- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I ran the example of mobilenet V1 as the following instructions:
https://github.com/movidius/ncappzoo/tree/master/tensorflow/mobilenets
But when I ran 'make', the error happend:
meom@meom:~/PycharmProjects/shenfang/NCScodes/ncappzoo/tensorflow/mobilenets$ make
TF_SRC_PATH is /home/meom/PycharmProjects/shenfang/NCScodes/ncappzoo/tensorflow/tensorflow
Downloading checkpoint files…
(mkdir -p model)
(cd model; wget -nc http://download.tensorflow.org/models/mobilenet_v1_1.0_224_2017_06_14.tar.gz;)
File ‘mobilenet_v1_1.0_224_2017_06_14.tar.gz’ already there; not retrieving.
(cd model; tar -xvf mobilenet_v1_1.0_224_2017_06_14.tar.gz;)
mobilenet_v1_1.0_224.ckpt.data-00000-of-00001
mobilenet_v1_1.0_224.ckpt.meta
mobilenet_v1_1.0_224.ckpt.index
TF_MODELS_PATH is /home/meom/PycharmProjects/shenfang/NCScodes/ncappzoo/tensorflow/models
Exporting GraphDef file…
(cd model; python3 /home/meom/PycharmProjects/shenfang/NCScodes/ncappzoo/tensorflow/models/research/slim/export_inference_graph.py \
--alsologtostderr \
--model_name=mobilenet_v1 \
--batch_size=1 \
--dataset_name=imagenet \
--image_size=224 \
--output_file=mobilenet_v1_1.0_224.pb;)
/usr/local/lib/python3.5/dist-packages/h5py/init.py:34: FutureWarning: Conversion of the second argument of issubdtype from
float
to np.floating
is deprecated. In future, it will be treated as np.float64 == np.dtype(float).type
.from ._conv import register_converters as _register_converters
INFO:tensorflow:Scale of 0 disables regularizer.
Freezing model for inference…
(cd model; python3 /home/meom/PycharmProjects/shenfang/NCScodes/ncappzoo/tensorflow/tensorflow/tensorflow/python/tools/freeze_graph.py \
--input_graph=mobilenet_v1_1.0_224.pb \
--input_binary=true \
--input_checkpoint=mobilenet_v1_1.0_224.ckpt \
--output_graph=mobilenet_v1_1.0_224_frozen.pb \
--output_node_name=MobilenetV1/Predictions/Reshape_1;)
/usr/local/lib/python3.5/dist-packages/h5py/init.py:34: FutureWarning: Conversion of the second argument of issubdtype from
float
to np.floating
is deprecated. In future, it will be treated as np.float64 == np.dtype(float).type
.from ._conv import register_converters as _register_converters
2019-01-03 10:47:32.615937: I tensorflow/core/platform/cpu_feature_guard.cc:137] Your CPU supports instructions that this TensorFlow binary was not compiled to use: SSE4.1 SSE4.2 AVX AVX2 FMA
2019-01-03 10:47:32.698358: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:892] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2019-01-03 10:47:32.698634: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1030] Found device 0 with properties:
name: GeForce GTX 1080 Ti major: 6 minor: 1 memoryClockRate(GHz): 1.62
pciBusID: 0000:01:00.0
totalMemory: 10.91GiB freeMemory: 10.03GiB
2019-01-03 10:47:32.698648: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1120] Creating TensorFlow device (/device:GPU:0) -> (device: 0, name: GeForce GTX 1080 Ti, pci bus id: 0000:01:00.0, compute capability: 6.1)
Converted 137 variables to const ops.
Profiling the model…
(cd model; mvNCProfile -s 12 mobilenet_v1_1.0_224_frozen.pb -in=input -on=MobilenetV1/Predictions/Reshape_1;)
mvNCProfile v02.00, Copyright @ Intel Corporation 2017
/usr/local/lib/python3.5/dist-packages/tensorflow/python/framework/tensor_util.py:509: DeprecationWarning: The binary mode of fromstring is deprecated, as it behaves surprisingly on unicode inputs. Use frombuffer instead
shape: (1, 224, 224, 3)
res.shape: (1, 1001)
TensorFlow output shape: (1, 1, 1001)
/usr/local/bin/ncsdk/Controllers/FileIO.py:65: UserWarning: You are using a large type. Consider reducing your data sizes for best performance
Blob generated
W: [ 0] ncDeviceOpen:528 ncDeviceOpen() XLinkBootRemote returned error 3
[Error 7] Toolkit Error: USB Failure. Code: Error opening device
Makefile:96: recipe for target 'profile' failed
make: *** [profile] Error 255
#
Please help me to solve this problem. I also cannot run 'make examples' after the installation.
- Tags:
- Tensorflow
Link Copied
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I found it works in NCS1 but in NCS2, the error happened.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I found it works in NCS1 but in NCS2, the error happened.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I found it works in NCS1 but in NCS2, the error happened.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I found it works in NCS1 but in NCS2, the error happened.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi @Meow10
Thank you for reaching out! Could you please provide additional information about your setup?
- Are you using the NCSDK1 or NCSDK2? (version 1.12, 2.08 or older)
- Which branch did you clone of the ncappzoo repository? (ncsdk2 or master)
- Are you using the original Neural Compute Stick or the Neural Compute Stick 2?
- Is the device listed when running lsusb?
- Are you running Ubuntu as a virtual machine?
- Have you tried connecting the device to a powered USB Hub or a different USB 3.0 port?
Regards,
Jesus
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
@Jesus_at_Intel The following are the answers to your questions:
- NCSDK2 2.08.01.02
- master
- the Neural Compute Stick 2
- The device is listed when running Isusb.
- Ubuntu
- I haven't tried.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi @Meow10
You will need to use the OpenVINO toolkit with the Intel Neural Compute Stick 2 as the NCSDK only support the original Neural Compute Stick.
Take a look at the Get Started Guide.
Hope this helps!
Regards,
Jesus
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page