MobileNetssd error

idata · ‎05-30-2018

first layer output no problem

mvNCCheck -w MobileNetSSD_deploy_compress.caffemodel -is 416 416 -i "/home/he/deeplearn/caffe/mobilev2ssd/caffe-ssd_3/data/Peddevkit/PED/JPEGImages/1527485592703wc-20180521-000232.jpg" MobileNetSSD_deploy_compress.prototxt -on conv1 -cs 0,1,2

mvNCCheck v02.00, Copyright @ Movidius Ltd 2016

/usr/local/lib/python3.5/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 "

/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…

USB: Myriad Execution Finished

USB: Myriad Connection Closing.

USB: Myriad Connection Closed.

Result: (64, 208, 208)

1) 1345170 69.4

2) 1380160 67.6

3) 1344990 67.4

4) 1345198 66.9

5) 1381180 66.5

Expected: (64, 208, 208)

1) 1345170 69.25

2) 1380160 67.7

3) 1344990 67.4

4) 1345198 66.94

5) 1381180 66.44

Obtained values

Obtained Min Pixel Accuracy: 0.25947652757167816% (max allowed=2%), Pass

Obtained Average Pixel Accuracy: 0.007905358506832272% (max allowed=1%), Pass

Obtained Percentage of wrong values: 0.0% (max allowed=0%), Pass

Obtained Pixel-wise L2 error: 0.014139951858627453% (max allowed=1%), Pass

Obtained Global Sum Difference: 15158.2119140625

but second layer output error

mvNCCheck -w MobileNetSSD_deploy_compress.caffemodel -is 416 416 -i "/home/he/deeplearn/caffe/mobilev2ssd/caffe-ssd_3/data/Peddevkit/PED/JPEGImages/1527485592703wc-20180521-000232.jpg" MobileNetSSD_deploy_compress.prototxt -on conv2 -cs 0,1,2

mvNCCheck v02.00, Copyright @ Movidius Ltd 2016

/usr/local/lib/python3.5/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 "

/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…

USB: Myriad Execution Finished

USB: Myriad Connection Closing.

USB: Myriad Connection Closed.

Result: (128, 104, 104)

1) 240215 nan

2) 954058 nan

3) 954065 nan

4) 402449 nan

5) 650229 nan

Expected: (128, 104, 104)

1) 1164665 37.2

2) 1166330 34.88

3) 1164769 34.4

4) 1162714 33.38

5) 1162610 32.16

/home/he/.local/lib/python3.5/site-packages/numpy/core/_methods.py:26: RuntimeWarning: invalid value encountered in reduce

return umr_maximum(a, axis, None, out, keepdims)

/usr/local/bin/ncsdk/Controllers/Metrics.py:75: RuntimeWarning: invalid value encountered in greater

diff)) / total_values * 100

Obtained values

Obtained Min Pixel Accuracy: nan% (max allowed=2%), Fail

Obtained Average Pixel Accuracy: nan% (max allowed=1%), Fail

Obtained Percentage of wrong values: 53.2986432137574% (max allowed=0%), Fail

Obtained Pixel-wise L2 error: nan% (max allowed=1%), Fail

Obtained Global Sum Difference: nan

secnd layer is

layer {

type: "Convolution"

bottom: "conv1"

top: "conv2/dw"

param {

lr_mult: 1.0

decay_mult: 1.0

}

convolution_param {

num_output: 64

bias_term: true

pad: 1

kernel_size: 3

stride: 2

group: 64

engine: CAFFE

weight_filler {

type: "msra"

}

layer {

type: "ReLU"

bottom: "conv2/dw"

top: "conv2/dw"

}

layer {

type: "Convolution"

bottom: "conv2/dw"

top: "conv2"

param {

lr_mult: 1.0

decay_mult: 1.0

}

convolution_param {

num_output: 128

bias_term: true

kernel_size: 1

weight_filler {

type: "msra"

}

layer {

type: "ReLU"

bottom: "conv2"

top: "conv2"

}

May it's input size is too big cause this problem ?

idata · ‎06-01-2018

@hexianyqan Others have experienced a similar issue. https://ncsforum.movidius.com/discussion/comment/1720/#Comment_1720. We are currently looking into this issue. Thanks.