Hi,

I have fine-tuned C3D model on ActivityNet video dataset using video frames. I am training on some parts of videos for examples frames [100..200] or [1234....1800] so my training list file might not start from frame 1 in order to do that I have just changed line 345 of VideoDataLayer: from: CHECK(ReadImageSequenceToVolumeDatum(file_list_[id].c_str(), 1, label_list_[id], new_length, new_height, new_width, sampling_rate, &datum)); to

CHECK(ReadImageSequenceToVolumeDatum(file_list_[id].c_str(), start_frm_list_[id], label_list_[id], new_length, new_height, new_width, sampling_rate, &datum));

When I test finetuned model and compute accuracy for list of test input data (formated as you described in your instruction) I get 0.3 accuracy.

However when I extract probability or accuracy of exact same list of test input data and same finetuned model I get always 0 accuracy (single precision and logprob around 5-6) and also based on extracted probabilities, I get very low 0.000005 accuracy. So extracted probabilities and accuracy are consistent but accuracy computed by finetuning_test is 30% which is reasonable.

Do you have any idea where this issue could possibly come from? Is it possible that the line I have changed make this issue and model need all the frames of one video clips to use for training/feature-extraction?

Best!

when I run "make all",there are errors: build_release/lib/libcaffe.so: undefined reference tocv::VideoCapture::set(int, double)' .build_release/lib/libcaffe.so: undefined reference to cv::VideoCapture::open(cv::String const&)' .build_release/lib/libcaffe.so: undefined reference tocv::VideoCapture::release()' .build_release/lib/libcaffe.so: undefined reference to cv::VideoCapture::~VideoCapture()' .build_release/lib/libcaffe.so: undefined reference tocv::VideoCapture::isOpened() const' .build_release/lib/libcaffe.so: undefined reference to cv::VideoCapture::get(int) const' .build_release/lib/libcaffe.so: undefined reference tocv::VideoCapture::read(cv::_OutputArray const&)' .build_release/lib/libcaffe.so: undefined reference to cv::VideoCapture::VideoCapture()' collect2: error: ld returned 1 exit status make: *** [.build_release/tools/compute_image_mean.bin] Error 1 make: *** Waiting for unfinished jobs.... .build_release/lib/libcaffe.so: undefined reference tocv::VideoCapture::set(int, double)' .build_release/lib/libcaffe.so: undefined reference to cv::VideoCapture::open(cv::String const&)' .build_release/lib/libcaffe.so: undefined reference tocv::VideoCapture::release()' .build_release/lib/libcaffe.so: undefined reference to cv::VideoCapture::~VideoCapture()' .build_release/lib/libcaffe.so: undefined reference tocv::VideoCapture::isOpened() const' .build_release/lib/libcaffe.so: undefined reference to cv::VideoCapture::get(int) const' .build_release/lib/libcaffe.so: undefined reference tocv::VideoCapture::read(cv::_OutputArray const&)' ` How can I fix this?

Thanks for sharing this nice source code. I managed to compile it on my laptop with GeForce 840M card with Ubuntu 16.04 and CUDA 7.5. However, make runtest produced several test failures and crashed with a core dump.

This is my Makefile.config:

## Refer to http://caffe.berkeleyvision.org/installation.html
# Contributions simplifying and improving our build system are welcome!

# CUDA directory contains bin/ and lib/ directories that we need.
CUDA_DIR := /usr

# CUDA architecture setting: going with all of them (up to CUDA 5.5 compatible).
# For the latest architecture, you need to install CUDA >= 6.0 and uncomment
# the *_50 lines below.
CUDA_ARCH := -gencode arch=compute_20,code=sm_20 \
        -gencode arch=compute_20,code=sm_21 \
        -gencode arch=compute_30,code=sm_30 \
        -gencode arch=compute_35,code=sm_35
        -gencode=arch=compute_50,code=sm_50 \
        -gencode=arch=compute_50,code=compute_50

# BLAS choice:
# atlas for ATLAS (default)
# mkl for MKL
# open for OpenBlas
BLAS := open
# Custom (MKL/ATLAS/OpenBLAS) include and lib directories.
# Leave commented to accept the defaults for your choice of BLAS
# (which should work)!
# BLAS_INCLUDE := /path/to/your/blas
# BLAS_LIB := /path/to/your/blas

# This is required only if you will compile the matlab interface.
# MATLAB directory should contain the mex binary in /bin.
MATLAB_DIR := /usr/local/MATLAB/R2016a/
# MATLAB_DIR := /Applications/MATLAB_R2012b.app

# NOTE: this is required only if you will compile the python interface.
# We need to be able to find Python.h and numpy/arrayobject.h.
PYTHON_INCLUDE := /usr/local/include/python2.7 \
        /usr/local/lib/python2.7/dist-packages/numpy/core/include
# Anaconda Python distribution is quite popular. Include path:
# PYTHON_INCLUDE := $(HOME)/anaconda/include \
        # $(HOME)/anaconda/include/python2.7 \
        # $(HOME)/anaconda/lib/python2.7/site-packages/numpy/core/include

# We need to be able to find libpythonX.X.so or .dylib.
PYTHON_LIB := /usr/local/lib
# PYTHON_LIB := $(HOME)/anaconda/lib

# Whatever else you find you need goes here.
INCLUDE_DIRS := $(PYTHON_INCLUDE) /usr/local/include /usr/include/hdf5/serial
LIBRARY_DIRS := $(PYTHON_LIB) /usr/local/lib /usr/lib

BUILD_DIR := build
DISTRIBUTE_DIR := distribute

# Uncomment for debugging.
# DEBUG := 1

# The ID of the GPU that 'make runtest' will use to run unit tests.
TEST_GPUID := 0

Below is the output of runtest:

$ make runtest
build/test/test_all.testbin 0 --gtest_shuffle
Cuda number of devices: 1
Setting to use device 0
Current device id: 0
Note: Randomizing tests' orders with a seed of 65971 .
[==========] Running 417 tests from 74 test cases.
[----------] Global test environment set-up.
[----------] 2 tests from SoftmaxWithLossLayerTest/0, where TypeParam = float
[ RUN      ] SoftmaxWithLossLayerTest/0.TestGradientCPU
[       OK ] SoftmaxWithLossLayerTest/0.TestGradientCPU (230 ms)
[ RUN      ] SoftmaxWithLossLayerTest/0.TestGradientGPU
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is 2.1941887140274048, which exceeds threshold_ * scale, where
computed_gradient evaluates to -0.10926592350006104,
estimated_gradient evaluates to 2.0849227905273438, and
threshold_ * scale evaluates to 0.020849227905273438.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,0
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is 1367.8038898706436, which exceeds threshold_ * scale, where
computed_gradient evaluates to -1.6719402074813843,
estimated_gradient evaluates to -1369.475830078125, and
threshold_ * scale evaluates to 13.694758415222168.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,1
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is 1871.9518399685621, which exceeds threshold_ * scale, where
computed_gradient evaluates to -0.20477624237537384,
estimated_gradient evaluates to -1872.1566162109375, and
threshold_ * scale evaluates to 18.721565246582031.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,2
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is 1937.4213381409645, which exceeds threshold_ * scale, where
computed_gradient evaluates to -0.52714818716049194,
estimated_gradient evaluates to -1937.948486328125, and
threshold_ * scale evaluates to 19.379484176635742.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,3
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is 2000.9808031320572, which exceeds threshold_ * scale, where
computed_gradient evaluates to -1.3934644460678101,
estimated_gradient evaluates to -2002.374267578125, and
threshold_ * scale evaluates to 20.02374267578125.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,4
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is 2067.3268105387688, which exceeds threshold_ * scale, where
computed_gradient evaluates to 0.57290428876876831,
estimated_gradient evaluates to -2066.75390625, and
threshold_ * scale evaluates to 20.667537689208984.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,5
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is 2132.0294613242149, which exceeds threshold_ * scale, where
computed_gradient evaluates to 0.8981136679649353,
estimated_gradient evaluates to -2131.13134765625, and
threshold_ * scale evaluates to 21.311313629150391.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,6
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is 2196.1026052832603, which exceeds threshold_ * scale, where
computed_gradient evaluates to 0.59381622076034546,
estimated_gradient evaluates to -2195.5087890625, and
threshold_ * scale evaluates to 21.955087661743164.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,7
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is 2258.0986850261688, which exceeds threshold_ * scale, where
computed_gradient evaluates to -1.7880337238311768,
estimated_gradient evaluates to -2259.88671875, and
threshold_ * scale evaluates to 22.598867416381836.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,8
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is 2323.8749594688416, which exceeds threshold_ * scale, where
computed_gradient evaluates to -0.38895654678344727,
estimated_gradient evaluates to -2324.263916015625, and
threshold_ * scale evaluates to 23.242639541625977.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,9
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is 2386.3956198692322, which exceeds threshold_ * scale, where
computed_gradient evaluates to -2.2457375526428223,
estimated_gradient evaluates to -2388.641357421875, and
threshold_ * scale evaluates to 23.88641357421875.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,10
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is 2450.3957705497742, which exceeds threshold_ * scale, where
computed_gradient evaluates to -2.6227841377258301,
estimated_gradient evaluates to -2453.0185546875, and
threshold_ * scale evaluates to 24.530185699462891.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,11
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is 2517.0021805465221, which exceeds threshold_ * scale, where
computed_gradient evaluates to -0.3943038284778595,
estimated_gradient evaluates to -2517.396484375, and
threshold_ * scale evaluates to 25.17396354675293.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,12
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is 2581.223292350769, which exceeds threshold_ * scale, where
computed_gradient evaluates to -0.55063343048095703,
estimated_gradient evaluates to -2581.77392578125, and
threshold_ * scale evaluates to 25.817739486694336.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,13
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is 2645.8555200994015, which exceeds threshold_ * scale, where
computed_gradient evaluates to -0.295847088098526,
estimated_gradient evaluates to -2646.1513671875, and
threshold_ * scale evaluates to 26.461513519287109.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,14
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is 2708.9229011535645, which exceeds threshold_ * scale, where
computed_gradient evaluates to -1.6061515808105469,
estimated_gradient evaluates to -2710.529052734375, and
threshold_ * scale evaluates to 27.105289459228516.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,15
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is 2774.1294550299644, which exceeds threshold_ * scale, where
computed_gradient evaluates to -0.77703911066055298,
estimated_gradient evaluates to -2774.906494140625, and
threshold_ * scale evaluates to 27.749063491821289.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,16
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is 2838.3582728505135, which exceeds threshold_ * scale, where
computed_gradient evaluates to -0.92566269636154175,
estimated_gradient evaluates to -2839.283935546875, and
threshold_ * scale evaluates to 28.392839431762695.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,17
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is 2904.4807693362236, which exceeds threshold_ * scale, where
computed_gradient evaluates to 0.81939238309860229,
estimated_gradient evaluates to -2903.661376953125, and
threshold_ * scale evaluates to 29.036613464355469.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,18
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is 2968.9486039280891, which exceeds threshold_ * scale, where
computed_gradient evaluates to 0.90954142808914185,
estimated_gradient evaluates to -2968.0390625, and
threshold_ * scale evaluates to 29.680389404296875.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,19
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is 3031.4549759030342, which exceeds threshold_ * scale, where
computed_gradient evaluates to -0.96152800321578979,
estimated_gradient evaluates to -3032.41650390625, and
threshold_ * scale evaluates to 30.324163436889648.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,20
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is 3096.3903590738773, which exceeds threshold_ * scale, where
computed_gradient evaluates to -0.40334209799766541,
estimated_gradient evaluates to -3096.793701171875, and
threshold_ * scale evaluates to 30.967935562133789.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,21
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is 3161.3736805468798, which exceeds threshold_ * scale, where
computed_gradient evaluates to 0.20204968750476837,
estimated_gradient evaluates to -3161.171630859375, and
threshold_ * scale evaluates to 31.611715316772461.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,22
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is 3225.3918676525354, which exceeds threshold_ * scale, where
computed_gradient evaluates to -0.15720461308956146,
estimated_gradient evaluates to -3225.549072265625, and
threshold_ * scale evaluates to 32.255489349365234.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,23
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is 3290.1285521090031, which exceeds threshold_ * scale, where
computed_gradient evaluates to 0.20155015587806702,
estimated_gradient evaluates to -3289.927001953125, and
threshold_ * scale evaluates to 32.899269104003906.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,24
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is 3354.6566629111767, which exceeds threshold_ * scale, where
computed_gradient evaluates to 0.35246369242668152,
estimated_gradient evaluates to -3354.30419921875, and
threshold_ * scale evaluates to 33.543041229248047.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,25
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is 3417.3534735441208, which exceeds threshold_ * scale, where
computed_gradient evaluates to -1.3284112215042114,
estimated_gradient evaluates to -3418.681884765625, and
threshold_ * scale evaluates to 34.186817169189453.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,26
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is nan, which exceeds threshold_ * scale, where
computed_gradient evaluates to 2.0507621765136719,
estimated_gradient evaluates to -nan, and
threshold_ * scale evaluates to 0.020507620647549629.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,27
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is nan, which exceeds threshold_ * scale, where
computed_gradient evaluates to -0.14714376628398895,
estimated_gradient evaluates to -nan, and
threshold_ * scale evaluates to 0.0099999997764825821.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,28
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is nan, which exceeds threshold_ * scale, where
computed_gradient evaluates to -0.092538051307201385,
estimated_gradient evaluates to -nan, and
threshold_ * scale evaluates to 0.0099999997764825821.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,29
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is nan, which exceeds threshold_ * scale, where
computed_gradient evaluates to 1.4348338842391968,
estimated_gradient evaluates to -nan, and
threshold_ * scale evaluates to 0.014348338358104229.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,30
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is nan, which exceeds threshold_ * scale, where
computed_gradient evaluates to -1.8200397491455078,
estimated_gradient evaluates to -nan, and
threshold_ * scale evaluates to 0.018200397491455078.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,31
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is nan, which exceeds threshold_ * scale, where
computed_gradient evaluates to -0.63687032461166382,
estimated_gradient evaluates to -nan, and
threshold_ * scale evaluates to 0.0099999997764825821.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,32
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is nan, which exceeds threshold_ * scale, where
computed_gradient evaluates to -0.27529597282409668,
estimated_gradient evaluates to -nan, and
threshold_ * scale evaluates to 0.0099999997764825821.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,33
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is nan, which exceeds threshold_ * scale, where
computed_gradient evaluates to 1.6539856195449829,
estimated_gradient evaluates to -nan, and
threshold_ * scale evaluates to 0.016539854928851128.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,34
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is nan, which exceeds threshold_ * scale, where
computed_gradient evaluates to 1.559444785118103,
estimated_gradient evaluates to -nan, and
threshold_ * scale evaluates to 0.015594447962939739.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,35
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is nan, which exceeds threshold_ * scale, where
computed_gradient evaluates to -0.47840580344200134,
estimated_gradient evaluates to -nan, and
threshold_ * scale evaluates to 0.0099999997764825821.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,36
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is nan, which exceeds threshold_ * scale, where
computed_gradient evaluates to -2.1084423065185547,
estimated_gradient evaluates to -nan, and
threshold_ * scale evaluates to 0.021084422245621681.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,37
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is nan, which exceeds threshold_ * scale, where
computed_gradient evaluates to -0.64734846353530884,
estimated_gradient evaluates to -nan, and
threshold_ * scale evaluates to 0.0099999997764825821.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,38
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is nan, which exceeds threshold_ * scale, where
computed_gradient evaluates to 1.2536804676055908,
estimated_gradient evaluates to -nan, and
threshold_ * scale evaluates to 0.012536804191768169.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,39
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is nan, which exceeds threshold_ * scale, where
computed_gradient evaluates to 0.70361709594726562,
estimated_gradient evaluates to -nan, and
threshold_ * scale evaluates to 0.0099999997764825821.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,40
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is nan, which exceeds threshold_ * scale, where
computed_gradient evaluates to 0.28150025010108948,
estimated_gradient evaluates to -nan, and
threshold_ * scale evaluates to 0.0099999997764825821.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,41
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is nan, which exceeds threshold_ * scale, where
computed_gradient evaluates to 2.1712429523468018,
estimated_gradient evaluates to -nan, and
threshold_ * scale evaluates to 0.021712429821491241.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,42
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is nan, which exceeds threshold_ * scale, where
computed_gradient evaluates to -0.31444552540779114,
estimated_gradient evaluates to -nan, and
threshold_ * scale evaluates to 0.0099999997764825821.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,43
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is nan, which exceeds threshold_ * scale, where
computed_gradient evaluates to 0.21716280281543732,
estimated_gradient evaluates to -nan, and
threshold_ * scale evaluates to 0.0099999997764825821.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,44
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is nan, which exceeds threshold_ * scale, where
computed_gradient evaluates to -0.14974787831306458,
estimated_gradient evaluates to -nan, and
threshold_ * scale evaluates to 0.0099999997764825821.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,45
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is nan, which exceeds threshold_ * scale, where
computed_gradient evaluates to -0.66217285394668579,
estimated_gradient evaluates to -nan, and
threshold_ * scale evaluates to 0.0099999997764825821.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,46
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is nan, which exceeds threshold_ * scale, where
computed_gradient evaluates to 1.8779197931289673,
estimated_gradient evaluates to -nan, and
threshold_ * scale evaluates to 0.018779197707772255.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,47
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is nan, which exceeds threshold_ * scale, where
computed_gradient evaluates to -0.46108913421630859,
estimated_gradient evaluates to -nan, and
threshold_ * scale evaluates to 0.0099999997764825821.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,48
./src/caffe/test/test_gradient_check_util.hpp:163: Failure
The difference between computed_gradient and estimated_gradient is nan, which exceeds threshold_ * scale, where
computed_gradient evaluates to -0.88383305072784424,
estimated_gradient evaluates to -nan, and
threshold_ * scale evaluates to 0.0099999997764825821.
debug: (top_id, top_data_id, blob_id, feat_id)=-1,-1,0,49
[  FAILED  ] SoftmaxWithLossLayerTest/0.TestGradientGPU, where TypeParam = float (54 ms)
[----------] 2 tests from SoftmaxWithLossLayerTest/0 (284 ms total)

[----------] 2 tests from SoftmaxLayerTest/1, where TypeParam = double
[ RUN      ] SoftmaxLayerTest/1.TestGradientCPU
[       OK ] SoftmaxLayerTest/1.TestGradientCPU (285 ms)
[ RUN      ] SoftmaxLayerTest/1.TestForwardCPU
[       OK ] SoftmaxLayerTest/1.TestForwardCPU (0 ms)
[----------] 2 tests from SoftmaxLayerTest/1 (285 ms total)

[----------] 8 tests from SplitLayerInsertionTest
[ RUN      ] SplitLayerInsertionTest.TestNoInsertionImageNet
[       OK ] SplitLayerInsertionTest.TestNoInsertionImageNet (2 ms)
[ RUN      ] SplitLayerInsertionTest.TestNoInsertion2
[       OK ] SplitLayerInsertionTest.TestNoInsertion2 (0 ms)
[ RUN      ] SplitLayerInsertionTest.TestNoInsertion1
[       OK ] SplitLayerInsertionTest.TestNoInsertion1 (0 ms)
[ RUN      ] SplitLayerInsertionTest.TestInsertion
[       OK ] SplitLayerInsertionTest.TestInsertion (0 ms)
[ RUN      ] SplitLayerInsertionTest.TestInputInsertion
[       OK ] SplitLayerInsertionTest.TestInputInsertion (0 ms)
[ RUN      ] SplitLayerInsertionTest.TestInsertionTwoTop
[       OK ] SplitLayerInsertionTest.TestInsertionTwoTop (0 ms)
[ RUN      ] SplitLayerInsertionTest.TestWithInPlace
[       OK ] SplitLayerInsertionTest.TestWithInPlace (0 ms)
[ RUN      ] SplitLayerInsertionTest.TestNoInsertionWithInPlace
[       OK ] SplitLayerInsertionTest.TestNoInsertionWithInPlace (0 ms)
[----------] 8 tests from SplitLayerInsertionTest (2 ms total)

[----------] 4 tests from SyncedMemoryTest
[ RUN      ] SyncedMemoryTest.TestInitialization
[       OK ] SyncedMemoryTest.TestInitialization (0 ms)
[ RUN      ] SyncedMemoryTest.TestGPUWrite
[       OK ] SyncedMemoryTest.TestGPUWrite (1 ms)
[ RUN      ] SyncedMemoryTest.TestCPUWrite
[       OK ] SyncedMemoryTest.TestCPUWrite (0 ms)
[ RUN      ] SyncedMemoryTest.TestAllocation
[       OK ] SyncedMemoryTest.TestAllocation (0 ms)
[----------] 4 tests from SyncedMemoryTest (1 ms total)

[----------] 10 tests from LRNLayerTest/1, where TypeParam = double
[ RUN      ] LRNLayerTest/1.TestGPUGradientAcrossChannels
F1006 11:55:23.656641 14987 lrn_layer.cu:96] Check failed: error == cudaSuccess (8 vs. 0)  invalid device function
*** Check failure stack trace: ***
    @     0x7f1423f725cd  google::LogMessage::Fail()
    @     0x7f1423f74433  google::LogMessage::SendToLog()
    @     0x7f1423f7215b  google::LogMessage::Flush()
    @     0x7f1423f74e1e  google::LogMessageFatal::~LogMessageFatal()
    @           0x62c532  caffe::LRNLayer<>::CrossChannelForward_gpu()
    @           0x435391  caffe::Layer<>::Forward()
    @           0x437b74  caffe::LRNLayerTest_TestGPUGradientAcrossChannels_Test<>::TestBody()
    @           0x55dd53  testing::internal::HandleExceptionsInMethodIfSupported<>()
    @           0x5571ca  testing::Test::Run()
    @           0x557318  testing::TestInfo::Run()
    @           0x5573f5  testing::TestCase::Run()
    @           0x5586cf  testing::internal::UnitTestImpl::RunAllTests()
    @           0x5589f3  testing::UnitTest::Run()
    @           0x4155bf  main
    @     0x7f141fb56830  __libc_start_main
    @           0x41a629  _start
    @              (nil)  (unknown)
Makefile:279: recipe for target 'runtest' failed
make: *** [runtest] Aborted (core dumped)

Got a make error

/usr/bin/g++ build/tools/compute_volume_mean_from_list.o build/lib/libcaffe.a -o build/tools/compute_volume_mean_from_list.bin -pthread -fPIC -DNDEBUG -O2 -I/usr/local/include/python2.7 -I/usr/local/lib/python2.7/dist-packages/numpy/core/include -I/usr/local/include -Ibuild/src -I./src -I./include -I/usr/local/cuda/include -I/scail/scratch/u/zhangst/coaction/install/OpenBLAS_lib/include/ -L/usr/local/lib -L/usr/local/lib -L/usr/lib -L/usr/local/cuda/lib64 -L/usr/local/cuda/lib -L/scail/scratch/u/zhangst/coaction/install/OpenBLAS_lib/lib -lcudart -lcublas -lcurand -lpthread -lglog -lprotobuf -lleveldb -lsnappy -lboost_system -lhdf5_hl -lhdf5 -lopencv_core -lopencv_highgui -lopencv_imgproc -lopenblas -Wall /usr/bin/ld: /usr/local/lib/libhdf5.a(H5PL.c.o): undefined reference to symbol 'dlclose@@GLIBC_2.2.5' //lib/x86_64-linux-gnu/libdl.so.2: error adding symbols: DSO missing from command line collect2: error: ld returned 1 exit status make: *** [build/tools/compute_volume_mean_from_list.bin] Error 1

Any idea what may caused it? Appreciate the help :)

Hello, I have the exact same issue even after all the images are saved in "%06d.jpg".

Here the error looks like F0622 11:35:55.689235 24543 video_data_layer.cpp:346] Check failed: ReadImageSequenceToVolumeDatum(file_list_[id].c_str(), start_frm_list_[id], label_list_[id], new_length, new_height, new_width, sampling_rate, &datum)

Kindly assist!

I am trying to compute the mean volume of the dataset HMDB-51 and I run this code:

GLOG_logtostderr=1 ../../build/tools/compute_volume_mean_from_list.bin c3d_finetuning_HMDB51/train2.lst 16 128 171 1 hmdb51_train_mean.binaryproto 10

My train.lst looks like this:

c3d_finetuning_HMDB51/data/April_09_brush_hair_u_nm_np1_ba_goo_0.avi 1 0 c3d_finetuning_HMDB51/data/April_09_brush_hair_u_nm_np1_ba_goo_0.avi 17 0 c3d_finetuning_HMDB51/data/April_09_brush_hair_u_nm_np1_ba_goo_0.avi 33 0 c3d_finetuning_HMDB51/data/April_09_brush_hair_u_nm_np1_ba_goo_0.avi 49 0

I1021 15:24:30.223497 28438 compute_volume_mean_from_list.cpp:53] using dropping rate 10 I1021 15:24:30.223736 28438 compute_volume_mean_from_list.cpp:80] Starting Iteration E1021 15:24:30.296300 28438 compute_volume_mean_from_list.cpp:112] Processed 2872 files. I1021 15:24:30.296344 28438 compute_volume_mean_from_list.cpp:119] Write to hmdb51_train_mean.binaryproto

I get an output file hmdb51_train_mean.binaryproto which is of size 10Bytes.

I have checked the 2 suggestions related to issue-30:

1. The correctness of the relative paths
2. Opencv and ffmpeg compiles with the --enable-shared flag on.

I have tried both and yet the same empty file is returned.

P.S: I do not have any other problems with the code. I am able to extract the features as well.

Thanks in advance!

I run ucf101_testing.sh, when going to batch 29, there is an error : Check failed: read status Testing must not miss any example. Before that, I have successfully trained the network on UCF101 with your trainlist. It's ok. From batch 1to 29, I can get the accuracy normally. But going to Batch 30, the error appears. I am confused. So I seek help here, thank you! @dutran

I was successful in training my 3D dataset by giving 5D tensor ( 10x1x32x32x32) to the network. Now I'm using python script to load the trained model and deploy.prototxt and giving the input test dataset in the form of 5D tensor ( 10x1x32x32x32) but caffe outputs a ValueError saying "could not broadcast input array from shape (10,1,32,32,32) into shape (10,1,32,32). Isn't it the right way to do? why my model is expecting a 10,1,32,32 shape?

E0529 11:24:15.056433 13957 test_net.cpp:51] Batch 1373, accuracy: 0.8
E0529 11:24:16.577850 13957 test_net.cpp:51] Batch 1374, accuracy: 0.833333
E0529 11:24:18.104264 13957 test_net.cpp:51] Batch 1375, accuracy: 0.833333
E0529 11:24:19.628096 13957 test_net.cpp:51] Batch 1376, accuracy: 0.833333
E0529 11:24:21.147753 13957 test_net.cpp:51] Batch 1377, accuracy: 0.833333
E0529 11:24:22.675658 13957 test_net.cpp:51] Batch 1378, accuracy: 0.733333
E0529 11:24:24.198937 13957 test_net.cpp:51] Batch 1379, accuracy: 0.8
E0529 11:24:25.720679 13957 test_net.cpp:51] Batch 1380, accuracy: 0.8
E0529 11:24:27.248215 13957 test_net.cpp:51] Batch 1381, accuracy: 0.866667
E0529 11:24:28.769614 13957 test_net.cpp:51] Batch 1382, accuracy: 0.833333
E0529 11:24:30.297591 13957 test_net.cpp:51] Batch 1383, accuracy: 0.766667
E0529 11:24:30.297626 13957 test_net.cpp:54] Test accuracy: 0.786777

I run ucf101_testing.sh I change batch size from 50 to 30, so I have 1384 batches in total Test accuracy is 0.786777, is this ok?

I have a 3D dataset stored in HDF5 file, where dimensions are 32x32x32 and chunk size is 10. When I try to load this data using HDF5 data layer, caffe's log prints that It has loaded 10x32x1x1x32. Whereas it should have loaded 10x32x32x32 in a single batch (screen-shots attached). Can someone please guide me about this issue?

Hi Du Tran, I checked several closed issues for "make runtest" but still couldn't find similar problem. After doing make all and make test without error, I did "make runtest" but always got stuck at test point below. At first I thought it was because having two GPUs, then I do "make runtest CUDA_VISIBLE_DEVICES=0" or export CUDA_VISIBLE_DEVICES=0 before doing make runtest (based on here and here). However, i still got stuck at the same point. Do you have any idea why I experienced this issue? Thank you.

Here is the test point that I got stuck: [----------] 5 tests from Deconvolution3DLayerTest/1, where TypeParam = double [ RUN ] Deconvolution3DLayerTest/1.TestSetup [ OK ] Deconvolution3DLayerTest/1.TestSetup (0 ms) [ RUN ] Deconvolution3DLayerTest/1.TestCPUGradient

or [----------] 5 tests from Deconvolution3DLayerTest/1, where TypeParam = double [ RUN ] Deconvolution3DLayerTest/1.TestSetup [ OK ] Deconvolution3DLayerTest/1.TestSetup (0 ms) [ RUN ] Deconvolution3DLayerTest/1.TestGPUSimpleDeconvolution3D [ OK ] Deconvolution3DLayerTest/1.TestGPUSimpleDeconvolution3D (1 ms) [ RUN ] Deconvolution3DLayerTest/1.TestCPUSimpleDeconvolution3D [ OK ] Deconvolution3DLayerTest/1.TestCPUSimpleDeconvolution3D (1 ms) [ RUN ] Deconvolution3DLayerTest/1.TestGPUGradient

I managed to solve the problem.. Thank you anyway.

Try to extract features with default values on CPU and got this error: Do you have any idea about this error @dutran @puneet29

os.getcwd() /app/examples/c3d_feature_extraction
file exists
number of frames 347
file per sec 30.0
[Info] num_frames=347, fps=30.0
input_file /app/examples/c3d_feature_extraction/resource/output_tmp/input.txt 
output_prefix_file /app/examples/c3d_feature_extraction/resource/output_tmp/output_prefix.txt 
feature_prototxt /app/examples/c3d_feature_extraction/resource/output_tmp/feature_extraction.prototxt
[Warning] frame_dir=/app/examples/c3d_feature_extraction/resource/output_tmp/RoadAccidents002_x264 does exist. Will overwrite
[Info] Extracting frame num=165.5
[Info] Extracting frame num=166.5
[Info] Extracting frame num=167.5
[Info] Extracting frame num=168.5
[Info] Extracting frame num=169.5
[Info] Extracting frame num=170.5
[Info] Extracting frame num=171.5
[Info] Extracting frame num=172.5
[Info] Extracting frame num=173.5
[Info] Extracting frame num=174.5
[Info] Extracting frame num=175.5
[Info] Extracting frame num=176.5
[Info] Extracting frame num=177.5
[Info] Extracting frame num=178.5
[Info] Extracting frame num=179.5
[Info] Extracting frame num=180.5
The temp directory does exist: /app/examples/c3d_feature_extraction/resource/output_tmp
caffe_root /opt/caffe/
[Info] Running C3D feature extraction: "/opt/caffe/build/tools/extract_features /app/examples/c3d_feature_extraction/resource/output_tmp/feature_extraction.prototxt /app/examples/c3d_feature_extraction/conv3d_deepnetA_sport1m_iter_1900000 -1 5 9999999 /app/examples/c3d_feature_extraction/resource/output_tmp/output_prefix.txt fc6-1"
E0906 02:28:49.145242    14 extract_features.cpp:62] Using CPU
[libprotobuf ERROR google/protobuf/text_format.cc:274] Error parsing text-format caffe.NetParameter: 2:1: Invalid control characters encountered in text.
[libprotobuf ERROR google/protobuf/text_format.cc:274] Error parsing text-format caffe.NetParameter: 2:27: Invalid control characters encountered in text.
[libprotobuf ERROR google/protobuf/text_format.cc:274] Error parsing text-format caffe.NetParameter: 2:28: Interpreting non ascii codepoint 182.
[libprotobuf ERROR google/protobuf/text_format.cc:274] Error parsing text-format caffe.NetParameter: 2:28: Message type "caffe.NetParameter" has no field named "DeepConv3DNet_3x3_Sport1M".
F0906 02:28:49.169955    14 upgrade_proto.cpp:88] Check failed: ReadProtoFromTextFile(param_file, param) Failed to parse NetParameter file: /app/examples/c3d_feature_extraction/conv3d_deepnetA_sport1m_iter_1900000
*** Check failure stack trace: ***
    @     0x7f27ed8885cd  google::LogMessage::Fail()
    @     0x7f27ed88a433  google::LogMessage::SendToLog()
    @     0x7f27ed88815b  google::LogMessage::Flush()
    @     0x7f27ed88ae1e  google::LogMessageFatal::~LogMessageFatal()
    @     0x7f27ede12891  caffe::ReadNetParamsFromTextFileOrDie()
    @     0x7f27eddfdc82  caffe::Net<>::Net()
    @           0x40b176  feature_extraction_pipeline<>()
    @     0x7f27ec9c0830  __libc_start_main
    @           0x405b79  _start
    @              (nil)  (unknown)
[Error] feature extraction failed!
Done in 27.59s.

The Caffe installation should be fine, as I can run the feature extraction code here, and this is the log:

root@045c38549154:/opt/caffe# ./scripts/download_model_binary.py models/bvlc_reference_caffenet
...100%, 232 MB, 578 KB/s, 411 seconds passedroot@045c38549154:/opt/caffe# mkdir ./examples/_temp
root@045c38549154:/opt/caffe# find pwd/examples/images -type f -exec echo {} \; > examples/_temp/temp.txt
root@045c38549154:/opt/caffe# sed "s/$/ 0/" examples/_temp/temp.txt > examples/_temp/file_list.txt
root@045c38549154:/opt/caffe# ./data/ilsvrc12/get_ilsvrc_aux.sh
Downloading...
--2019-09-06 02:56:21--  http://dl.caffe.berkeleyvision.org/caffe_ilsvrc12.tar.gz
Resolving dl.caffe.berkeleyvision.org (dl.caffe.berkeleyvision.org)... 169.229.222.251
Connecting to dl.caffe.berkeleyvision.org (dl.caffe.berkeleyvision.org)|169.229.222.251|:80... connected.
HTTP request sent, awaiting response... 200 OK
Length: 17858008 (17M) [application/octet-stream]
Saving to: caffe_ilsvrc12.tar.gz

caffe_ilsvrc12.tar.gz   100%[===============================>]  17.03M   500KB/s    in 33s     

2019-09-06 02:56:55 (528 KB/s) - caffe_ilsvrc12.tar.gz saved [17858008/17858008]

Unzipping...
Done.
root@045c38549154:/opt/caffe# cp examples/feature_extraction/imagenet_val.prototxt examples/_temp
root@045c38549154:/opt/caffe# ./build/tools/extract_features models/bvlc_reference_caffenet/bvlc_reference_caffenet.caffemodel examples/_temp/imagenet_val.prototxt fc6 examples/_temp/features 10 leveldb
E0906 02:59:06.618793    27 extract_features.cpp:62] Using CPU
E0906 02:59:07.137260    27 extract_features.cpp:133] Extracting Features
E0906 02:59:49.265465    27 extract_features.cpp:176] Extracted features of 500 query images for feature blob fc6
E0906 02:59:49.265573    27 extract_features.cpp:181] Successfully extracted the features!

I have caffe version 1.0.0 installed and it compiled correctly with video_data_layer included. When I run c3d_resnet18_sports1m_r2_iter_2800000.caffemodel for feature extraction I get the error for Video Data layer, which is included in the prototxt file. The error trace is:

F0905 11:26:43.151420 12241 layer_factory.hpp:81] Check failed: registry.count(type) == 1 (0 vs. 1) Unknown layer type: VideoData (known types: AbsVal, Accuracy, ArgMax, BNLL, BatchNorm, BatchReindex, Bias, BlackHole, Concat, ContrastiveLoss, Convolution, Convolution3D, Crop, Data, Deconvolution, Deconvolution3D, Dropout, DummyData, ELU, Eltwise, Embed, EuclideanLoss, Exp, Filter, Flatten, HDF5Data, HDF5Output, HingeLoss, Im2col, InfogainLoss, InnerProduct, Input, L2Normalize, LRN, LSTM, LSTMUnit, Log, MVN, MemoryData, MultinomialLogisticLoss, PReLU, Parameter, Pooling, Pooling3D, Power, Python, RNN, ReLU, Reduction, Reshape, SPP, Scale, Sigmoid, SigmoidCrossEntropyLoss, Silence, Slice, Softmax, SoftmaxWithLoss, Split, TanH, Threshold, Tile) *** Check failure stack trace: *** @ 0x7fd85fc320cd google::LogMessage::Fail() @ 0x7fd85fc33f33 google::LogMessage::SendToLog() @ 0x7fd85fc31c28 google::LogMessage::Flush() @ 0x7fd85fc34999 google::LogMessageFatal::~LogMessageFatal() @ 0x7fd86004d3b4 caffe::Net<>::Init() @ 0x7fd86004ef37 caffe::Net<>::Net() @ 0x55ece66bb679 feature_extraction_pipeline<>() @ 0x7fd85f2b5b97 __libc_start_main @ 0x55ece66badda _start Aborted (core dumped)

Please help me resolve this issue, I've been at it for days. Thank you.

I am trying to train UCF-101 dataset on the C3D-v1.1 but i face a weird error. The frame which is being accessed is one number more than the last frame as shown below:

I0729 16:32:09.606221 526955 solver.cpp:273] Solving C3D-UCF101Net I0729 16:32:09.606225 526955 solver.cpp:274] Learning Rate Policy: step E0729 16:32:09.634727 526978 image_io.cpp:339] Could not open or find file UCF-101/Hammering/v_Hammering_g13_c07//000020.jpg I0729 16:32:09.642668 526955 blocking_queue.cpp:49] Waiting for data E0729 16:32:09.661631 526978 image_io.cpp:339] Could not open or find file UCF-101/Skijet/v_Skijet_g19_c03//000044.jpg E0729 16:32:09.716790 526978 image_io.cpp:339] Could not open or find file UCF-101/RockClimbingIndoor/v_RockClimbingIndoor_g09_c01//000052.jpg E0729 16:32:09.747853 526978 image_io.cpp:339] Could not open or find file UCF-101/StillRings/v_StillRings_g18_c01//000044.jpg E0729 16:32:09.804396 526978 image_io.cpp:339] Could not open or find file UCF-101/Bowling/v_Bowling_g19_c01//000025.jpg E0729 16:32:09.822487 526978 image_io.cpp:339] Could not open or find file UCF-101/SoccerJuggling/v_SoccerJuggling_g19_c05//000052.jpg E0729 16:32:09.889794 526978 image_io.cpp:339] Could not open or find file UCF-101/WallPushups/v_WallPushups_g19_c02//000019.jpg E0729 16:32:09.964848 526978 image_io.cpp:339] Could not open or find file UCF-101/PlayingTabla/v_PlayingTabla_g17_c02//000054.jpg E0729 16:32:10.011976 526978 image_io.cpp:339] Could not open or find file UCF-101/HulaHoop/v_HulaHoop_g17_c03//000022.jpg E0729 16:32:10.023752 526978 image_io.cpp:339] Could not open or find file UCF-101/CleanAndJerk/v_CleanAndJerk_g12_c07//000032.jpg E0729 16:32:10.075891 526978 image_io.cpp:339] Could not open or find file UCF-101/RockClimbingIndoor/v_RockClimbingIndoor_g08_c05//000098.jpg E0729 16:32:10.137146 526978 image_io.cpp:339] Could not open or find file UCF-101/CuttingInKitchen/v_CuttingInKitchen_g13_c01//000031.jpg E0729 16:32:10.144898 526978 image_io.cpp:339] Could not open or find file UCF-101/WritingOnBoard/v_WritingOnBoard_g20_c07//000024.jpg E0729 16:32:10.206794 526978 image_io.cpp:339] Could not open or find file UCF-101/HandstandWalking/v_HandstandWalking_g09_c02//000031.jpg E0729 16:32:10.244976 526978 image_io.cpp:339] Could not open or find file UCF-101/ApplyEyeMakeup/v_ApplyEyeMakeup_g12_c03//000030.jpg E0729 16:32:10.256954 526978 image_io.cpp:339] Could not open or find file UCF-101/Haircut/v_Haircut_g09_c01//000046.jpg E0729 16:32:10.395485 526978 image_io.cpp:339] Could not open or find file UCF-101/PlayingSitar/v_PlayingSitar_g17_c07//000094.jpg E0729 16:32:10.517381 526978 image_io.cpp:339] Could not open or find file UCF-101/RockClimbingIndoor/v_RockClimbingIndoor_g17_c02//000083.jpg I0729 16:33:49.253451 526955 solver.cpp:219] Iteration 0 (0 iter/s, 99.647s/20 iters), loss = 4.53648 I0729 16:33:49.253582 526955 solver.cpp:238] Train net output #0: loss = 4.53648 (* 1 = 4.53648 loss) I0729 16:33:49.253629 526955 sgd_solver.cpp:105] Iteration 0, lr = 0.01 E0729 16:33:49.530026 526978 image_io.cpp:339] Could not open or find file UCF-101/Archery/v_Archery_g13_c07//000040.jpg E0729 16:33:49.591573 526978 image_io.cpp:339] Could not open or find file UCF-101/PlayingFlute/v_PlayingFlute_g05_c01//000066.jpg E0729 16:33:49.605487 526978 image_io.cpp:339] Could not open or find file UCF-101/ParallelBars/v_ParallelBars_g15_c04//000033.jpg E0729 16:33:49.616542 526978 image_io.cpp:339] Could not open or find file UCF-101/PlayingTabla/v_PlayingTabla_g13_c02//000060.jpg E0729 16:33:49.678442 526978 image_io.cpp:339] Could not open or find file UCF-101/PlayingDaf/v_PlayingDaf_g08_c03//000063.jpg E0729 16:33:49.690623 526978 image_io.cpp:339] Could not open or find file UCF-101/PommelHorse/v_PommelHorse_g09_c01//000043.jpg

For example, in the path UCF-101/PommelHorse/v_PommelHorse_g09_c01/ the total frames that are present are 42 but frame number 43 is being opened. This is similar to all the errors.

Please tell me how to solve this issue as I am new to Caffe.

I am currently working on a computer vision problem statement for which I use C3D on CPU only mode. It works great, but feature extraction on my dataset of videos (1900 videos) takes 4 to 5 days. I want to now build a real time prototype of my model. Could you suggest minimum GPU configuration for this to be possible? By real-time, I mean 1 sec of a 240x320 video should be processed within much lesser than 1 sec so that further processing can be done on features to send out real time alerts. Thanks in advance.

I am getting this error while running "make" in C3D-v1.1, and got the similar error even in 1.0. Please let me know if anyone have encountered the same and ways to fix it.

Full Error from here:

CXX .build_release/src/caffe/proto/caffe.pb.cc In file included from /usr/include/c++/5/mutex:35:0, from /usr/local/include/google/protobuf/stubs/mutex.h:33, from /usr/local/include/google/protobuf/stubs/common.h:52, from .build_release/src/caffe/proto/caffe.pb.h:9, from .build_release/src/caffe/proto/caffe.pb.cc:4: /usr/include/c++/5/bits/c++0x_warning.h:32:2: error: #error This file requires compiler and library support for the ISO C++ 2011 standard. This support must be enabled with the -std=c++11 or -std=gnu++11 compiler options. #error This file requires compiler and library support
^ In file included from /usr/local/include/google/protobuf/stubs/common.h:52:0, from .build_release/src/caffe/proto/caffe.pb.h:9, from .build_release/src/caffe/proto/caffe.pb.cc:4: /usr/local/include/google/protobuf/stubs/mutex.h:58:8: error: ‘mutex’ in namespace ‘std’ does not name a type std::mutex mu_; ^ /usr/local/include/google/protobuf/stubs/mutex.h: In member function ‘void google::protobuf::internal::WrappedMutex::Lock()’: /usr/local/include/google/protobuf/stubs/mutex.h:51:17: error: ‘mu_’ was not declared in this scope void Lock() { mu_.lock(); } ^ /usr/local/include/google/protobuf/stubs/mutex.h: In member function ‘void google::protobuf::internal::WrappedMutex::Unlock()’: /usr/local/include/google/protobuf/stubs/mutex.h:52:19: error: ‘mu_’ was not declared in this scope void Unlock() { mu_.unlock(); } ^ /usr/local/include/google/protobuf/stubs/mutex.h: At global scope: /usr/local/include/google/protobuf/stubs/mutex.h:61:7: error: expected nested-name-specifier before ‘Mutex’ using Mutex = WrappedMutex; ^ /usr/local/include/google/protobuf/stubs/mutex.h:66:28: error: expected ‘)’ before ‘*’ token explicit MutexLock(Mutex *mu) : mu_(mu) { this->mu_->Lock(); } ^ /usr/local/include/google/protobuf/stubs/mutex.h:69:3: error: ‘Mutex’ does not name a type Mutex const mu_; ^ /usr/local/include/google/protobuf/stubs/mutex.h: In destructor ‘google::protobuf::internal::MutexLock::~MutexLock()’: /usr/local/include/google/protobuf/stubs/mutex.h:67:24: error: ‘class google::protobuf::internal::MutexLock’ has no member named ‘mu_’ ~MutexLock() { this->mu_->Unlock(); } ^ /usr/local/include/google/protobuf/stubs/mutex.h: At global scope: /usr/local/include/google/protobuf/stubs/mutex.h:80:33: error: expected ‘)’ before ‘’ token explicit MutexLockMaybe(Mutex *mu) : ^ In file included from /usr/local/include/google/protobuf/arena.h:48:0, from .build_release/src/caffe/proto/caffe.pb.h:23, from .build_release/src/caffe/proto/caffe.pb.cc:4: /usr/include/c++/5/typeinfo:39:37: error: expected ‘}’ before end of line /usr/include/c++/5/typeinfo:39:37: error: expected unqualified-id before end of line /usr/include/c++/5/typeinfo:39:37: error: expected ‘}’ before end of line /usr/include/c++/5/typeinfo:39:37: error: expected ‘}’ before end of line /usr/include/c++/5/typeinfo:39:37: error: expected ‘}’ before end of line /usr/include/c++/5/typeinfo:39:37: error: expected declaration before end of line Makefile:588: recipe for target '.build_release/src/caffe/proto/caffe.pb.o' failed make: *** [.build_release/src/caffe/proto/caffe.pb.o] Error 1