(-215:Assertion failed) 1 <= blobs.size() && blobs.size() <= 2

Question

I'm trying to import my own pre-trained Caffe Googlenet model using OpenCV v.3.4.3, where i run a Caffe test after training using the model deploy file and everything was working fine. However when feeding the OpenCv net (after loading it) with image blob i get an exception.

OpenCv Code:

Net net = Dnn.readNetFromCaffe("deploy.prototxt","bvlc_googlenet.caffemodel");
Mat image = Imgcodecs.imread(input.getAbsolutePath(), Imgcodecs.IMREAD_COLOR);
Mat blob = Dnn.blobFromImage(image);
System.out.println(image);
System.out.println(blob);
net.setInput(blob);
Mat result = net.forward().reshape(1);

Output Error:

Mat [ 24*15*CV_8UC3, isCont=true, isSubmat=false, nativeObj=0x1bcd0740, dataAddr=0x1a9d1880 ]
Mat [ -1*-1*CV_32FC1, isCont=true, isSubmat=false, nativeObj=0x1bcd0eb0, dataAddr=0x1a4e4340 ]
Exception in thread "main" CvException [org.opencv.core.CvException: cv::Exception: OpenCV(3.4.3) Z:\build tools\opencv-3.4.3\modules\dnn\src\layers\fully_connected_layer.cpp:73: error: (-215:Assertion failed) 1 <= blobs.size() && blobs.size() <= 2 in function 'cv::dnn::FullyConnectedLayerImpl::FullyConnectedLayerImpl'
]
    at org.opencv.dnn.Net.forward_1(Native Method)
    at org.opencv.dnn.Net.forward(Net.java:62)
    at test.OpenCVTests.main(OpenCVTests.java:54)

Caffe-train-val-model.prototxt Caffe-deploy-model.prototxt

Thanks in advance!

Answer 1

This issue was solved for me here: https://github.com/opencv/opencv/issues/12578#issuecomment-422304736

"there is no loss3/classifier_retrain from Caffe-deploy-model.prototxt in Caffe-train-val-model.prototxt. If you tried to run this model several times in Caffe you'll get different outputs for the same input because Caffe fills missed weights randomly."

Copyright (github.com/dkurt)