Deep learning models typically employ algorithms known as neural networks, which are said to be inspired by the way actual biological nervous systems (such as the brain) process information. It enables computers to recognize all data points as to what each represents and learn patterns. 

Today, the principal software tool for deep learning models is TensorFlow as it permits developers to create large-scale neural networks with numerous layers. 

TensorFlow is mainly used for the following purposes: 

• Classification
• Perception
• Understanding 
• Discovering 
• Prediction  
• Creation

As noted in the Watson documentation, the challenge with deploying complex machine learning models such as a TensorFlow model is that these models are very computationally expensive and time-consuming to train. Some solutions (to this challenge) include GPU acceleration, distributed computing, or a combination of both. The IBM Cloud platform and Watson Studio offers both of these.

It also points: IBM Watson Studio permits one to leverage the computational power available on the cloud to speed up the training time of the more complex machine learning models, and thus reduce the time from hours or days, down to minutes.

In the next sections, we will explore several exercises demonstrating various ways of using TensorFlow with IBM Watson Studio.