I am new to the machine learning field, but I wanted to try and implement a simple classification algorithm with Keras. Unfortunately, I have a very small set of data, so I thought to try to apply transfer learning to the problem; however, I couldn't find anything on this online, so I wanted to understand which are the best places to look for a pre-trained neural network. Do you have any suggestion in this regard? Which website is best for getting an idea on how to start a machine learning project?


2 Answers 2


Keras itself provides some of the successful image processing neural networks pretrained on the ImageNet: https://keras.io/applications/

Other deep learning libraries also offer some pretrained models, notably:

Many pretrained models for various platforms can also be found at https://www.gradientzoo.com.

Moreover, if you are interested in some particular network architecture, authors sometimes provide pretrained models themselves, e.g. ResNeXt.


Since the question title is generic (and not specific to computer vision), I will give the NLP-related answer as well, in case it helps someone who stumbles upon looking for pretrained vector embeddings:

The two most popular pre-trained vector embeddings can be found on these links:

There are also a couple of less popular and/or more recent ones:

  • $\begingroup$ Do you think I should edit the title? Also, I am actually looking for a pre-trained texture classification model, but reckoning such a question to be too specific, I thought a general idea on where people look for pre-trained models would have been a good starting point $\endgroup$
    – Eggman
    Feb 6, 2018 at 8:38
  • $\begingroup$ As it stands, it is a general question, so assuming there aren't others like this on the website, this may be a useful source for everyone. However, I don't think someone with the answer to the specific question you have in mind will be drawn to this just from the title. $\endgroup$
    – Zhubarb
    Feb 6, 2018 at 8:44

Not the answer you're looking for? Browse other questions tagged or ask your own question.