s9xie / hed

code for Holistically-Nested Edge Detection
Other
1.79k stars 528 forks source link

Make errors #14

Open yhlleo opened 8 years ago

yhlleo commented 8 years ago

When I make the project, using the command make all, some errors occur as follows:

src/caffe/layers/cudnn_conv_layer.cu(81): error: argument of type "cudnnAddMode_t" is incompatible with parameter of type "const void *"
          detected during instantiation of "void caffe::CuDNNConvolutionLayer<Dtype>::Forward_gpu(const std::vector<caffe::Blob<Dtype> *, std::allocator<caffe::Blob<Dtype> *>> &, const std::vector<caffe::Blob<Dtype> *, std::allocator<caffe::Blob<Dtype> *>> &) [with Dtype=float]" 
(157): here

...

src/caffe/layers/cudnn_conv_layer.cu(140): error: too few arguments in function call
          detected during instantiation of "void caffe::CuDNNConvolutionLayer<Dtype>::Backward_gpu(const std::vector<caffe::Blob<Dtype> *, std::allocator<caffe::Blob<Dtype> *>> &, const std::vector<__nv_bool, std::allocator<__nv_bool>> &, const std::vector<caffe::Blob<Dtype> *, std::allocator<caffe::Blob<Dtype> *>> &) [with Dtype=double]" 
(157): here

20 errors detected in the compilation of "/tmp/tmpxft_0000153f_00000000-16_cudnn_conv_layer.compute_50.cpp1.ii".

These problems don't occur in the caffe project. It seams that there are some issues on the cudnn. The version of related packages in my computer is:

Would you like to help me to find any problems? Thank you very much!

yhlleo commented 8 years ago

If I run the file solve.py (after download the data and caffemodels), some errors occur as follow:

E0505 16:49:01.775686 21977 common.cpp:113] Cannot create Cublas handle. Cublas won't be available.
E0505 16:49:01.779266 21977 common.cpp:120] Cannot create Curand generator. Curand won't be available.
F0505 16:49:01.782297 21977 common.cpp:151] Check failed: error == cudaSuccess (30 vs. 0)  unknown error

Do you know why?

yhlleo commented 8 years ago

I re-make my caffe project:

make clean
make all -j8
make test -j8
make runtest -j8
make distribute
make pycaffe
make matcaffe

The above issue is solved, but a new issue arising:

I0505 20:59:33.863049  6492 layer_factory.hpp:77] Creating layer data
F0505 20:59:33.863072  6492 layer_factory.hpp:81] Check failed: registry.count(type) == 1 (0 vs. 1) Unknown layer type: ImageLabelmapData (known types: AbsVal, Accuracy, ArgMax, BNLL, BatchNorm, BatchReindex, Bias, Concat, ContrastiveLoss, Convolution, Crop, Data, Deconvolution, Dropout, DummyData, ELU, Eltwise, Embed, EuclideanLoss, Exp, Filter, Flatten, HDF5Data, HDF5Output, HingeLoss, Im2col, ImageData, InfogainLoss, InnerProduct, Input, LRN, Log, MVN, MemoryData, MultinomialLogisticLoss, PReLU, Pooling, Power, Python, ReLU, Reduction, Reshape, SPP, Scale, Sigmoid, SigmoidCrossEntropyLoss, Silence, Slice, Softmax, SoftmaxWithLoss, Split, TanH, Threshold, Tile, WindowData)
*** Check failure stack trace: ***
weiliu620 commented 8 years ago

The author of this project seem have defined their own layers for ImageLabelmapData (see layers folder in the source code). It should be there. I don't see the problem in my build.

And, the authors seems not monitoring this project any more. Need to decide if you should invest time on it!

yhlleo commented 8 years ago

@weiliu620

To fix there issues, I change the version of my cudnn from v4 into v3. Now, the project works well.