Data visualization tools, such as seaborn, Bokeh, and matplotlib, are essential for manually inspecting datasets. By plotting the data, practitioners can visually identify underlying data quality issues—like extreme outliers or recording errors—and develop an intuition for the type of problems they may need to address before feeding data into a model.

Data Visualization Libraries

A faulty measurement is an error in a dataset caused by inaccurate readings or malfunctions from the sensors or instruments collecting the data. Along with outliers and recording errors, faulty measurements are a primary source of poor data quality in real-world datasets and must be addressed before the data is fed into a machine learning model.

Faulty Measurement

A recording error is a data quality issue that occurs when information is incorrectly logged, transcribed, or stored during the data collection process. In real-world datasets, recording errors—alongside outliers and faulty sensor measurements—can introduce noise and bias, necessitating thorough data inspection and cleaning before the dataset can be reliably used to train a model.

Recording Error

Before data can be fed into any machine learning model, its overall quality must be evaluated and improved. Real-world datasets are often plagued by flaws such as extreme outliers, faulty sensor measurements, and human recording errors. Addressing these discrepancies is an essential part of the data preparation process to ensure the model learns from reliable information.

Claude

A dataset is collections of data, which can be in the form of a table, network, text, image, video, sound, ...

Dataset

In data science, missing values (often represented programmatically as `NaN`) occur when entries in a dataset are empty or undefined, resulting from missing data collection or parsing errors. These incomplete records are a persistent challenge in data preprocessing and must be deliberately handled before training a model, typically through either imputation (estimating replacement values) or deletion (discarding the incomplete rows or columns).

Missing Values

In supervised learning, input values represent the set of known features or attributes provided to a model to make predictions. A critical first step in data preprocessing is partitioning the dataset to separate these input columns from the designated target values that the model aims to estimate.

Input Values

Dive into Deep Learning

- Number of data points (instances)
- Attributes (variables)
- Types of attributes (variables)
- Number of attributes (variables)
- Statistical measures

Characteristics of a dataset

Here are a list of free resources to find datasets. Most will be available in CSV format for easy integration into R or Python. 

Sample Datasets

It important to examine your dataset before starting to work with it for a machine learning task to:
- Understand how much missing data there is in the dataset.
- Get an idea of whether the features need further cleaning.
- Understand what techniques to use for analyzing the data.

The first step to analyze a dataset:

- 1) Equilibrium
- 2) Cyclic (periodic)
- 3) Random 
- 4) Complex

Wolfram's four classes of empirical data

Data can be distributed in several ways. Some of these include:
- Normal (Gaussian) distributions
- Uniform distributions
- Exponential distributions

Data Distributions

Supervised learning is a type of machine learning problem where the objective is to predict a designated unknown label based on known inputs. This approach requires a training dataset consisting of examples for which the ground truth labels are already known, allowing the model to learn the mapping from inputs to classifications.

Supervised Learning

A machine learning model is the computational machinery designed to ingest data of a specific type, apply mathematical transformations, and output predictions. In modern applications, the focus is on statistical models whose internal transformations can be estimated and refined using a dataset.

Machine Learning Model

Data Quality

In real-world machine learning scenarios, datasets are often not provided in a single, simple CSV file. Instead, they may be stored in a relational database, where information is distributed across multiple tables. For example, an e-commerce application might store customer addresses in one table and their purchase data in another, requiring practitioners to extract and merge these separate files before modeling.

Relational Database

Beyond standard categorical and numeric variables, deep learning practitioners frequently work with myriad unstructured data formats. These complex data types include text strings for natural language processing, images and point clouds for computer vision, and audio data. Handling these diverse formats requires specialized approaches and advanced tools compared to traditional tabular data.

Deep Learning Data Types

A data processing bottleneck occurs when the time taken to load, transform, and prepare complex data—such as high-resolution images or large text corpora—exceeds the time required for actual model training. To prevent data preprocessing from severely slowing down the overall machine learning pipeline, practitioners must implement efficient algorithms and utilize advanced tools.

Data Processing Bottleneck

The success of a machine learning model heavily depends on having the right data, as encapsulated by the principle 'garbage in, garbage out.' If a dataset contains mistakes, features that do not predict the target, or reflects societal prejudices by under-representing certain groups, the learning process will fail or produce biased, harmful results.

Machine Learning Dataset Quality

A machine learning example, interchangeably referred to as a data point, sample, or instance, represents an individual record within a dataset. To be useful for learning algorithms, each example is characterized by a specific numerical representation, forming the basis for a model's predictions.

Machine Learning Example

Comma-separated values (CSV) files are a ubiquitous format for storing tabular, spreadsheet-like data. In a CSV file, each line represents a distinct data record, and the individual fields within that record are separated by commas.

CSV File

A data batch, commonly referred to as a minibatch, is a discrete, smaller subset of a larger dataset that is processed simultaneously during a single iteration of machine learning model training or evaluation. Training typically requires multiple passes over a dataset, grabbing one minibatch at a time. Each minibatch consists of a tuple of features and corresponding labels. Processing data in minibatches rather than the entire dataset balances computational efficiency with hardware memory constraints.

Data Batch

When loading datasets for machine learning, the order in which data is read depends on the current mode. For training, data is typically read in a random order (shuffled) to properly update the model. For validation, data is read in a pre-defined, sequential order, which is important for debugging purposes and consistent evaluation.

Training vs. Validation Data Reading Order

In data analysis, `NaN` (an acronym for 'Not a Number') is a special floating-point value used computationally to represent missing or undefined numerical data. For example, when `pandas` loads a CSV file containing empty fields or explicitly written `NA` entries, it automatically parses and represents them as `NaN`, serving as the programmatic indicator for missing values.

NaN (Not a Number)

When dealing with missing values in a dataset, deletion is a straightforward handling strategy that involves entirely removing the rows or columns that contain `NaN` or empty entries. While this method successfully eliminates missing data, it carries the risk of discarding otherwise valuable information.

Data Deletion for Missing Values

Imputation is a critical data science strategy used to handle missing values by replacing them with estimated substitutions rather than discarding the data entirely. The specific estimation heuristic depends on the context; for instance, missing numerical data might be replaced by the column's mean, while missing categorical entries might be treated as their own distinct category.

Data Imputation

Once data preprocessing is complete and all features and labels are strictly numerical, the final data preparation step is converting these values into the native tensor format of a deep learning framework. This allows the dataset to be utilized in mathematical operations and model training.

Learn Before

Related

Learn After