liveBook · Manning

Missing value

As explained in Deep Learning with Python, Third Edition

An absent data field in a sample that must be handled explicitly during preprocessing.

A missing value is an absent data field that cannot be passed directly to a neural network without explicit preprocessing. Treat it as part of preparing raw data for model input, alongside vectorization and normalization. section 6.2.1 section 5.4.1

Use these practical rules when preparing data:

  • Inspect and clean the dataset before training; missing values belong to data curation, not to the model’s implicit guesswork. section 5.4.1
  • Handle missing values explicitly during preprocessing, before inputs are converted into the tensors consumed by the network. section 6.2.1
  • Treat the exact handling strategy as domain specific when the general workflow does not provide further detail. section 6.2.1

How Deep Learning with Python, Third Edition covers it

How deep it goes
Identifies missing values as a preprocessing task but defers detailed handling strategies to a later treatment.
What it assumes
Assumes raw datasets and neural-network inputs, while placing missing-value handling among preprocessing basics common to data domains.
What you can do after
Apply it by identifying missing values as a data-cleaning requirement before supplying tensors to a neural network.
How it connects
Connects missing values to data preprocessing, dataset curation, tensor inputs, and neural networks.
Cover of Deep Learning with Python, Third Edition
liveBookmanning.com
Read Deep Learning with Python, Third Edition on liveBook Get the book at manning.com

Also explained in

Compare how 35 books explain missing value →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Deep Learning with Python, Third Edition. The section links point to the parts of the book it draws on.