liveBook · Manning

TensorFlow

As explained in Fight Fraud with Machine Learning

TensorFlow is an open-source platform and library for numerical computation and building, training, and deploying machine-learning models.

TensorFlow is a deep-learning library for building, training, evaluating, monitoring, inferencing, and deploying neural-network models. In TensorFlow 2.x, Keras is integrated as its official high-level API. section 4.2.2

A TensorFlow model can specify:

  • An architecture with input, hidden, and output layers. section 4.2.2
  • An optimizer, loss function, and evaluation metric for training and assessment. section 4.2.2
  • A model summary that exposes layer shapes and parameter counts. section 4.2.2

Small binary-classification network

A Sequential model accepts eight inputs, applies three nine-unit sigmoid hidden layers, and produces one sigmoid output. Compiling with Adam, binary cross entropy, and accuracy prepares it for training and evaluation. section 4.2.2

How Fight Fraud with Machine Learning covers it

How deep it goes
Explains model construction and compilation with a concrete network, stopping short of detailed training and deployment procedures.
Examples
Uses an eight-input Sequential network with three hidden layers, sigmoid activations, Adam, binary cross entropy, and accuracy.
What it assumes
Assumes familiarity with neural-network layers and training choices while demonstrating the TensorFlow implementation directly.
What you can do after
Build a compact binary-classification neural network with TensorFlow's integrated Keras API.
How it connects
Connects TensorFlow with Keras, PyTorch, classical machine-learning libraries, and ONNX model conversion.

Also explained in

Compare how 5 books explain TensorFlow →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Fight Fraud with Machine Learning. The section links point to the parts of the book it draws on.