Bureaucrats, dave_user, Administrators
4,650
edits
Changes
→Performances
!Prediction time
[s]
|-
|SBCX (i.MX6Q)
|interactive
|3.0
|-
|SBCX (i.MX6Q)
|interactive
|2.6
|-
|SBCX (i.MX6Q)
|interactive
|2.5
|-|SBCX (i.MX6Q)|4|userspace(1 GHz)|2.5
|}
As shown, the prediction time doesn't change significantly to increasing of the number of cores. This should confirm TensorFlow computing engine doesn'timplement a parallel computing engine on this architecture.
