Can I use a transfer learning model anywhere?

[Christian_Simonis]

@tbhaxor

Ohk to confirm again, you mean “any” model trained on image classification dataset will work for any image classification problem related to any business.

According to me it should not work well because the features extraction in the CNN layers which are usually freezed will of that dataset on which it was initially trained.

I agree with you, @tbhaxor.

Transfer Learning works well, e.g. in applications of:

• if you learn a transfer of style
• if you have (implicit) representative features learned in the pre-trained model which help in the fine-tuning, see also @shanup’s excellent example above!

In the end a deep learning model can learn what is contained in the data. If the data is sufficiently representative, compared to the new specialised tasks, transfer learning should be worth a look! Then you can leverage the pre-trained model (which is usually also super expensive and difficult to create in the first place).

Best regards
Christian