comiple error about the ndarray or something else

[dangweili]

There is a problem when compile the mxnet on centos6.5 without cudadnn and the cuda version is 6.5.
The error is about the ndarray or something else, I am not sure about it. Could anyone else help me fix this problem ? Thank you very much.

build/src/ndarray/ndarray.o: In function std::_Function_handler<void ()(mxnet::RunContext, mxnet::engine::CallbackOnComplete), void mxnet::Engine::PushSync<mxnet::ClipOp(mxn et::NDArray const&, float const&, float const&, mxnet::NDArray*)::{lambda(mxnet::RunContext)#2}>(mxnet::ClipOp(mxnet::NDArray const&, float const&, float const&, mxnet::NDArr ay*)::{lambda(mxnet::RunContext)#2}, mxnet::Context, std::vector<mxnet::engine::Var*, std::allocator<mxnet::engine::Var> > const&, std::vector<mxnet::engine::Var*, std::alloc ator<mxnet::engine::Var> > const, mxnet::FnProperty, int)::{lambda(mxnet::RunContext, mxnet::engine::CallbackOnComplete)#1}>::_M_invoke(std::_Any_data const&, mxnet::RunConte xt, mxnet::engine::CallbackOnComplete)': ndarray.cc:(.text+0x7aa): undefined reference tovoid mxnet::ndarray::EvalClipmshadow::gpu(mshadow::TBlob const&, float const&, float const&, mshadow::TBlob_, mxnet::RunCo
ntext)'
build/src/ndarray/ndarray.o: In function std::_Function_handler<void ()(mxnet::RunContext, mxnet::engine::CallbackOnComplete), void mxnet::Engine::PushSync<mxnet::SetValueOp (float const&, mxnet::NDArray_)::{lambda(mxnet::RunContext)#2}>(mxnet::SetValueOp(float const&, mxnet::NDArray_)::{lambda(mxnet::RunContext)#2}, mxnet::Context, std::vector<m xnet::engine::Var_, std::allocatormxnet::engine::Var > const&, std::vector<mxnet::engine::Var*, std::allocator<mxnet::engine::Var> > const, mxnet::FnProperty, int)::{lambda (mxnet::RunContext, mxnet::engine::CallbackOnComplete)#1}>::_M_invoke(std::_Any_data const&, mxnet::RunContext, mxnet::engine::CallbackOnComplete)': ndarray.cc:(.text+0x8b1): undefined reference to void mxnet::ndarray::Evalmshadow::gpu(float const&, mshadow::TBlob*, mxnet::RunContext)'
build/src/ndarray/ndarray.o: In functionstd::_Function_handler<void ()(mxnet::RunContext, mxnet::engine::CallbackOnComplete), void mxnet::Engine::PushSync<mxnet::CopyFromTo (mxnet::NDArray const&, mxnet::NDArray*, int)::{lambda(mxnet::RunContext)#2}>(mxnet::CopyFromTo(mxnet::NDArray const&, mxnet::NDArray_, int)::{lambda(mxnet::RunContext)#2}, m xnet::Context, std::vector<mxnet::engine::Var_, std::allocatormxnet::engine::Var > const&, std::vector<mxnet::engine::Var*, std::allocator<mxnet::engine::Var> > const, mxne t::FnProperty, int)::{lambda(mxnet::RunContext, mxnet::engine::CallbackOnComplete)#1}>::_M_invoke(std::_Any_data const&, mxnet::RunContext, mxnet::engine::CallbackOnComplete) ': ndarray.cc:(.text+0x9e5): undefined reference to void mxnet::ndarray::Copy<mshadow::cpu, mshadow::gpu>(mshadow::TBlob const&, mshadow::TBlob*, mxnet::Context, mxnet::Context
, mxnet::RunContext)'
build/src/ndarray/ndarray.o: In functionstd::_Function_handler<void ()(mxnet::RunContext, mxnet::engine::CallbackOnComplete), void mxnet::Engine::PushSync<mxnet::CopyFromTo (mxnet::NDArray const&, mxnet::NDArray*, int)::{lambda(mxnet::RunContext)#3}>(mxnet::CopyFromTo(mxnet::NDArray const&, mxnet::NDArray_, int)::{lambda(mxnet::RunContext)#3}, m xnet::Context, std::vector<mxnet::engine::Var_, std::allocatormxnet::engine::Var > const&, std::vector<mxnet::engine::Var*, std::allocator<mxnet::engine::Var> > const, mxne t::FnProperty, int)::{lambda(mxnet::RunContext, mxnet::engine::CallbackOnComplete)#1}>::_M_invoke(std::_Any_data const&, mxnet::RunContext, mxnet::engine::CallbackOnComplete) ': ndarray.cc:(.text+0xb55): undefined reference to void mxnet::ndarray::Copy<mshadow::gpu, mshadow::cpu>(mshadow::TBlob const&, mshadow::TBlob*, mxnet::Context, mxnet::Context
, mxnet::RunContext)'
build/src/ndarray/ndarray.o: In functionstd::_Function_handler<void ()(mxnet::RunContext, mxnet::engine::CallbackOnComplete), void mxnet::Engine::PushSync<mxnet::CopyFromTo (mxnet::NDArray const&, mxnet::NDArray*, int)::{lambda(mxnet::RunContext)#4}>(mxnet::CopyFromTo(mxnet::NDArray const&, mxnet::NDArray_, int)::{lambda(mxnet::RunContext)#4}, m xnet::Context, std::vector<mxnet::engine::Var_, std::allocatormxnet::engine::Var > const&, std::vector<mxnet::engine::Var*, std::allocator<mxnet::engine::Var> > const, mxne t::FnProperty, int)::{lambda(mxnet::RunContext, mxnet::engine::CallbackOnComplete)#1}>::_M_invoke(std::_Any_data const&, mxnet::RunContext, mxnet::engine::CallbackOnComplete) ': ndarray.cc:(.text+0xcc5): undefined reference to void mxnet::ndarray::Copy<mshadow::gpu, mshadow::gpu>(mshadow::TBlob const&, mshadow::TBlob*, mxnet::Context, mxnet::Context
, mxnet::RunContext)'

................................................................
................................................................
build/src/operator/softmax_activation.o: In function mxnet::op::SoftmaxActivationProp::CreateOperator(mxnet::Context) const': softmax_activation.cc:(.text+0x3bc): undefined reference tomxnet::Operator* mxnet::op::CreateOpmshadow::gpu(mxnet::op::SoftmaxActivationParam)'
build/src/operator/softmax_output.o: In function mxnet::op::SoftmaxOutputProp::CreateOperator(mxnet::Context) const': softmax_output.cc:(.text+0x7a1): undefined reference tomxnet::Operator* mxnet::op::CreateOpmshadow::gpu(mxnet::op::SoftmaxOutputParam)'
build/src/operator/swapaxis.o: In function mxnet::op::SwapAxisProp::CreateOperator(mxnet::Context) const': swapaxis.cc:(.text+0x38d): undefined reference tomxnet::Operator* mxnet::op::CreateOpmshadow::gpu(mxnet::op::SwapAxisParam)'
build/src/operator/upsampling.o: In function mxnet::op::UpSamplingProp::CreateOperator(mxnet::Context) const': upsampling.cc:(.text+0x38a5): undefined reference tomxnet::Operator* mxnet::op::CreateOpmshadow::gpu(mxnet::op::UpSamplingParam)'
build/src/optimizer/sgd.o: In function std::_Function_handler<void ()(mxnet::RunContext, mxnet::engine::CallbackOnComplete), void mxnet::Engine::PushSync<mxnet::opt::SGDOpt: :Update(int, mxnet::NDArray*, mxnet::NDArray const*, float, float)::{lambda(mxnet::RunContext)#4}>(mxnet::opt::SGDOpt::Update(int, mxnet::NDArray*, mxnet::NDArray const*, flo at, float)::{lambda(mxnet::RunContext)#4}, mxnet::Context, std::vector<mxnet::engine::Var*, std::allocator<mxnet::engine::Var> > const&, std::vector<mxnet::engine::Var*, std: :allocator<mxnet::engine::Var> > const, mxnet::FnProperty, int)::{lambda(mxnet::RunContext, mxnet::engine::CallbackOnComplete)#1}>::_M_invoke(std::_Any_data const&, mxnet::Ru nContext, mxnet::engine::CallbackOnComplete)': sgd.cc:(.text._ZNSt17_Function_handlerIFvN5mxnet10RunContextENS0_6engine18CallbackOnCompleteEEZNS0_6Engine8PushSyncIZNS0_3opt6SGDOpt6UpdateEiPNS0_7NDArrayEPKS9_ffEUlS1_E2_EEv T_NS0_7ContextERKSt6vectorIPNS2_3VarESaISI_EESM_NS0_10FnPropertyEiEUlS1_S3_E_E9_M_invokeERKSt9_Any_dataS1_S3_[_ZNSt17_Function_handlerIFvN5mxnet10RunContextENS0_6engine18Call backOnCompleteEEZNS0_6Engine8PushSyncIZNS0_3opt6SGDOpt6UpdateEiPNS0_7NDArrayEPKS9_ffEUlS1_E2_EEvT_NS0_7ContextERKSt6vectorIPNS2_3VarESaISI_EESM_NS0_10FnPropertyEiEUlS1_S3_E_E 9_M_invokeERKSt9_Any_dataS1_S3_]+0x9c): undefined reference tomxnet::opt::call_sgd_update_gpu(mxnet::RunContext, mshadow::TBlob, mshadow::TBlob, float, float, mxnet::opt::S
GDParam const&)'
build/src/optimizer/sgd.o: In function std::_Function_handler<void ()(mxnet::RunContext, mxnet::engine::CallbackOnComplete), void mxnet::Engine::PushSync<mxnet::opt::SGDOpt$ :Update(int, mxnet::NDArray*, mxnet::NDArray const*, float, float)::{lambda(mxnet::RunContext)#3}>(mxnet::opt::SGDOpt::Update(int, mxnet::NDArray*, mxnet::NDArray const*, fl$ at, float)::{lambda(mxnet::RunContext)#3}, mxnet::Context, std::vector<mxnet::engine::Var*, std::allocator<mxnet::engine::Var> > const&, std::vector<mxnet::engine::Var*, std$ :allocator<mxnet::engine::Var> > const, mxnet::FnProperty, int)::{lambda(mxnet::RunContext, mxnet::engine::CallbackOnComplete)#1}>::_M_invoke(std::_Any_data const&, mxnet::R$ nContext, mxnet::engine::CallbackOnComplete)': sgd.cc:(.text._ZNSt17_Function_handlerIFvN5mxnet10RunContextENS0_6engine18CallbackOnCompleteEEZNS0_6Engine8PushSyncIZNS0_3opt6SGDOpt6UpdateEiPNS0_7NDArrayEPKS9_ffEUlS1_E1_EE$T_NS0_7ContextERKSt6vectorIPNS2_3VarESaISI_EESM_NS0_10FnPropertyEiEUlS1_S3_E_E9_M_invokeERKSt9_Any_dataS1_S3_[_ZNSt17_Function_handlerIFvN5mxnet10RunContextENS0_6engine18Cal$ backOnCompleteEEZNS0_6Engine8PushSyncIZNS0_3opt6SGDOpt6UpdateEiPNS0_7NDArrayEPKS9_ffEUlS1_E1_EEvT_NS0_7ContextERKSt6vectorIPNS2_3VarESaISI_EESM_NS0_10FnPropertyEiEUlS1_S3_E_$ 9_M_invokeERKSt9_Any_dataS1_S3_]+0xbd): undefined reference tomxnet::opt::call_sgd_mom_update_gpu(mxnet::RunContext, mshadow::TBlob, mshadow::TBlob, mshadow::TBlob, float, float, mxnet::opt::SGDParam const&)'
collect2: error: ld returned 1 exit status
make: *** [bin/im2rec] Error 1

[dangweili]

I think I should add message below to give a better description about the error state.

g++ -DMSHADOW_FORCE_STREAM -Wall -O3 -I./mshadow/ -I./dmlc-core/include -fPIC -Iinclude -msse3 -funroll-loops -Wno-unused-parameter -Wno-unknown-pragmas -I/data1/NLPRMNT/lidangwei/local/cuda-6.5/include -DMSHADOW_USE_CBLAS=1 -DMSHADOW_USE_MKL=0 -DMSHADOW_RABIT_PS=0 -DMSHADOW_DIST_PS=0 -DMXNET_USE_OPENCV=1 pkg-config --cflags opencv -fopenmp -DMXNET_USE_NVRTC=0 -std=c++0x -o bin/im2rec tools/im2rec.cc build/src/resource.o build/src/c_api/c_api.o build/src/c_api/c_api_error.o build/src/c_api/c_predict_api.o build/src/common/mxrtc.o build/src/common/tblob_op_registry.o build/src/engine/engine.o build/src/engine/naive_engine.o build/src/engine/threaded_engine.o build/src/engine/threaded_engine_perdevice.o build/src/engine/threaded_engine_pooled.o build/src/io/io.o build/src/io/iter_csv.o build/src/io/iter_image_recordio.o build/src/io/iter_mnist.o build/src/kvstore/kvstore.o build/src/ndarray/ndarray.o build/src/ndarray/ndarray_function.o build/src/ndarray/unary_function.o build/src/operator/activation.o build/src/operator/batch_norm.o build/src/operator/block_grad.o build/src/operator/cast.o build/src/operator/concat.o build/src/operator/convolution.o build/src/operator/crop.o build/src/operator/cross_device_copy.o build/src/operator/cudnn_batch_norm.o build/src/operator/deconvolution.o build/src/operator/dropout.o build/src/operator/elementwise_binary_op.o build/src/operator/elementwise_binary_scalar_op.o build/src/operator/elementwise_sum.o build/src/operator/embedding.o build/src/operator/fully_connected.o build/src/operator/identity_attach_KL_sparse_reg.o build/src/operator/leaky_relu.o build/src/operator/lrn.o build/src/operator/native_op.o build/src/operator/ndarray_op.o build/src/operator/operator.o build/src/operator/pooling.o build/src/operator/regression_output.o build/src/operator/reshape.o build/src/operator/slice_channel.o build/src/operator/softmax_activation.o build/src/operator/softmax_output.o build/src/operator/swapaxis.o build/src/operator/upsampling.o build/src/optimizer/optimizer.o build/src/optimizer/sgd.o build/src/storage/storage.o build/src/symbol/graph_executor.o build/src/symbol/graph_memory_allocator.o build/src/symbol/static_graph.o build/src/symbol/symbol.o dmlc-core/libdmlc.a -pthread -lm -lcudart -lcublas -lcurand -L/data1/NLPRMNT/lidangwei/local/cuda-6.5/lib64 -L/data1/NLPRMNT/lidangwei/local/cuda-6.5/lib -lrt pkg-config --libs opencv

[dangweili]

@mli @piiswrong could you give me some suggestions?

[piiswrong]

it's hard to tell what happened.
Try removing the repo and do a fresh clone, then git submodule update && make clean && make

[dangweili]

I reinstall the repo and change the atlas to mkl, then it works.
The mxnet cannot find the atlas library, although it is in the system path (/usr/lib64/atlas).
So how to add the other include and library path in the config.mk ?

[dangweili]

After the compile is ok, there still exist a problem in running.
I test the example of python example/image-classification/train_mnist.py
It returns a Segmentation fault.
So I try another example Inception_BN classification, It stills return a segmentation error.
I donot know why.

python example/image-classification/train_mnist.py

/data1/NLPRMNT/lidangwei/MxNet/mxnet/example/image-classification/train_mnist.py(131)()
-> train_model.fit(args, net, get_iterator(data_shape))
(Pdb) c
2016-02-25 18:49:29,276 Node[0] start with arguments Namespace(batch_size=128, data_dir='mnist/', gpus=None, kv_store='local', load_epoch=None, lr=0.1, lr_factor=1, lr_factor_epoch=1, model_prefix=None, network='mlp', num_epochs=10, num_examples=60000)
[Segmentation fault

[dangweili]

Here I list the example code given in example/notebook/...
The code will be segmentation fault at model = mx.model.Feed.....
The compile is ok. So why?

#!/usr/bin/env python
import mxnet as mx
import logging
import numpy as np
from skimage import io, transform
import pdb
pdb.set_trace()
logger = logging.getLogger()
logger.setLevel(logging.DEBUG)

Load the pre-trained model

prefix = "Inception/Inception_BN"
num_round = 39
model = mx.model.FeedForward.load(prefix, num_round, ctx=mx.gpu(), numpy_batch_size=1)

[dangweili]

I have worked out the problem, this issue could be closed!