faster rcnn 训练pascal_voc数据集

cd py_faster_rcnn/data

wget http://host.robots.ox.ac.uk/pascal/VOC/voc2007/VOCtrainval_06-Nov-2007.tar wget http://host.robots.ox.ac.uk/pascal/VOC/voc2007/VOCtest_06-Nov-2007.tar wget http://host.robots.ox.ac.uk/pascal/VOC/voc2007/VOCdevkit_08-Jun-2007.tar[/code] 
tar xvf VOCtrainval_06-Nov-2007.tar
tar xvf VOCtest_06-Nov-2007.tar
tar xvf VOCdevkit_08-Jun-2007.tar

//建立软连接
ln -s $VOCdevkit VOCdevkit2007

//下载Imagenet预训练的模型，
cd $FRCN_ROOT
./data/scripts/fetch_imagenet_models.sh

//开始训练，数据集使用pascal_voc
cd $FRCN_ROOT
./experiments/scripts/faster_rcnn_end2end.sh [GPU_ID] [NET] [--set ...]

//例如：
./experiments/scripts/faster_rcnn_end2end.sh 0 VGG_CNN_M_1024 pascal_voc \
  --set EXP_DIR foobar RNG_SEED 42 TRAIN.SCALES "[400, 500, 600, 700]"

#./experiments/scripts/faster_rcnn_end2end.sh
#文件内容，主要是三个方面：
#1、选取数据集格式
#2、调用训练脚本
#3、调用测试脚本、
 #!/bin/bash

set -x
set -e

export PYTHONUNBUFFERED="True"

GPU_ID=$1
NET=$2
NET_lc=${NET,,}
DATASET=$3

array=( $@ )
len=${#array[@]}
EXTRA_ARGS=${array[@]:3:$len}
EXTRA_ARGS_SLUG=${EXTRA_ARGS// /_}

#选取数据集格式，并没有包含数据集地址，只是采用定义好的数据集文件夹框架
#数据集用voc或者coco
case $DATASET in
  pascal_voc)
    TRAIN_IMDB="voc_2007_trainval"
    TEST_IMDB="voc_2007_test"
    PT_DIR="pascal_voc"
    ITERS=70000   #迭代次数

    ;;
  coco)
    # This is a very long and slow training schedule
    # You can probably use fewer iterations and reduce the
    # time to the LR drop (set in the solver to 350,000 iterations).
    TRAIN_IMDB="coco_2014_train"
    TEST_IMDB="coco_2014_minival"
    PT_DIR="coco"
    ITERS=490000
    ;;
  *)
    echo "No dataset given"
    exit
    ;;
esac

LOG="experiments/logs/faster_rcnn_end2end_${NET}_${EXTRA_ARGS_SLUG}.txt.`date +'%Y-%m-%d_%H-%M-%S'`"
exec &> >(tee -a "$LOG")
echo Logging output to "$LOG"

#调用训练脚本
time ./tools/train_net.py --gpu ${GPU_ID} \
  --solver models/${PT_DIR}/${NET}/faster_rcnn_end2end/solver.prototxt \
#其中PT_DIR = pascal_voc, Net = ZF

  --weights data/imagenet_models/${NET}.v2.caffemodel \
#预训练模型

  --imdb ${TRAIN_IMDB} \
  --iters ${ITERS} \
  --cfg experiments/cfgs/faster_rcnn_end2end.yml \
  ${EXTRA_ARGS}
set +x
NET_FINAL=`grep -B 1 "done solving" ${LOG} | grep "Wrote snapshot" | awk '{print $4}'`
set -x

#训练完成，调用测试脚本
time ./tools/test_net.py --gpu ${GPU_ID} \
  --def models/${PT_DIR}/${NET}/faster_rcnn_end2end/test.prototxt \
  --net ${NET_FINAL} \
  --imdb ${TEST_IMDB} \
  --cfg experiments/cfgs/faster_rcnn_end2end.yml \
  ${EXTRA_ARGS}

#训练脚本分析
train_net: "models/pascal_voc/ZF/faster_rcnn_end2end/train.prototxt"
#http://ethereon.github.io/netscope/#/editor 网页的网络可视化工具

#其中存在两组训练参数，一个是 RPN 一个是RCNN训练参数
base_lr: 0.001
lr_policy: "step"
gamma: 0.1
stepsize: 50000
display: 20
average_loss: 100
momentum: 0.9
weight_decay: 0.0005

#base_lr: 0.001
#lr_policy: "exp"
#gamma: 0.999539589  # (0.00001/0.001)^(1/10000)
#display: 1
#average_loss: 100
#momentum: 0.9
#weight_decay: 0.0005

# We disable standard caffe solver snapshotting and implement our own snapshot
# function
snapshot: 0
# We still use the snapshot prefix, though
snapshot_prefix: "zf_faster_rcnn"
iter_size: 2