i.XM8MP EVK：NNAPI run ssd_mobilenet_v1_coco in Android

Geo — Thu, 04 Feb 2021 05:25:06 GMT

My Environment
Hardware: NXP i.XM8MP EVK A01
Software: Android version 10

I try to use NNAPI load ssd_mobilenet to inference in Android. Load model later inference first frame that inference cost time will be more. The second frame, later inference cost time will be normal.

Is this reasonable?

// calculate inference time
long l2 = System.currentTimeMillis();
mTflite.runForMultipleInputsOutputs(new Object[]{localObject}, paramBitmap);
long l1 = System.currentTimeMillis();

*inference time*	ssd_mobilenet.tflite	ssd_mobilenet_quant.tflite
CPU（1 thread）	361ms	307ms
CPU（4 thread）	140ms	112 ms
NNAPI (first frame)	2234 ms	12722ms
NNAPI (second frame)	1101ms	27ms

laterat some eventual time in the futureMore (Definitions, Synonyms, Translation)

Re: i.XM8MP EVK：NNAPI run ssd_mobilenet_v1_coco in Android

manish_bajaj — Thu, 25 Feb 2021 01:09:42 GMT

@Geo,

There is appnote on warm time in NXP.com.

https://www.nxp.com/docs/en/application-note/AN12964.pdf

Check it out for the more detail.

-Manish

topic Re: i.XM8MP EVK：NNAPI run ssd_mobilenet_v1_coco in Android in eIQ Machine Learning Software

i.XM8MP EVK：NNAPI run ssd_mobilenet_v1_coco in Android

Re: i.XM8MP EVK：NNAPI run ssd_mobilenet_v1_coco in Android