Functional - Keras
Here we're going to be going over the Keras Functional API. We're going to talk about complex multi-input and multi-output models, different nodes from those models, sharing layers and more.
Main topics:
1. First Example - a densely connected network: we have a dense layer, it takes some inputs and it maps them from one dimension to another dimension.
2. Multi-input and multi-output models - I go ahead and get a new layer, its called a merge layer which takes two outputs and it concatenates them together. I then go through the summary of this. I then talk about how compilation, fit and evaluation change because of the models being multi-input and multi-output.
3. Sharing layers - We have a single input, we have a layer that we share, we apply this layer to the inputs, and then we apply it again to its output which then spits out some prediction. This allows you to share layer weights and which is super cool and incredibly common in Deep Learning. Also, this way we have much fewer parameters.
4. I finally go over the concept of the layer "node".
Links:
1) Link to my Scikit Learn tutorial - A Bit of DataScience and Scikit Learn: • Intro to Scikit Learn
2) The Hitchhiker's Guide to Python - one of the best handbooks to the installation, configuration, and usage of Python that I have come across: http://docs.python-guide.org/en/latest/
3) Link to Keras: https://keras.io
4) Link to TensorFlow: https://www.tensorflow.org
5) GitHub link to a-bit-of-deep-learning-and-keras notebooks: https://github.com/knathanieltucker/a...
6) Link to the History of Deep Learning video will be up soon!
Смотрите видео Functional - Keras онлайн без регистрации, длительностью часов минут секунд в хорошем качестве. Это видео добавил пользователь Data Talks 02 Октябрь 2017, не забудьте поделиться им ссылкой с друзьями и знакомыми, на нашем сайте его посмотрели 16,606 раз и оно понравилось 260 людям.