Deep Learning
This repo is the continuation of the Machine-Learning repo. Here I'm gonna to upload all the examples and exercises which I'll do to learn deep learning techniques and all the problems I'll solve using these last few.
Disclaimer
- Tensorflow 2.0 is here! Check HERE. Right now, most of the files in this repo are written using the old version of TF.
- Sometimes GitHub doesn't render properly the notebooks: if the load fails, go on nbviewer to see them.
What are we talking about?
Deep Learning is a subfield of machine learning concerned with algorithms inspired by the structure and function of the brain called artificial neural networks.
Why Deep Learning?
Deep learning has proven to outperform all other algorithms in a lot of situations. Of course, in some cases using it could be not only excessive but also a bad idea, so don't use it randomly (Random Forest and SVM, for example, are great models in many contexts).
BUT...
Deep Learning needs A LOT OF DATA and A LOT OF COMPUTING POWER to perform well. That's why even if the idea behind NNs was developed in the '50s, they became so popular in the last years.
How many types of Neural Networks exist?
More than you think, HERE you can find a well explained list.