What is this max and min value? how to find it?
scale = (max_value - min_value)/255
zero_point = max_value - max_value/scale
What is this max and min value? how to find it?
scale = (max_value - min_value)/255
zero_point = max_value - max_value/scale
do i have to repost in VIM3 section ??
I am following that document but it gives this information after step 2
Step 1 :
$pegasus import darknet
âmodel ./model/${NAME}.cfg
âweights ./model/${NAME}.weights
âoutput-model ${NAME}.json
âoutput-data ${NAME}.data \
$pegasus generate inputmeta
âmodel ${NAME}.json
âinput-meta-output ${NAME}_inputmeta.yml
âchannel-mean-value â0 0 0 0.0039â
âsource-file dataset.txt
Step 2 :
$pegasus quantize
âquantizer asymmetric_affine
âqtype uint8
ârebuild
âwith-input-meta ${NAME}_inputmeta.yml
âmodel ${NAME}.json
âmodel-data ${NAME}.data
I want to know how it is decided what -channel-mean-value to provide during step 1.
The issue is I am testing a Yolo model, but the accuracy on NPU is very low same model gives me an accuracy of about 74% on CPU and GPU.
how I am using the model :
to convert : sudo ./convert --model-name yolov3_tiny0_0036 --platform darknet --model yolov3-tiny-416.cfg --weights yolov3-tiny-416.weights --mean-values â0 0 0 0.00390625â --quantized-dtype asymmetric_affine --qtype uint8 --source-files ./data/dataset/dataset0.txt --kboard VIM3 --print-level 0
I used âchannel-mean-value â0 0 0 0.0039â value because it was used in all the yolo examples in âksnn/examples/darknet at master · khadas/ksnn · GitHubâ . But it is giving poor accuracy 1.6%