How to Understand TFLITE Model Type

I would like to learn if a TFLITE model is an object detection or image classification model. What method would you recommend for the aim?

Hi @murkoc, If your are taking about the pre trained models in tfhub you can see the model description to determine if it is an object detection or image classification model. Thank You.


No it is mine model and would like to check that output.

Hi @murkoc, like you want to know how to make the predictions using tflite model or know the tflite model output shape? Thank You.

The shape and some other outputs are able to be checked already but couldn’t see the model type if it was an object detection model or image classification model. If you need checking them please reply back.

Hi @murkoc, If your model output is in three components: The bounding boxes — x1, y1, width, height, class of the bounding box, probability score for that prediction then it was object detection. If it was 2D i.e vector with a length equal to the number of classes and value denoting the score that the image belongs to this class then it was image classification. Thank You.

Thank you!! I am collecting informations possibly solve my big issue and will try to make predicitions on having the tflite model with desired outputs.