DAZL Documentation | Data Analytics A-to-Z Processing Language

Contents

Quick Index Pages (1)

Steps (34)

Recipes (24)

Topic Maps (18)

Examples (18)

Tutorials (6)

Reference (7)

DAZL Language Reference Guide

general

slug: reference-dazl-language-reference-guide

DAZL Steps by Functional Area

DAZL provides a wide range of steps for every stage of the data pipeline — from loading and transforming data to analyzing, visualizing, and deploying models. Steps marked (planned) are coming soon.

Data Management

Steps for loading, transforming, and reshaping data.

  • load step – Load data from an external dataset or file.
  • loadInline step – Load data defined directly inside the YAML script.
  • filter step – Keep or exclude rows based on conditions.
  • calculate step – Create new columns or compute values.
  • sort step – Sort data by one or more columns.
  • keep step – Keep only the specified columns.
  • drop step – Remove specific columns from the dataset.
  • transpose step – Switch rows and columns.
  • lengthen step – Convert wide datasets into long format.
  • widen step – Convert long datasets into wide format.
  • combine step – Merge multiple datasets together.
  • compare step – Compare two datasets row by row or column by column.
  • release step – Remove temporary datasets from memory.
  • attributes step – View or modify metadata about columns.
  • lag step – Shift values in a column forward or backward.

Presentation Tools

Steps for visualizing and presenting data.

  • print step – Print tables or data subsets to output.
  • chart step – Create charts from data (bar, line, pie, etc.).
  • catalog step – Organize datasets or results into catalogs.
  • dashboard step – Combine charts and tables into dashboards.
  • Planned: [step-cards] [step-calendar] [step-report] [step-cubeview]

Statistical Primitives

Steps for statistical summaries and analysis.

  • freq step – Count occurrences of values in columns.
  • univariate step – Compute basic statistics for a single column.
  • timeSeries step – Analyze time-series data.
  • Planned: [step-corr] [step-rank] [step-reg] [step-xfreq] [step-kmeans]

Business Analytics Primitives

Steps for advanced analytics like segmentation, cubes, and contribution analysis.

  • classify step – Categorize rows based on rules or models.
  • cube step – Create multidimensional summaries.
  • rfm step – Compute recency, frequency, and monetary metrics.
  • basket step – Perform market basket or association analysis.
  • contribution step – Calculate contributions of items to totals.
  • index step – Compute index values for comparisons.
  • pareto step – Identify top contributors to cumulative results.
  • Planned: [step-describe]

Machine Learning

  • trainModel step – Train a predictive or classification model.
  • useModel step – Apply a trained model to new data.

Network Graph

Steps for graph or network-based analytics.

  • Planned: [step-edges] [step-findneighbors] [step-traverse] [step-centrality]

Linear Algebra

Steps for vector operations and clustering.

  • Planned: [step-buildvectors] [step-comparevectors] [step-clustervectors]

Reasoning

Steps for problem-solving, pattern recognition, and argument mapping.

  • Planned: [step-findsolution] [step-findpatterns] [step-testpatterns] [step-applypatterns] [step-buildargumentmap]

External Communications

  • Planned: [step-callal] [step-sendemail]

Workflow Control

Steps for controlling execution flow and iterating over data.

  • executeIf step – Run steps conditionally.
  • forEach step – Repeat steps for each element in a collection.
  • exit step – Stop pipeline execution early.

See Also