Martina Pugliese
Martina Pugliese
Tensors are generalisations of vectors and matrices to higher dimensions and furnish a representation which is very convenient to working with neural networks. In fact, TensorFlow, the Google Machine Learning...
From the notebook I had on this: This page is an overview of some of the common activation functions used in working with neural networks, each with its own advantaged...
From here https://www2.eecs.berkeley.edu/Research/Projects/CS/vision/shape/sc_digits.html the gist of the idea (see image) This article explains https://medium.com/machine-learning-world/shape-context-descriptor-and-fast-characters-recognition-c031eac726f9 Some slides (referref by the Medium article) https://github.com/creotiv/Python-Shape-Context/blob/master/info/ShapeContexts425.pdf
## Hashing an image TODO https://blog.iconfinder.com/detecting-duplicate-images-using-python-cb240b05a3b6 - explains why cryptographic hashing doesn't work, and how to do dhash
* add the list of algorithms suited * add binary classification description
In the page about learning paradigms under ML- algorithms. You'd have to explain * supervised * unsupervised * semi-supervised * reinforcement?
https://crypto.stackexchange.com/questions/18612/how-is-sha1-different-from-md5