Tensorflow Model Garden Tutorial with different models

General Discussion
, ,
#1

Hi, I’m trying to follow this colab notebook. Google Colab

I want to try different models. I try to change the model line like below:
exp_config = exp_factory.get_exp_config(‘coco_spinenet49_mobile_tpu’)
But it does not work.
I get model names from following link: https://github.com/tensorflow/models/blob/master/official/vision/MODEL_GARDEN.md

But I also wonder that can I use model zoo? : https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/tf2_detection_zoo.md

Thank you for your interest.

#3

Hi @Taha_Er
I know nothing of all this to be honest. But looking quickly around:
Can you please try out :
exp_config = exp_factory.get_exp_config('maskrcnn_spinenet_coco')
And I assume you have to adjust model parameters/settings thereafter in the notebook.

1 Like
#4

Hey @Taha_Er,

Please find the different experiments supported here from official model garden. Once you load you change backbone and heads based on your requirements. Hope this helped.

Thanks.

1 Like
#5

Thank you very much for your answer, as @Siva_Sravana_Kumar_N mention above, solution is using something from this link: tfm.core.exp_factory.get_exp_config  |  TensorFlow v2.14.0

1 Like
#6

Thank you very much, your solution is worked. Now I’m trying to learn which parameters should I switch before running model.

1 Like