Caffe error Cannot copy param 0 weights from layer, shape mismatch

最新推荐文章于 2021-08-24 15:52:16 发布
最新推荐文章于 2021-08-24 15:52:16 发布
阅读量7.7k 收藏 2
点赞数 3
分类专栏： 深度学习
版权声明：本文为博主原创文章，遵循 CC 4.0 BY-SA 版权协议，转载请附上原文出处链接和本声明。
本文链接：https://blog.csdn.net/zouyu1746430162/article/details/53787737
版权

I am getting following error while running test app. (when I am using trained model with my test application)

F0518 18:21:13.978204 13437 net.cpp:766] Cannot copy param 0 weights from layer 'fc6'; shape mismatch.  Source param shape is 4096 2304 (9437184); target param shape is 4096 9216 (37748736). To learn this layer's parameters from scratch rather than copying from a saved net, rename the layer.
*** Check failure stack trace: ***
Aborted (core dumped)

Can anybody tell the reason for this ?

deploy.prototxt

name: "CaffeNet"
layer {
  name: "data"
  type: "Input"
  top: "data"
  input_param { shape: { dim: 10 dim: 3 dim: 227 dim: 227 } }
}
layer {
  name: "conv1"
  type: "Convolution"
  bottom: "data"
  top: "conv1"
  convolution_param {
    num_output: 96
    kernel_size: 11
    stride: 4
  }
}
layer {
  name: "relu1"
  type: "ReLU"
  bottom: "conv1"
  top: "conv1"
}
layer {
  name: "pool1"
  type: "Pooling"
  bottom: "conv1"
  top: "pool1"
  pooling_param {
    pool: MAX
    kernel_size: 3
    stride: 2
  }
}
layer {
  name: "norm1"
  type: "LRN"
  bottom: "pool1"
  top: "norm1"
  lrn_param {
    local_size: 5
    alpha: 0.0001
    beta: 0.75
  }
}
layer {
  name: "conv2"
  type: "Convolution"
  bottom: "norm1"
  top: "conv2"
  convolution_param {
    num_output: 256
    pad: 2
    kernel_size: 5
    group: 2
  }
}
layer {
  name: "relu2"
  type: "ReLU"
  bottom: "conv2"
  top: "conv2"
}
layer {
  name: "pool2"
  type: "Pooling"
  bottom: "conv2"
  top: "pool2"
  pooling_param {
    pool: MAX
    kernel_size: 3
    stride: 2
  }
}
layer {
  name: "norm2"
  type: "LRN"
  bottom: "pool2"
  top: "norm2"
  lrn_param {
    local_size: 5
    alpha: 0.0001
    beta: 0.75
  }
}
layer {
  name: "conv3"
  type: "Convolution"
  bottom: "norm2"
  top: "conv3"
  convolution_param {
    num_output: 384
    pad: 1
    kernel_size: 3
  }
}
layer {
  name: "relu3"
  type: "ReLU"
  bottom: "conv3"
  top: "conv3"
}
layer {
  name: "conv4"
  type: "Convolution"
  bottom: "conv3"
  top: "conv4"
  convolution_param {
    num_output: 384
    pad: 1
    kernel_size: 3
    group: 2
  }
}
layer {
  name: "relu4"
  type: "ReLU"
  bottom: "conv4"
  top: "conv4"
}
layer {
  name: "conv5"
  type: "Convolution"
  bottom: "conv4"
  top: "conv5"
  convolution_param {
    num_output: 256
    pad: 1
    kernel_size: 3
    group: 2
  }
}
layer {
  name: "relu5"
  type: "ReLU"
  bottom: "conv5"
  top: "conv5"
}
layer {
  name: "pool5"
  type: "Pooling"
  bottom: "conv5"
  top: "pool5"
  pooling_param {
    pool: MAX
    kernel_size: 3
    stride: 2
  }
}
layer {
  name: "fc6"
  type: "InnerProduct"
  bottom: "pool5"
  top: "fc6"
  inner_product_param {
    num_output: 4096
  }
}
layer {
  name: "relu6"
  type: "ReLU"
  bottom: "fc6"
  top: "fc6"
}
layer {
  name: "drop6"
  type: "Dropout"
  bottom: "fc6"
  top: "fc6"
  dropout_param {
    dropout_ratio: 0.5
  }
}
layer {
  name: "fc7"
  type: "InnerProduct"
  bottom: "fc6"
  top: "fc7"
  inner_product_param {
    num_output: 4096
  }
}
layer {
  name: "relu7"
  type: "ReLU"
  bottom: "fc7"
  top: "fc7"
}
layer {
  name: "drop7"
  type: "Dropout"
  bottom: "fc7"
  top: "fc7"
  dropout_param {
    dropout_ratio: 0.5
  }
}
layer {
  name: "fc8"
  type: "InnerProduct"
  bottom: "fc7"
  top: "fc8"
  inner_product_param {
    num_output: 2
  }
}
layer {
  name: "prob"
  type: "Softmax"
  bottom: "fc8"
  top: "prob"
}

train_val.prototxt

name: "CaffeNet"
layer {
  name: "data"
  type: "Data"
  top: "data"
  top: "label"
  include {
    phase: TRAIN
  }
  transform_param {
    mirror: true
    crop_size: 256
    mean_file: "data/lmvhmv/imagenet_mean.binaryproto"
  }
# mean pixel / channel-wise mean instead of mean image
#  transform_param {
#    crop_size: 126
#    mean_value: 104
#    mean_value: 117
#    mean_value: 123
#    mirror: true
#  }
  data_param {
    source: "examples/imagenet/lmvhmv1_train_lmdb"
    batch_size: 10
    backend: LMDB
  }
}
layer {
  name: "data"
  type: "Data"
  top: "data"
  top: "label"
  include {
    phase: TEST
  }
  transform_param {
    mirror: false
    crop_size: 256
    mean_file: "data/ilsvrc12/imagenet_mean.binaryproto"
  }
# mean pixel / channel-wise mean instead of mean image
#  transform_param {
#    crop_size: 256
#    mean_value: 104
#    mean_value: 117
#    mean_value: 123
#    mirror: true
#  }
  data_param {
    source: "examples/imagenet/lmvhmv1_test_lmdb"
    batch_size: 10
    backend: LMDB
  }
}
layer {
  name: "conv1"
  type: "Convolution"
  bottom: "data"
  top: "conv1"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  convolution_param {
    num_output: 96
    kernel_size: 11
    stride: 4
    weight_filler {
      type: "gaussian"
      std: 0.01
    }
    bias_filler {
      type: "constant"
      value: 1
    }
  }
}
layer {
  name: "relu1"
  type: "ReLU"
  bottom: "conv1"
  top: "conv1"
}
layer {
  name: "pool1"
  type: "Pooling"
  bottom: "conv1"
  top: "pool1"
  pooling_param {
    pool: MAX
    kernel_size: 3
    stride: 2
  }
}
layer {
  name: "norm1"
  type: "LRN"
  bottom: "pool1"
  top: "norm1"
  lrn_param {
    local_size: 5
    alpha: 0.0001
    beta: 0.75
  }
}
layer {
  name: "conv2"
  type: "Convolution"
  bottom: "norm1"
  top: "conv2"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  convolution_param {
    num_output: 256
    pad: 2
    kernel_size: 5
    group: 2
    weight_filler {
      type: "gaussian"
      std: 0.01
    }
    bias_filler {
      type: "constant"
      value: 1
    }
  }
}
layer {
  name: "relu2"
  type: "ReLU"
  bottom: "conv2"
  top: "conv2"
}
layer {
  name: "pool2"
  type: "Pooling"
  bottom: "conv2"
  top: "pool2"
  pooling_param {
    pool: MAX
    kernel_size: 3
    stride: 2
  }
}
layer {
  name: "norm2"
  type: "LRN"
  bottom: "pool2"
  top: "norm2"
  lrn_param {
    local_size: 5
    alpha: 0.0001
    beta: 0.75
  }
}
layer {
  name: "conv3"
  type: "Convolution"
  bottom: "norm2"
  top: "conv3"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  convolution_param {
    num_output: 384
    pad: 1
    kernel_size: 3
    weight_filler {
      type: "gaussian"
      std: 0.01
    }
    bias_filler {
      type: "constant"
      value: 1
    }
  }
}
layer {
  name: "relu3"
  type: "ReLU"
  bottom: "conv3"
  top: "conv3"
}
layer {
  name: "conv4"
  type: "Convolution"
  bottom: "conv3"
  top: "conv4"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  convolution_param {
    num_output: 384
    pad: 1
    kernel_size: 3
    group: 2
    weight_filler {
      type: "gaussian"
      std: 0.01
    }
    bias_filler {
      type: "constant"
      value: 1
    }
  }
}
layer {
  name: "relu4"
  type: "ReLU"
  bottom: "conv4"
  top: "conv4"
}
layer {
  name: "conv5"
  type: "Convolution"
  bottom: "conv4"
  top: "conv5"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  convolution_param {
    num_output: 256
    pad: 1
    kernel_size: 3
    group: 2
    weight_filler {
      type: "gaussian"
      std: 0.01
    }
    bias_filler {
      type: "constant"
      value: 1
    }
  }
}
layer {
  name: "relu5"
  type: "ReLU"
  bottom: "conv5"
  top: "conv5"
}
layer {
  name: "pool5"
  type: "Pooling"
  bottom: "conv5"
  top: "pool5"
  pooling_param {
    pool: MAX
    kernel_size: 3
    stride: 2
  }
}
layer {
  name: "fc6"
  type: "InnerProduct"
  bottom: "pool5"
  top: "fc6"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  inner_product_param {
    num_output: 4096
    weight_filler {
      type: "gaussian"
      std: 0.005
    }
    bias_filler {
      type: "constant"
      value: 1
    }
  }
}
layer {
  name: "relu6"
  type: "ReLU"
  bottom: "fc6"
  top: "fc6"
}
layer {
  name: "drop6"
  type: "Dropout"
  bottom: "fc6"
  top: "fc6"
  dropout_param {
    dropout_ratio: 0.5
  }
}
layer {
  name: "fc7"
  type: "InnerProduct"
  bottom: "fc6"
  top: "fc7"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  inner_product_param {
    num_output: 4096
    weight_filler {
      type: "gaussian"
      std: 0.005
    }
    bias_filler {
      type: "constant"
      value: 1
    }
  }
}
layer {
  name: "relu7"
  type: "ReLU"
  bottom: "fc7"
  top: "fc7"
}
layer {
  name: "drop7"
  type: "Dropout"
  bottom: "fc7"
  top: "fc7"
  dropout_param {
    dropout_ratio: 0.5
  }
}
layer {
  name: "fc8"
  type: "InnerProduct"
  bottom: "fc7"
  top: "fc8"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  inner_product_param {
    num_output: 2
    weight_filler {
      type: "gaussian"
      std: 0.01
    }
    bias_filler {
      type: "constant"
      value: 1
    }
  }
}
layer {
  name: "accuracy"
  type: "Accuracy"
  bottom: "fc8"
  bottom: "label"
  top: "accuracy"
  include {
    phase: TEST
  }
}
layer {
  name: "loss"
  type: "SoftmaxWithLoss"
  bottom: "fc8"
  bottom: "label"
  top: "loss"
}
caffe
shareimprove this question



Share the prototxt file that you used to train and test to pinpoint the issue. It should be mainly because the dimensions of the input image is not the same in both the test prototxt and train prototxt. Check the height, width and channel count of both the test and train prototxt – Anoop K. Prabhu May 18 at 13:25


I was able to train caffe properly, this is happening in test application where I am using trained caffe model. – user_12 May 18 at 13:46


Ok, Now its working when I corrected dim value in deploy.txt its working – user_12 May 18 at 13:59


Though the issue is solved, I am writing an answer to help others who come up with the same issue later. – Anoop K. Prabhu May 18 at 15:40

1 Answer

up vote 4 down vote accepted

The training was done on an image of dimension 256x256x3, but test prototxt tries to read in 227x227x3 image. The convolutional layers will not show any issue as the dimension of input blobs to convolutional layers doesnot matter. The fully connected layers will crash when it finds that the input blob have a different dimension. In other words, the dimension of the weights of a fully connected layer is related to the dimension of the input blobs.

shareimprove this answer

Your Answer

北漠苍狼1746430162
关注
  • 3
    点赞
  • 踩
  • 2
    收藏
    觉得还不错? 一键收藏
  • 0
    评论
专栏目录
Cannot copy param 0 weights from layer 'conv6'; shape mismatch. Source param shape is 21 512 1 1
w_q_q2017的博客
10-12 1115
今天在使用别人的模型训练自己的数据时候，遇到了如下错误： Cannot copy param 0 weights from layer 'conv6'; shape mismatch.  Source param shape is 21 512 1 1 (10752); target param shape is 5 512 1 1 (2560). To learn this layer's p...
caffe源码阅读《六》base_conv_layer
念去去的博客
04-07 427
BaseConvolutionLayer是所有卷积层的基类。为什么卷积层还需要基类呢？因为再caffe里除了 ConvolutionLayer还有 DeconvolutionLayer BaseConvolutionLayer 类 BaseConvolutionLayer 构造函数 explicit BaseConvolutionLayer(const LayerParameter& ...
Cannot copy param 0 weights from layer 'conv1'; shape mismatch.
一个做图像文本的深度学习人
04-09 3159
当我使用训练好的caffemodel准备测试一下图片的时候，我定义了deploy.prototxt文件，然后利用bat脚本进行测试，发现报错，错误如下：错误解决：打开我的deploy.prototxt文件，把Input_param中的数据改掉，把1通道改为3通道，shape参数的意思：第一个参数表示batch_size，第二个参数表示通道（彩色为3通道），剩下两个表示图像大小，我用的是黑白图像，所...
Cannot copy param 0 weights from layer ‘loss1/fc’; shape mismatch
weixin_44705725的博客
08-05 320
在转换googlenet模型时候总是出现这个问题，困扰 好久了，百度上找也找不到解决办法，今天终于解决了 ERROR： Cannot copy param 0 weights from layer 'loss1/fc'; shape mismatch. Source param shape is 1 1 1024 2048 (2097152); target param shape is ...
Cannot copy param 0 weights from layer 'fc6'; shape mismatch.
机器学习的小学生
05-17 6518
问题：Cannot copy param 0 weights from layer ‘fc6’; shape mismatch. Source param shape is 4096 9216 (37748736); target param shape is 4096 2304 (9437184). To learn this layer’s parameters from scratch ra
cannot copy param 0 weights from layer
jane_6091的博客
11-07 1013
I have this problem when i train my data by caffe . F1107 12:42:42.507861 2369 net.cpp:774] Cannot copy param 0 weights from layer 'conv4_3_norm_mbox_conf'; shape mismatch. Source param shape is 4...
Caffe解读（三）
cookie_234的博客
03-25 6826
Caffe 梳理 net.cpp
Caffe: Net类解析(1)--原创
ricky5000的博客
03-31 5006
Net类是Caffe中的一个核心类, 他将各个Layer和Blob组织到一起, 完成前向和反响传播计算 下面详细解读Net.cpp的代码: ----------------------------------------------------------------------------------------------------------------------------
caffe运行时常见错误
sunshine__0411的博客
02-23 1008
这篇文章记录了我运行python时遇到的错误，以及我的解决方法，可能不够全面，欢迎大家一起讨论，补充 1. import caffe 报错：No module named caffe 原因：没有添加caffe/python目录到bash shell中 ，或者是添加了没有source 解决：在～/.bashrc文件中添加export PYTHONPATH="/home/caffe-...
Caffe模型调试中的错误总结
BockSong的博客
08-16 1603
错误：Cannot copy param 0 weights from layer '...'; shape mismatch 即model和prototxt定义的模型不一样。通常是fine-tune时图片尺寸有了调整，或者改了网络结构想要train时去掉weight参数即可。   错误：Check failed: error == cudaSuccess (2 vs. 0) out of ...
在自己的数据集上微调Light CNN
honghu549599AAA的专栏
11-29 3212
1.编写微调的网络结构lightcnn_train_test.prototxt2.准备训练图像数据，分成训练集和验证集合注意 要将图像灰度化3.运行训练脚本文件code/caffe-master/build/tools/caffe train \ --solver=code/caffe-master/lightCNNFace/LCNN_solver.prototxt \ --weig
Caffe学习：pycaffe利用caffemodel进行分类=>批处理
from_jiarenyf
09-30 4085
Caffe学习：pycaffe利用caffemodel进行分类=>批处理
pycaffe使用gpu推理的时候出错的可能解决方案。以及显存不够的可能解决方案。
york1996的博客
08-24 711
math_functions.cu:27] Check failed: status == CUBLAS_STATUS_SUCCESS (13 vs. 0) CUBLAS_STATUS_EXECUTION_FAILED 交换下面两行 caffe.set_device(gpu) # 设置用哪个gpu caffe.set_mode_gpu() # 用了这个才是真的在gpu上推理 并且，如果第一个gpu已经快满了，虽然你用的是有显存剩余的显卡。但是可能还会报错，这个时候，你可以再代码前面
matlab添加路径报错,Win10 + Caffe + CPU + MATLAB （包括各种问题详细解决）（二）
weixin_30199989的博客
03-16 836
转载自：https://blog.csdn.net/u014546828/article/details/80447583这里介绍一下 Win10 下，如何搭建 Caffe，仅有 CPU。因为我需要用 MATLAB 调试，所以需要对 Caffe 做关于 MATLAB 的配置。很多博客都写了有关这方面的配置过程，写的也很好，但对于我这种初学者可能还不够细。我就想写一个特别细致的配置过程。但太细又会显...
MATLAB调试caffe,在MATLAB下调试Caffe
weixin_42314846的博客
03-18 1036
Caffe本身是C++、CUDA语言编写的。在调试模型、参数时，根据运行log、snapshot很难实时反馈当前训练的权值情况，也难以捕捉算法存在的bug。MATLAB则是非常适合算法设计、快速迭代的利器，只需要做少量工作就能编写出复杂的算法，调试非常方便，位于workspace中的变量随时都能打印，无论是一维、二维还是三维数据，都能直观显示，从而有利于定位算法设计问题，减少调试时间。Caffe中...
为了caffe（七）做kaggle上的mnist
热门推荐
ddqqfree123的专栏
08-31 2万+
为了caffe（七）做kaggle上的mnist   因为蠢啊，因为不懂才写这么详细啊   第0步： 刚刚不能验证注册kaggle,注册需要ＶＰＮ翻墙一下             第一步： 　老样子　   dong@dong-Lenovo-G480:~$ sudo modprobe nvidia_352_uvm [sudo] password for don
Source param shape is 4096 9216 (37748736); target param shape is 4096 25088 (102760448).
北漠苍狼的专栏(QQ:1746430162)
12-21 3236
I1221 15:49:07.614038 28846 net.cpp:228] relu1_1 does not need backward computation. I1221 15:49:07.614042 28846 net.cpp:228] conv1_1 does not need backward computation. I1221 15:49:07.614045 28846
caffe的Matlab接口的使用方法
weixin_30814329的博客
01-15 399
编译MatCaffe 使用如下命令编译MatCaffe make all matcaffe 之后，你可以用以下命令测试MatCaffe： make mattest 如果你在运行上面命令时，遇到如下错误：libstdc++.so.6 version ‘GLIBCXX_3.4.15’ not found，说明你的Matlab库不匹...
Caffe学习笔记（1）--Python接口
Jesse_Mx的博客
02-28 1万+
前言使用caffe也有一小段时间了，但是对于caffe的python接口总是一知半解，最近终于能静下心来，仔细阅读了caffe官方例程，并写下此博客。博文主要对caffe自带的分类例程00-classification.ipynb做了详细的注释，相信能加强这方面的理解。准备工作加载必要的库import numpy as np #加载numpy import matplotlib.pyplot as
Ubuntu16安装caffe踩坑与解决：DSO missing from command line
"个人在安装caffe的过程中遇到的问题和解决方案，主要是在Ubuntu 16上，环境为anaconda3和cuda10.1，涉及到opencv的安装以及与caffe编译的冲突" 在安装Caffe时，作者遇到了从Ubuntu 14升级到16后的一系列问题。...
评论
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付：10.00元
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付元
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明：

1.余额是钱包充值的虚拟货币，按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载，可以购买VIP、付费专栏及课程。

余额充值