Skip to content

Execution from command line

The pydre entrypoint provides a command-line interface for processing driving data through project files and generating analysis results.

Command Line Usage

If installed via uv:

uv run pydre -p PROJECT_FILE [-d DATAFILES [DATAFILES ...]] [-o OUTPUT_FILE] [-l LOG_LEVEL]

If installed via pip or another situation where pydre is in your PATH:

pydre -p PROJECT_FILE [-d DATAFILES [DATAFILES ...]] [-o OUTPUT_FILE] [-l LOG_LEVEL]

Required Arguments

  • -p, --projectfile: Path to the project configuration file (TOML or JSON format)
  • Defines filters, ROIs (Regions of Interest), metrics, and configuration settings

Optional Arguments

  • -d, --datafiles: One or more data files to process
  • Can be specified as a space-separated list

  • Overrides or adds to any data files defined in the project file

  • -o, --outputfile: Name of the output file (CSV format)

  • Default: out.csv

  • Overrides the output file name specified in the project file

  • -l, --warninglevel: Logging level

  • Valid options: DEBUG, INFO, SUCCESS, WARNING, ERROR, CRITICAL
  • Default: WARNING

Examples

Basic Usage

Process a project file with the data files defined in it:

pydre -p projects/analysis.toml

Specifying Data Files

Process specific data files with a project file:

pydre -p projects/analysis.toml -d data/drive1.dat data/drive2.dat

Changing Output File

Save results to a custom output file:

pydre -p projects/analysis.toml -o results/custom_output.csv

Verbose Logging

Use INFO level logging for more detailed information:

pydre -p projects/analysis.toml -l INFO

Process Flow

  1. The program loads the project file (TOML or JSON)
  2. If specified, additional data files are added to the project
  3. For each data file:
  4. Data is loaded
  5. Filters are applied in the order defined in the project
  6. ROIs are processed to split data into relevant segments
  7. Metrics are calculated for each ROI
  8. Results are aggregated and saved to the output CSV file

Return Codes

  • 0: Successful execution
  • 1: Error occurred during execution (check logs for details)