LMDB是Cafffe中应用的一种数据库，我们常常需要对LMDB进行读写操作，本文介绍如何采用bash脚本进行LMDB的数据源制作操作。

脚本例子

#!/usr/bin/env sh
# Create the face_48 lmdb inputs
# N.B. set the path to the face_48 train + val data dirs

EXAMPLE=/home/tyd/下载/face_detect  # 数据集存放的路经
DATA=/home/tyd/下载/face_detect
TOOLS=/home/tyd/caffe/build/tools  # caffe安装的路经

TRAIN_DATA_ROOT=/home/tyd/下载/face_detect/train/  # 训练数据集存放路经
VAL_DATA_ROOT=/home/tyd/下载/face_detect/val/   # 测试集存放的路经

# Set RESIZE=true to resize the images to 60 x 60. Leave as false if images have
# already been resized using another tool.
RESIZE=true
if $RESIZE; then  # 对数据resize操作
  RESIZE_HEIGHT=227
  RESIZE_WIDTH=227
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_face_48.sh to the path" \
       "where the face_48 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_face_48.sh to the path" \
       "where the face_48 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 \  # 训练集的txt文件, 存放图片目录和图片类别
    $EXAMPLE/face_train_lmdb  # 训练集lmdb存放位置

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 \ # 验证集的txt文件, 存放图片目录和图片类别
    $EXAMPLE/face_val_lmdb  # 验证集lmdb存放位置

echo "Done."
Status API Training Shop Blog About

根据上面的脚本示例,按照脚本中的中文注释的地方修改自己实际的路经位置.然后运行脚本:sh **.sh 文件即可.

python 生成LMDB

网络上也有用python制作LMDB数据源的,可以参考:https://zhuanlan.zhihu.com/p/23485774