Testing ncsdk for individual CNN

[idata]

I test individual example (e.g. GoogleLeNet) by running "make all" inside ncsdk/examples/caffe/GoogleLeNet. The output is as below. In overall, there are two error with code index: 7 and 25. I am skeptical those errors involving my USB port which must be USB3.0. However, if I run "run.py" script, it seems that Neuron USB stick work properly. I check usb support on my machine (using lsusb) and it seems that my computer supports USB 3.0 and also recognize Neuron stick (03e7:2150).

My question here is if I am doing somthing wrong when plugging Neuron stick or my installation is incorrect !

/Tung

//**** Results after running "lsusb"

[QUOTE]Bus 002 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub*_Bus 001 Device 022: ID 03e7:2150_*Bus 001 Device 007: ID 05e3:0748 Genesys Logic, Inc.Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub[/1QUOTE]

//**** Results after running ./run.py for GoogleLeNet

[QUOTE]Device 0 Address: 8 - VID/PID 03e7:2150Starting wait for connect with 2000ms timeoutFound Address: 8 - VID/PID 03e7:2150Found EP 0x81 : max packet size is 512 bytesFound EP 0x01 : max packet size is 512 bytesFound and opened devicePerforming bulk write of 830212 bytes…Successfully sent 830212 bytes of data in 70.770983 ms (11.187521 MB/s)Boot successful, device address 8Found Address: 8 - VID/PID 03e7:f63bdoneBooted 8 -> VSC

------- predictions --------

prediction 0 (probability 0.99609) is n03272010 electric guitar label index is: 546prediction 1 (probability 0.0035095) is n02676566 acoustic guitar label index is: 402prediction 2 (probability 0.0) is n02396427 wild boar, boar, Sus scrofa label index is: 342prediction 3 (probability 0.0) is n02391049 zebra label index is: 340prediction 4 (probability 0.0) is n02389026 sorrel label index is: 339[/QUOTE]

//**** Result of "make all" ****

[QUOTE]\nmaking prereqs(cd ../../data/ilsvrc12; make)make[1]: Entering directory /home/hoangt/WORK/TOOLS/DeepLearning/movidious/ncsdk/examples/data/ilsvrc12' make[1]: Leaving directory/home/hoangt/WORK/TOOLS/DeepLearning/movidious/ncsdk/examples/data/ilsvrc12'\nmaking prototxtDownloading Prototxt file--2017-11-22 00:30:09-- https://raw.githubusercontent.com/BVLC/caffe/master/models/bvlc_googlenet/deploy.prototxtResolving raw.githubusercontent.com (raw.githubusercontent.com)… 151.101.40.133Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|151.101.40.133|:443… connected.HTTP request sent, awaiting response… 200 OKLength: 35862 (35K) [text/plain]Saving to: ‘./deploy.prototxt’

100%[==============================================================================================>] 35,862 --.-K/s in 0.006s

2017-11-22 00:30:09 (5.84 MB/s) - ‘./deploy.prototxt’ saved [35862/35862]

Adding input shape to prototxt file.

\nmaking profilemvNCProfile deploy.prototxt -s 12mvNCProfile v02.00, Copyright @ Movidius Ltd 2016

__ WARNING: using empty weights __

Layer inception_3b/1x1 forced to im2col_v2, because its output is used in concatLayer inception_3b/pool_proj forced to im2col_v2, because its output is used in concat……./usr/local/bin/ncsdk/Controllers/FileIO.py:52: UserWarning: You are using a large type. Consider reducing your data sizes for best performance "Consider reducing your data sizes for best performance\033[0m")## [Error 7] Toolkit Error: USB Failure. Code: Error opening device\nmaking caffemodelDownloading caffemodel file--2017-11-22 00:30:11-- http://dl.caffe.berkeleyvision.org/bvlc_googlenet.caffemodelResolving dl.caffe.berkeleyvision.org (dl.caffe.berkeleyvision.org)… 169.229.222.251Connecting to dl.caffe.berkeleyvision.org (dl.caffe.berkeleyvision.org)|169.229.222.251|:80… connected.HTTP request sent, awaiting response… 200 OKLength: 53533754 (51M) [application/octet-stream]Saving to: ‘./bvlc_googlenet.caffemodel’

100%[==============================================================================================>] 53,533,754 6.61MB/s in 7.5s

2017-11-22 00:30:19 (6.79 MB/s) - ‘./bvlc_googlenet.caffemodel’ saved [53533754/53533754]

\nmaking check

mvNCCheck -w bvlc_googlenet.caffemodel -i ../../data/images/nps_electric_guitar.png -s 12 -id 546 deploy.prototxt -S 255 -M 110mvNCCheck v02.00, Copyright @ Movidius Ltd 2016

/usr/local/lib/python3.4/dist-packages/skimage/transform/_warps.py:84: UserWarning: The default mode, 'constant', will be changed to 'reflect' in skimage 0.15.

warn("The default mode, 'constant', will be changed to 'reflect' in "Layer inception_3b/1x1 forced to im2col_v2, because its output is used in concatLayer inception_3b/pool_proj forced to im2col_v2, because its output is used in concat…../usr/local/bin/ncsdk/Controllers/FileIO.py:52: UserWarning: You are using a large type. Consider reducing your data sizes for best performance "Consider reducing your data sizes for best performance\033[0m")USB: Transferring Data…[Error 25] Myriad Error: "mvncStatus.TIMEOUT".\nmaking compilemvNCCompile -w bvlc_googlenet.caffemodel -s 12 deploy.prototxtmvNCCompile v02.00, Copyright @ Movidius Ltd 2016

Layer inception_3b/1x1 forced to im2col_v2, because its output is used in concat

Layer inception_3b/pool_proj forced to im2col_v2, because its output is used in concat…../usr/local/bin/ncsdk/Controllers/FileIO.py:52: UserWarning: You are using a large type. Consider reducing your data sizes for best performance "Consider reducing your data sizes for best performance\033[0m")\nmaking cppg++ cpp/run.cpp cpp/fp16.c -o cpp/run_cpp -lmvnc[/QUOTE]

[idata]

It seems to be an issue with a malloc library.

if you do that (ubuntu 16.04)

sudo apt-get install libtcmalloc-minimal4

export LD_PRELOAD="/usr/lib/libtcmalloc_minimal.so.4"

then

make examples

I should be ok

[idata]

Hi ,

After I installed the package of "libtcmalloc-minimal4" , and export LD_PRELOAD variable,I still got the same warning message:/usr/local/bin/ncsdk/Controllers/FileIO.py:52: UserWarning: You are using a large type. Consider reducing your data sizes for best performance"Consider reducing your data sizes for best performance\033[0m").

And it always occurs the exception error when run image-classifier.py

Traceback (most recent call last): File "image-classifier.py", line 45, in graph = device.AllocateGraph( blob ) File "/usr/local/lib/python3.5/dist-packages/mvnc/mvncapi.py", line 201, in AllocateGraph raise Exception(Status(status))Exception: mvncStatus.ERROR

Could you help teach me to fix the issue? thanks.

ps. I plug into NCS stick USB2.0 port, run on 16.04 system.

regards,

SamHsieh

[idata]

@samhsieh Usually similar problems are observed when using virtual machines. Is that your case or you are running native ubuntu 16.04 installation? Also you can try using USB hub (powered) or try different port on your computer and see if any difference.

[idata]

Hi georgievm_cms,

Thanks for your kind reminder.

Yes, it is true that it runs on VM environment.Though I could open/closed NCS device successfully. It seems that FileIO is not compatible with on VM environment.Anyway, I try to install the "NCSDK" into Raspberry Pi3.