Skip to content

TensorFlow

TensorFlow is an open-source framework used in the fields of [[artificial-intelligence]], [[machine-learning]], and [[deep-learning]]^[600-developer__big-data__ai__tensorflow-01.md].

Etymology and Structure

The name "TensorFlow" is a portmanteau derived from its two primary components:

  • Tensor: Refers to a multidimensional array, which serves as the framework's fundamental data structure^[600-developer__big-data__ai__tensorflow-01.md].
  • Flow: Refers to a flow, which represents the computational model^[600-developer__big-data__ai__tensorflow-01.md].

Core Concepts

In the context of the framework, Deep Learning is specifically characterized by its emphasis on feature extraction^[600-developer__big-data__ai__tensorflow-01.md].

Workflow

When utilizing neural networks for classification tasks within TensorFlow, the operational process typically follows these steps^[600-developer__big-data__ai__tensorflow-01.md]:

  1. Extract entity features.
  2. Define the neural network structure.
  3. Train the model and adjust parameters.
  4. Perform prediction.

Sources

^[600-developer__big-data__ai__tensorflow-01.md]