Can a pretrained Deep learning model of objects (chair, table) be used to do transfer learning and classify telecom equipment?

Question

I want to classify telecom devices: switches, routers, etc. I know that there are pre-trained model available online: https://github.com/tensorflow/models

Will it be possible to use transfer learning using those? Or do I have to use a pre-trained model of telecom devices? My reasoning is that the algorithm learns "forms"
Which pre-trained model will you recommend me to do transfer learning?

Thank you

Sebastian Dziadzio · Accepted Answer · 2020-03-12T16:43:28.097

1

An intermediate solution would be to use a pre-trained image classification model such as ResNet, freeze the weights of a few initial layers and re-train on your dataset. That way the initial layers will serve as a feature extractor. This should work well as long as your data is similar to ImageNet (in terms of size, number and order of channels, preprocessing etc).

edited Mar 12 '20 at 16:43

answered Mar 12 '20 at 16:23

Sebastian Dziadzio

126
4

Can a pretrained Deep learning model of objects (chair, table) be used to do transfer learning and classify telecom equipment?

1 Answers1