Faster Rcnn调试日记

版权声明：转载请联系博主 https://blog.csdn.net/zxtalentwolf/article/details/80627234

/usr/bin

/home/zhangxin/py-faster-rcnn/caffe-fast-rcnn/include/caffe/util/cudnn.hpp(107): error: too few arguments in function call

/home/zhangxin/py-faster-rcnn/caffe-fast-rcnn/include/caffe/util/cudnn.hpp(126): error: argument of type "int" is incompatible with parameter of type "cudnnNanPropagation_t"

/home/zhangxin/py-faster-rcnn/caffe-fast-rcnn/include/caffe/util/cudnn.hpp(126): error: too few arguments in function call

3 errors detected in the compilation of "/tmp/tmpxft_00001201_00000000-5_absval_layer.cpp4.ii".
CMake Error at cuda_compile_1_generated_absval_layer.cu.o.Release.cmake:279 (message):
  Error generating file
  /home/zhangxin/py-faster-rcnn/caffe-fast-rcnn/build/src/caffe/CMakeFiles/cuda_compile_1.dir/layers/./cuda_compile_1_generated_absval_layer.cu.o


make[2]: *** [src/caffe/CMakeFiles/cuda_compile_1.dir/layers/cuda_compile_1_generated_absval_layer.cu.o] Error 1
make[1]: *** [src/caffe/CMakeFiles/caffe.dir/all] Error 2
make: *** [all] Error 2

这里是cudn版本呢不对。要更新caffe才能支持cudnn6

cd caffe-fast-rcnn
# 将BLVC/caffe添加为远程仓库，命名为caffe
git remote add caffe https://github.com/BVLC/caffe.git
# 获取BLVC/caffe代码
git fetch caffe
# 将BLVC/caffe的master分支合并到当前分支
# -X theirs 指定在合并遇到冲突时以BLVC/caffe版本为主
git merge -X theirs caffe/master

• [x] 制作VOC格式数据集

发现frrcnn要imdb数据，寻找imdb数据转换的方式

修改py-faster-rcnn/models/pascal_voc/VGG16/faster_rcnn_alt_opt/stage1_fast_rcnn_train.pt

layer {  
  name: 'data'  
  type: 'Python'  
  top: 'data'  
  top: 'rois'  
  top: 'labels'  
  top: 'bbox_targets'  
  top: 'bbox_inside_weights'  
  top: 'bbox_outside_weights'  
  python_param {  
    module: 'roi_data_layer.layer'  
    layer: 'RoIDataLayer'  
    param_str: "'num_classes': 2" #按训练集类别改，该值为类别数+1  
  }  
}

layer {  
  name: "cls_score"  
  type: "InnerProduct"  
  bottom: "fc7"  
  top: "cls_score"  
  param { 
  lr_mult: 1.0
  }  
  param {
  lr_mult: 2.0 
  }  
  inner_product_param {  
    num_output: 2 #按训练集类别改，该值为类别数+1  
    weight_filler {  
      type: "gaussian"  
      std: 0.01  
    }  
    bias_filler {  
      type: "constant"  
      value: 0  
    }  
  }  
}

修改py-faster-rcnn/models/pascal_voc/VGG16/faster_rcnn_alt_opt/stage1_rpn_train.pt

layer {  
  name: 'input-data'  
  type: 'Python'  
  top: 'data'  
  top: 'im_info'  
  top: 'gt_boxes'  
  python_param {  
    module: 'roi_data_layer.layer'  
    layer: 'RoIDataLayer'  
    param_str: "'num_classes': 2" #按训练集类别改，该值为类别数+1  
  }  
}  

修改py-faster-rcnn/models/pascal_voc/VGG16/faster_rcnn_alt_opt/stage2_fast_rcnn_train.pt

layer {  
  name: 'data'  
  type: 'Python'  
  top: 'data'  
  top: 'rois'  
  top: 'labels'  
  top: 'bbox_targets'  
  top: 'bbox_inside_weights'  
  top: 'bbox_outside_weights'  
  python_param {  
    module: 'roi_data_layer.layer'  
    layer: 'RoIDataLayer'  
    param_str: "'num_classes': 2" #按训练集类别改，该值为类别数+1  
  }  
}

layer {  
  name: "cls_score"  
  type: "InnerProduct"  
  bottom: "fc7"  
  top: "cls_score"  
  param { 
  lr_mult: 1.0 
  }  
  param { 
  lr_mult: 2.0 
  }  
  inner_product_param {  
    num_output: 2 #按训练集类别改，该值为类别数+1  
    weight_filler {  
      type: "gaussian"  
      std: 0.01  
    }  
    bias_filler {  
      type: "constant"  
      value: 0  
    }  
  }  
}


layer {  
  name: "bbox_pred"  
  type: "InnerProduct"  
  bottom: "fc7"  
  top: "bbox_pred"  
  param { 
  lr_mult: 1.0
  }  
  param { 
  lr_mult: 2.0 
  }  
  inner_product_param {  
    num_output: 8 #按训练集类别改，该值为（类别数+1）*4,四个顶点坐标  
    weight_filler {  
      type: "gaussian"  
      std: 0.001  
    }  
    bias_filler {  
      type: "constant"  
      value: 0  
    }  
  }  
}

修改py-faster-rcnn/models/pascal_voc/VGG16/faster_rcnn_alt_opt/stage2_rpn_train.pt

layer {  
  name: 'input-data'  
  type: 'Python'  
  top: 'data'  
  top: 'im_info'  
  top: 'gt_boxes'  
  python_param {  
    module: 'roi_data_layer.layer'  
    layer: 'RoIDataLayer'  
    param_str: "'num_classes': 2" #按训练集类别改，该值为类别数+1  
  }  
}

修改py-faster-rcnn/models/pascal_voc/VGG16/faster_rcnn_alt_opt/faster_rcnn_test.pt

layer {  
  name: "cls_score"  
  type: "InnerProduct"  
  bottom: "fc7"  
  top: "cls_score"  
  inner_product_param {  
    num_output: 2 #按训练集类别改，该值为类别数+1  
  }  
}  

修改py-faster-rcnn/models/pascal_voc/VGG16/faster_rcnn_alt_opt/faster_rcnn_test.pt

layer {  
  name: "cls_score"  
  type: "InnerProduct"  
  bottom: "fc7"  
  top: "cls_score"  
  inner_product_param {  
    num_output: 2 #按训练集类别改，该值为类别数+1  
  }  
}

修改py-faster-rcnn/lib/datasets/pascal_voc.py

class pascal_voc(imdb):  
    def __init__(self, image_set, year, devkit_path=None):  
        imdb.__init__(self, 'voc_' + year + '_' + image_set)  
        self._year = year  
        self._image_set = image_set  
        self._devkit_path = self._get_default_path() if devkit_path is None \  
                            else devkit_path  
        self._data_path = os.path.join(self._devkit_path, 'VOC' + self._year)  
        self._classes = ('__background__', # always index 0  
                         captcha' # 有几个类别此处就写几个，我是两个
                      )  

修改py-faster-rcnn/lib/datasets/imdb.py
将append_flipped_images函数改为如下形式:

def append_flipped_images(self):  
        num_images = self.num_images  
        widths = [PIL.Image.open(self.image_path_at(i)).size[0]  
                  for i in xrange(num_images)]  
        for i in xrange(num_images):  
            boxes = self.roidb[i]['boxes'].copy()  
            oldx1 = boxes[:, 0].copy()  
            oldx2 = boxes[:, 2].copy()  
            boxes[:, 0] = widths[i] - oldx2 - 1  
            print boxes[:, 0]  
            boxes[:, 2] = widths[i] - oldx1 - 1  
            print boxes[:, 0]  
            assert (boxes[:, 2] >= boxes[:, 0]).all()  
            entry = {'boxes' : boxes,  
                     'gt_overlaps' : self.roidb[i]['gt_overlaps'],  
                     'gt_classes' : self.roidb[i]['gt_classes'],  
                     'flipped' : True}  
            self.roidb.append(entry)  
        self._image_index = self._image_index * 2

训练

　　训练前还需要注意几个地方

1. cache问题

   　　假如你之前训练了官方的VOC2007的数据集或其他的数据集，是会产生cache的问题的，建议在重新训练新的数据之前将其删除

(1) py-faster-rcnn/output
(2) py-faster-rcnn/data/cache

1. 训练参数

　　py-faster-rcnn/models/pascal_voc/VGG16/faster_rcnn_alt_opt/stage_fast_rcnn_solver*.pt

base_lr: 0.001
lr_policy: 'step'
step_size: 30000
display: 20

　　迭代次数在文件py-faster-rcnn/tools/train_faster_rcnn_alt_opt.py中进行修改

line 80

max_iters = [80000, 40000, 80000, 40000]

　　分别对应rpn第1阶段，fast rcnn第1阶段，rpn第2阶段，fast rcnn第2阶段的迭代次数，自己修改即可，不过注意这里的值不要小于上面的solver里面的step_size的大小，大家自己修改吧。

开始训练:

cd py-faster-rcnn
./experiments/scripts/faster_rcnn_alt_opt.sh 0 VGG16 pascal_voc 
　　指明使用第一块GPU(0),模型是VGG16,训练数据是pascal_voc(voc2007)，没问题的话应该可以迭代训练了

训练faster-rcnn出现错误“KeyError: ‘max_overlaps’”解决

解决方法：
删除py-faster-rcnn/data/VOCdevkit2007/annotations_cache这个文件夹；
删除py-faster-rcnn/data/cache文件夹。

结果

训练完毕，得到我们的训练模型，我们就可以使用它来进行我们的object detection了，具体是:
1. 将py-faster-rcnn/output/faster_rcnn_alt_opt/voc_2007_trainval/VGG16_faster_rcnn_final.caffemodel，拷贝到py-faster-rcnn/data/faster_rcnn_models下
2. 将你需要进行test的images放在py-faster-rcnn/data/demo下
3. 修改py-faster-rcnn/tools/demo.py文件
“`
CLASSES = (‘background‘, ‘captcha’) #参考你自己的类别写

NETS = {‘vgg16’: (‘VGG16’,
‘VGG16_faster_rcnn_final.caffemodel’), #改成你训练得到的model的name
‘zf’: (‘ZF’,
‘ZF_faster_rcnn_final.caffemodel’)
}

im_names = [‘1559.jpg’,’1564.jpg’] # 改成自己的test image的name
“`

models —> 三种模型, ZF(S)/VGG1024(M)/VGG16(L)
需要一一测试
- [ ] ZF(S)
- [ ] VGG1024(M)
- [ ] VGG16

注意VGG需要更多显存
结果如下