https://community.intel.com/t5/Intel-Distribution-of-OpenVINO/Classid-Labels/m-p/1179225#M17869 <P>Hi James,</P><P>You should be able to search the internet for the associated labels and annotation files. But labels and annotation are not associated with the models, they are associated with the dataset which was used to train the models, for instance, Coco, ImageNet, etc...</P><P>Well, OpenVINO doesn't&nbsp;ship with labels file for all supported models, so, for example, you can&nbsp;pull the ILSVRC2012 sysnset_words.txt from&nbsp;<A href="http://dl.caffe.berkeleyvision.org/caffe_ilsvrc12.tar.gz">http://dl.caffe.berkeleyvision.org/caffe_ilsvrc12.tar.gz</A>&nbsp;and rename&nbsp;it as googlenet-v1.labels. Note that the ILSVRC2012 synset_words.txt won't work for GoogLeNet V2 since V2 was trained on ILSVRC2015, you'd have to fetch the 2015 labels file when running GoogLeNet V2.</P><P>You have to run the classification sample with --lables option followed by the lables file downloaded from respective sites.</P> Tue, 09 Jul 2019 11:36:06 GMT HemanthKum_G_Intel 2019-07-09T11:36:06Z https://community.intel.com/t5/Intel-Distribution-of-OpenVINO/Classid-Labels/m-p/1179224#M17868 <P>New to the compute stick.&nbsp; I've been using the google coral and jetson nano.</P><P>I was running the alexnet and googlenet caffe model and got outputs with no labels.&nbsp; Is there a list somewhere or a way to input labels?</P><P>googlenet</P><P>classid probability<BR />------- -----------<BR />616 &nbsp; &nbsp; 0.4355762 &nbsp;<BR />50 &nbsp; &nbsp; &nbsp;0.1069827 &nbsp;<BR />836 &nbsp; &nbsp; 0.0557207 &nbsp;<BR />36 &nbsp; &nbsp; &nbsp;0.0534245 &nbsp;<BR />429 &nbsp; &nbsp; 0.0518655 &nbsp;<BR />494 &nbsp; &nbsp; 0.0363212 &nbsp;<BR />452 &nbsp; &nbsp; 0.0321387 &nbsp;<BR />544 &nbsp; &nbsp; 0.0257031 &nbsp;<BR />600 &nbsp; &nbsp; 0.0205605 &nbsp;<BR />772 &nbsp; &nbsp; 0.0144706&nbsp;</P><P>alexnet</P><P>classid probability<BR />------- -----------<BR />88 &nbsp; &nbsp; &nbsp;0.9810570 &nbsp;<BR />12 &nbsp; &nbsp; &nbsp;0.0157526 &nbsp;<BR />7 &nbsp; &nbsp; &nbsp; 0.0009286 &nbsp;<BR />90 &nbsp; &nbsp; &nbsp;0.0009172 &nbsp;<BR />14 &nbsp; &nbsp; &nbsp;0.0005626 &nbsp;<BR />136 &nbsp; &nbsp; 0.0002139 &nbsp;<BR />92 &nbsp; &nbsp; &nbsp;0.0001659 &nbsp;<BR />129 &nbsp; &nbsp; 0.0001552 &nbsp;<BR />91 &nbsp; &nbsp; &nbsp;0.0000463 &nbsp;<BR />17 &nbsp; &nbsp; &nbsp;0.0000393 &nbsp;</P><P>&nbsp;</P><P>&nbsp;</P> Mon, 08 Jul 2019 09:19:20 GMT https://community.intel.com/t5/Intel-Distribution-of-OpenVINO/Classid-Labels/m-p/1179224#M17868 JVond 2019-07-08T09:19:20Z https://community.intel.com/t5/Intel-Distribution-of-OpenVINO/Classid-Labels/m-p/1179225#M17869 <P>Hi James,</P><P>You should be able to search the internet for the associated labels and annotation files. But labels and annotation are not associated with the models, they are associated with the dataset which was used to train the models, for instance, Coco, ImageNet, etc...</P><P>Well, OpenVINO doesn't&nbsp;ship with labels file for all supported models, so, for example, you can&nbsp;pull the ILSVRC2012 sysnset_words.txt from&nbsp;<A href="http://dl.caffe.berkeleyvision.org/caffe_ilsvrc12.tar.gz">http://dl.caffe.berkeleyvision.org/caffe_ilsvrc12.tar.gz</A>&nbsp;and rename&nbsp;it as googlenet-v1.labels. Note that the ILSVRC2012 synset_words.txt won't work for GoogLeNet V2 since V2 was trained on ILSVRC2015, you'd have to fetch the 2015 labels file when running GoogLeNet V2.</P><P>You have to run the classification sample with --lables option followed by the lables file downloaded from respective sites.</P> Tue, 09 Jul 2019 11:36:06 GMT https://community.intel.com/t5/Intel-Distribution-of-OpenVINO/Classid-Labels/m-p/1179225#M17869 HemanthKum_G_Intel 2019-07-09T11:36:06Z