# BigData / TensorFlow

Tensorflow is a computational framework from Google for creating machine learning models. It provides different toolkits that allow you to create models by choosing a different level of abstraction ranging from lower-level APIs such as series of mathematical operations to the higher level APIS such as tf.estimator for predefined architectures like Neural Networks, linear regressors.

**Artificial intelligence** (AI) refers to the study of how to train computers so that they can perform things that at present humans can do better. Therefore it is an intelligence where we want to add the capabilities to a machine that human contain.

**Machine Learning** is the learning in which a machine can learn on its own without being explicitly programmed. It is an application of AI that provides the system the ability to automatically learn and improve from experience.

**Machine** learning uses algorithms that parse data, learn from that data, and then apply what is learned to make informed decisions. For example, youtube service displays the list of recommendations based on the video that we currently watch.

Deep learning is just a subset of machine learning. Deep learning structures algorithms in layers to create an **artificial neural network** (ANN) that can learn and make intelligent decisions on its own.

- tf.Variable,
- tf.constant,
- tf.placeholder,
- and tf.SpareTensor.

A tensor is a generalization of vectors and matrices to potentially higher dimensions. Internally, TensorFlow represents each tensors as a n-dimensional arrays of base datatypes.

The rank of a tensor is its number of dimensions. The term rank may also be referred as order or degree or n-dimension.

Rank | Math entity |
---|---|

0 | Scalar (magnitude only) |

1 | Vector (magnitude and direction) |

2 | Matrix (table of numbers) |

3 | 3-Tensor (cube of numbers) |

n | n-Tensor |