生成lmdb文件和均值文件

一、首先准备好训练数据和测试数据，每一类图片放到一个文件夹中

生成lmdb文件和均值文件

二、生成标签文件：这个代码对应的是七分类

"<span style=""font-family:Arial;font-size:18px;"">"
"<span style=""font-size:18px;"">"
"<span style=""font-size:18px;"">"
import os

def IsSubString(SubStrList,Str):
flag=True
for substr in SubStrList:
if not(substr in Str):
flag=False

return flag
def GetFileList(FindPath,FlagStr=[]):
FileList=[]
FileNames=os.listdir(FindPath)
if len(FileNames)>0:
for fn in FileNames:
if len(FlagStr)>0:
if IsSubString(FlagStr,fn):
fullfilename=os.path.join(FindPath,fn)
FileList.append(fullfilename)
else:
fullfilename=os.path.join(FindPath,fn)
FileList.append(fullfilename)

if len(FileList)>0:
FileList.sort()

return FileList

train_txt=open('train.txt','w')

imgfile=GetFileList('/home/ths/6_operation_32_data/database/train/train_bilinear/')
for img in imgfile:
str1=img+' '+'0'+'\n'
train_txt.writelines(str1)

imgfile=GetFileList('/home/ths/6_operation_32_data/database/train/train_constract/')
for img in imgfile:
str2=img+' '+'1'+'\n'

train_txt.writelines(str2)

imgfile=GetFileList('/home/ths/6_operation_32_data/database/trian/train_gaussion/')
for img in imgfile:
str3=img+' '+'2'+'\n'
train_txt.writelines(str3)

imgfile=GetFileList('/home/ths/6_operation_32_data/database/train/train_jpeg/')
for img in imgfile:
str4=img+' '+'3'+'\n'
train_txt.writelines(str4)

imgfile=GetFileList('/home/ths/6_operation_32_data/database/train/train_mean/')
for img in imgfile:
str5=img+' '+'4'+'\n'
train_txt.writelines(str5)

imgfile=GetFileList('/home/ths/6_operation_32_data/database/train/train_med/')
for img in imgfile:
str6=img+' '+'5'+'\n'
train_txt.writelines(str6)

imgfile=GetFileList('/home/ths/6_operation_32_data/database/train/train_org/')
for img in imgfile:
str7=img+' '+'6'+'\n'
train_txt.writelines(str7)
train_txt.close()

test_txt=open('val.txt','w')

imgfile=GetFileList('/home/ths/6_operation_32_data/database/val/val_bilinear/')
for img in imgfile:
str8=img+' '+'0'+'\n'
test_txt.writelines(str8)

imgfile=GetFileList('/home/ths/6_operation_32_data/database/val/val_constract/')
for img in imgfile:
str9=img+' '+'1'+'\n'
test_txt.writelines(str9)

imgfile=GetFileList('/home/ths/6_operation_32_data/database/val/val_gaussion/')
for img in imgfile:
str10=img+' '+'2'+'\n'
test_txt.writelines(str10)

imgfile=GetFileList('/home/ths/6_operation_32_data/database/train/train_jpeg/')
for img in imgfile:
str11=img+' '+'3'+'\n'
test_txt.writelines(str11)

imgfile=GetFileList('/home/ths/6_operation_32_data/database/val/val_mean/')
for img in imgfile:
str12=img+' '+'4'+'\n'
test_txt.writelines(str12)

imgfile=GetFileList('/home/ths/6_operation_32_data/database/val/val_med/')
for img in imgfile:
str13=img+' '+'5'+'\n'
test_txt.writelines(str13)

imgfile=GetFileList('/home/ths/6_operation_32_data/database/val/val_org/')
for img in imgfile:
str14=img+' '+'6'+'\n'
test_txt.writelines(str14)
test_txt.close()

print("success")

生成lmdb文件和均值文件

生成的标签文件前面有路径，把路径用查找替换功能删掉，只保留上图中的信息即可。因为生成lmdb文件的时候可以自动读出路径信息。

三、生成lmdb文件

在caffe-master下新建My_Files文件夹，将caffe-master/examples/imagenet文件夹下的create_imagenet.sh复制过去

要是用服务器运行这个代码的时候记得删掉所有汉字哦

#!/usr/bin/env sh
# Create the imagenet lmdb inputs
# N.B. set the path to the imagenet train + val data dirs

EXAMPLE=/home/ths/caffe-master/My_Files #生成模型训练数据文件，即create_imagenet.sh文件的位置
DATA=/home/ths/6_operation_32_data/database #python脚本处理数据路径，即生成的文件列表.txt文件所在的文件夹
TOOLS=/home/ths/caffe-master/.build_release/tools #caffe所在的文件夹

TRAIN_DATA_ROOT=/home/ths/6_operation_32_data/database/train/ #待处理的训练数据所在的文件夹
VAL_DATA_ROOT=/home/ths/6_operation_32_data/database/val/ #待处理的训练数据所在的文件夹

# 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."

生成lmdb文件和均值文件

四、生成均值文件

用服务器跑要删掉汉字哦！！！

将caffe-master/examples/imagenet文件夹下的make_imagenet_mean.sh到My_Files文件夹

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

EXAMPLE=/home/ths/caffe-master/My_Files #lmdb文件所在的文件夹
DATA=/home/ths/6_operation_32_data/database #生成的imagenet_mean.binaryproto的存放路径
TOOLS=/home/ths/caffe-master/.build_release/tools #caffe所在的文件夹

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

echo "Done."

生成lmdb文件和均值文件