TensorFlow Architecture

Tensorflow architecture works in three parts:

1. Preprocessing the data

Data transformation or preprocessing function is an initial step to extract data from dataset to tensors(accepts and returns a dictionary of tensors)

2. Build the model

TensorFlow provides flexible and efficient serving systems for building machine learning models for production environments. They serve out-of-the-box integration with these models, which can  extended to serve other types of models and data

3. Train and estimate the model

After successful model has built this has to be trained well using training data and should be validated using validation data apart from testing data

It is called Tensorflow because it takes input as a multi-dimensional array, also known as tensors. You can construct a sort of flowchart of operations (called a Graph) that you want to perform on that input. The input goes in at one end, and then it flows through this system of multiple operations and comes out the other end as output.

This is why it is called TensorFlow because the tensor goes in it flows through a list of operations, and then it comes out the other side.