Create an object of class workflow by using the
dlhdl.Workflow class.
Set the deep learning and bitstream for the workflow object.
Call the estimate function for the workflow object.
The speed and latency are stored in a structure struct and
displayed on the screen.
For example:
snet = vgg19; hW = dlhdl.Workflow('Network', snet, 'Bitstream', 'arria10soc_single'); result = hW.estimate('Performance');
The result of the estimation is:
Deep Learning Processor Estimator Performance Results
LastLayerLatency(cycles) LastLayerLatency(seconds) FramesNum Total Latency Frames/s
------------- ------------- --------- --------- ---------
Network 172441964 1.14961 1 172441964 0.9
conv_module 162622207 1.08415
conv1_1 4528942 0.03019
conv1_2 17788981 0.11859
pool1 2360417 0.01574
conv2_1 8510437 0.05674
conv2_2 15432208 0.10288
pool2 1242064 0.00828
conv3_1 7660645 0.05107
conv3_2 14177125 0.09451
conv3_3 14177125 0.09451
conv3_4 14177125 0.09451
pool3 671713 0.00448
conv4_1 6957812 0.04639
conv4_2 13621492 0.09081
conv4_3 13621492 0.09081
conv4_4 13621492 0.09081
pool4 391652 0.00261
conv5_1 3396733 0.02264
conv5_2 3396733 0.02264
conv5_3 3396733 0.02264
conv5_4 3396733 0.02264
pool5 94553 0.00063
fc_module 9819757 0.06547
fc6 8160258 0.05440
fc7 1331586 0.00888
fc8 327913 0.00219
* The clock frequency of the DL processor is: 150MHz