Error in stream_infer.py

[idata]

Hi,

I had a error message in stream_infer.py of ncappzoo/apps/stream_infer folder.

The stream_infer.py was from

https://github.com/movidius/ncappzoo.My operation

In ncappzoo/apps/stream_infer,

$ make run

The message are blow.

making run python3 stream_infer.py Traceback (most recent call last): File "stream_infer.py", line 13, in <module> gi.require_version('Gst', '1.0') File "/usr/lib/python3/dist-packages/gi/__init__.py", line 118, in require_version raise ValueError('Namespace %s not available' % namespace) ValueError: Namespace Gst not available Makefile:12: recipe for target 'run' failed make: *** [run] Error 1

Please give me any advice to solve the problem.

My Environment

Raspberry PI3, Raspbian stretch with desktop

Movidius NC SDK version: 1.10.01.00

$ uname -a

Linux raspberrypi 4.9.59-v7+ #1047 SMP Sun Oct 29 12:19:23 GMT 2017 armv7l GNU/Linux

I already checked that camera was working by raspivid and omxplayer command.

And I refered that plugin information

https://github.com/movidius/ncappzoo/blob/master/apps/stream_infer/readme.md

Followings are dpkg results.

$ dpkg -l gstreamer1.0-lib*ii gstreamer1.0-libav:armhf 1.10.4-1 armhf

$ dpkg -l gstreamer1.0-plug*

ii gstreamer1.0-plugins-bad:armhf 1.10.4-1 armhf GStreamer plugins from the "bad" setii gstreamer1.0-plugins-base:armhf 1.10.4-1 armhf GStreamer plugins from the "base" setii gstreamer1.0-plugins-good:armhf 1.10.4-1 armhf GStreamer plugins from the "good" setun gstreamer1.0-plugins-ugly (No description)

Thanks

[idata]

You are getting further than me. I am getting an error: Makefile:7: recipe for target 'squeezenet' failed

[idata]

try

sudo apt-get install gstreamer1.0-tools python-gst-1.0

[idata]

Hi, ChazyD

I tried your comment. The above error message was no longer displayed.Thank you very much.

[idata]

Hi,

I had a error message in stream_infer.py of ncappzoo/apps/stream_infer folder when i ran the app on Pi3.

It could recognized stream image and show object & accuracy at first a few minutes.

After it occurred the exception error, the stream data still could preview image, but the another "Predict" thread will be dead.

Could you help solve the issue or give me some suggestion to look into the issue ? Thanks.

BR,

SamHsieh

Log message

Found stale device, resetting Device 0 Address: 1.2.2.2 - VID/PID 03e7:2150 ./stream_infer.py:335: DeprecationWarning: Gtk.Widget.set_double_buffered is deprecated gDrawAreaSink.set_double_buffered(True) Starting wait for connect with 2000ms timeout Found Address: 1.2.2.2 - VID/PID 03e7:2150 Found EP 0x81 : max packet size is 512 bytes Found EP 0x01 : max packet size is 512 bytes Found and opened device Performing bulk write of 865212 bytes… Successfully sent 865212 bytes of data in 102.927434 ms (8.016623 MB/s) Boot successful, device address 1.2.2.2 Found Address: 1.2.2.2 - VID/PID 03e7:f63b done Booted 1.2.2.2 -> VSC Exception in thread Thread-1: Traceback (most recent call last): File "/usr/lib/python3.5/threading.py", line 914, in _bootstrap_inner self.run() File "/usr/lib/python3.5/threading.py", line 862, in run self._target(*self._args, **self._kwargs) File "./stream_infer.py", line 168, in input_thread gGraph.LoadTensor(preprocessed_image_buf ,"frame %s" % frame_number) File "/usr/local/lib/python3.5/dist-packages/mvnc/mvncapi.py", line 251, in LoadTensor raise Exception(Status(status)) Exception: mvncStatus.ERROR

[idata]

@samhsieh I had a similar problem and what solved it for me was recompiling the graph file for the network being used by using the mvNCCompile command. With each NCSDK release, the graph files for each network must be recompiled with that same NCSDK version.

[idata]

Hi Tome,

Thanks for your suggestion.

I confirmed the latest NCSDK (Ver.1.11.00.02), the graph files for the network which be compiled the same NCSDK version.Anyway, I try to rebuild the graph file for "squeezenet" again.

[idata]

@Tome_at_Intel,

Thanks for your instructions. I had verified successfully the "stream_infer" app without exception.

BTW, the "street_cam" APP also recognizes accurately the images, including car, bus, motorbike, person, and so on.