label"
  include {
    phase: TRAIN
  }
  transform_param {
    mirror: true
    crop_size: 227
    mean_file: "../mydata/imagenet_mean.binaryproto"  #此处需要修改
  }
  data_param {
    source: "../mydata/ilsvrc12_train_lmdb"           #此处需要修改    
    batch_size: 256
    backend: LMDB
  }
}
layer {
  name: "data"
  type: "Data"
  top: "data"
  top: "label"
  include {
    phase: TEST
  }
  transform_param {
    mirror: false
    crop_size: 227
    mean_file: "../mydata/imagenet_mean.binaryproto" #此处需要修改  
  }
  data_param {
    source: "../mydata/ilsvrc12_val_lmdb"            #此处需要修改
    batch_size: 50
    backend: LMDB
  }
}

接着由于我们copy的是imagenet内的训练文件，里面对应的是models/bvlc_reference_caffenet内的文件，而我们需要训练的是bvlc_alexnet内的文件，所以还需要修改/mydata目录下的train_caffenet.sh:

#!/usr/bin/env sh
set -e

../build/tools/caffe train \
    --solver=../models/bvlc_alexnet/solver.prototxt $@

和resume_training.sh：

#!/usr/bin/env sh
set -e

../build/tools/caffe train \
    --solver=../models/bvlc_alexnet/solver.prototxt \
    --snapshot=../models/bvlc_alexnet/caffenet_train_10000.solverstate.h5 \
    $@

好了，准备工作已经全部完成，只需要执行train_caffenet.sh即可。(注意以上各种文件的路径是否加载正确，要以当前目录为准，不要单纯的安装文件的形式去修改，否则会找不到需要加载的各种文件而报错！！！)

训练过程中如果出现Check failed: error == cudaSuccess (2 vs. 0) out of memory的问题证明在train_val.prototxt文件中train和val的batch_size太大了，一次性读入的图片超出了显存，所以适当的修改batch_size的值。
caffe跑试验遇到错误：Check failed: error == cudaSuccess (2 vs. 0) out of memory

caffe将各种原始图片数据集转换为lmdb格式并训练网络

1.Caltech-UCSD Birds200 鸟类图像数据

相关文章

网友评论

延伸阅读

深度阅读

栏目导航

热点阅读

机器学习TF/coffe

CAFFE