i'm using opencv 3.4.1 DNN in java in order to load a "LeNet" model trained using keras and tensorflow in python. The model is saved as a tensorflow frozen model ".pb" where i'm using the following line of code to load the it:
Dnn cvDnn = new org.opencv.dnn.Dnn();
Net net = cvDnn.readNetFromTensorflow("C:\\Users\\kr\\Desktop\\Plate_Recognition_frozen.pb");
where the Error says:
OpenCV(3.4.1) Error: Unspecified error (Input layer not found: convolution2d_1_b_1) in cv::dnn::experimental_dnn_v4::`anonymous-namespace'::TFImporter::connect, file C:\build\master_winpack-bindings-win64-vc14-static\opencv\modules\dnn\src\tensorflow\tf_importer.cpp, line 553
Exception in thread "main" CvException [org.opencv.core.CvException: cv::Exception: OpenCV(3.4.1) C:\build\master_winpack-bindings-win64-vc14-static\opencv\modules\dnn\src\tensorflow\tf_importer.cpp:553: error: (-2) Input layer not found: convolution2d_1_b_1 in function cv::dnn::experimental_dnn_v4::`anonymous-namespace'::TFImporter::connect
]
at org.opencv.dnn.Dnn.readNetFromTensorflow_1(Native Method)
at org.opencv.dnn.Dnn.readNetFromTensorflow(Dnn.java:163)
at opencv.Main.main(Main.java:44)
Any help would be appreciated, thanks in advance.
Related
I am doing a text extraction on specific regions using Yolov5. I trained a model and convert it to onnx formate for OpenCV readable formate.. but when I load the model weight this error occurs.. all stack, GitHub issues are not able to solve my problem. I download graded my torch version according to Github resolved issues but my issue is still there.
Kindly if anyone has an idea about this error contact me plz.
I am very glad to receive your message.
Best regards. Error is below
[ERROR:0] global D:\a\opencv-python\opencv-python\opencv\modules\dnn\src\onnx\onnx_importer.cpp (720) cv::dnn::dnn4_v20211004::ONNXImporter::handleNode DNN/ONNX: ERROR during processing node with 1 inputs and 1 outputs: [Identity]:(onnx::Resize_445)
Traceback (most recent call last):
File "C:\Users\Python_Coder\Desktop\YOLO_T_ID\Custom-OCR-with-YOLO\Custom_OCRs.py", line 176, in <module>
net = cv2.dnn.readNet(modelWeights)
cv2.error: OpenCV(4.5.4) D:\a\opencv-python\opencv-python\opencv\modules\dnn\src\onnx\onnx_importer.cpp:739: error: (-2:Unspecified error) in function 'cv::dnn::dnn4_v20211004::ONNXImporter::handleNode'
> Node [Identity]:(onnx::Resize_445) parse error: OpenCV(4.5.4) D:\a\opencv-python\opencv-python\opencv\modules\dnn\src\dnn.cpp:5653: error: (-215:Assertion failed) inputs.size() in function 'cv::dnn::dnn4_v20211004::Layer::getMemoryShapes'
OpenCV error with PyTorch model loading
I download the caffe source code from Github and compile it as C++ static library, and I test the OpenCV face detection Caffe model with the static library, but the lib report below error:
[libprotobuf ERROR D:\ThirdPartyLibrary\protobuf\src\google\protobuf\text_format.cc:296] Error parsing text-format caffe.NetParameter: 984:14: Message type "caffe.LayerParameter" has no field named "norm_param".
F0328 02:08:05.225075 24332 upgrade_proto.cpp:88] Check failed: ReadProtoFromTextFile(param_file, param) Failed to parse NetParameter file: D:/DATA/PreTrainedModel/cv_facedet/deploy.prototxt
is it means that norm_param field is implement only by OpenCV, not a standard Caffe field?
norm_param is an argument of Normalization layer from SSD Caffe framework: https://github.com/weiliu89/caffe/blob/8a65ae316b34e7c8fdefa6e18bf08a23b78caa0e/src/caffe/proto/caffe.proto#L523
Origin repo: https://github.com/weiliu89/caffe/tree/ssd
There is no SSD object detection networks in the origin Caffe.
Probably, it has different name in Caffe: normalize_bbox_param. See this discussion.
as I'm learning Apple's Vision and CoreML framework but got stuck on how to use my own re-trained models. I tried training a VG16 model with Keras based on this tutorial. Everything looks OK except for some Keras version warnings. Then I tried converting the resulted model with CoreMLTools with the following code
coremlModel = coremltools.converters.keras.convert(
kmodel,
input_names = 'image',
image_input_names = 'image',
output_names = 'classLabelProbs',
class_labels = ['cats', 'dogs'],
)
During the conversion it gave me some version compatible warnings but otherwise it was successful:
WARNING:root:Keras version 2.0.6 detected. Last version known to be fully compatible of Keras is 2.0.4 .
WARNING:root:TensorFlow version 1.2.1 detected. Last version known to be fully compatible is 1.1.1 .
So I loaded this model into Apple's Vision+ML example code, but every time I tried to classify an image it failed with errors
Vision+ML Example[2090:2012481] Error: The VNCoreMLTransform request failed
Vision+ML Example[2090:2012481] Didn't get VNClassificationObservations
Error Domain=com.apple.vis Code=3 "The VNCoreMLTransform request failed" UserInfo={NSLocalizedDescription=The VNCoreMLTransform request failed, NSUnderlyingError=0x1c025d130 {Error Domain=com.apple.CoreML Code=0 "Dimensions of layer 'classLabelProbs' is not the same size as the number of class labels." UserInfo={NSLocalizedDescription=Dimensions of layer 'classLabelProbs' is not the same size as the number of class labels.}}}
I was guessing this is because the pre-trained VGG16 model has already had 1000 categories, so I tried 1000 categories and 1000 + 2 (cats and dogs) categories, but still got the same problem.
Did I miss anything? I'd greatly appreciate any clue and help.
I've tried the example in the opencv_contrib tutorial of dnn: http://docs.opencv.org/master/d5/de7/tutorial_dnn_googlenet.html
But it return two errors, I've search for hours and didn't get any solution. Does anyone could give me some help?
OpenCV Error: Assertion failed (retval == 0) in cv::ocl::Kernel::set, file D:\opencv32\opencv\modules\core\src\ocl.cpp, line 3366
OpenCV Error: Assertion failed (The following error occured while making forward() for layer "loss3/classifier": retval == 0) in cv::ocl::Kernel::set, file D:\opencv32\opencv\modules\core\src\ocl.cpp, line 3366
I am training my datasets in Liblinear: train heart_scale, but i am getting this error
can't open /home/linguistics/.train//train.ini
ERROR: Init file not found (/home/linguistics/.train//train.ini)
train>
Reference:
README file from Liblinear's source code. I downloaded it from here:
http://www.csie.ntu.edu.tw/~cjlin/cgi-bin/liblinear.cgi?+http://www.csie.ntu.edu.tw/~cjlin/liblinear+tar.gz
try ./train dataset_name instead of train dataset_name