CLI Reference

ResPredAI provides a command-line interface for all operations. This section documents all available commands.

Getting Help

# Show all available commands
respredai --help

# Show help for a specific command
respredai run --help

Commands Overview

Command	Description
run	Run nested cross-validation pipeline
train	Train models on full dataset for cross-dataset validation
evaluate	Evaluate trained models on new data
feature-importance	Extract and visualize feature importance
create-config	Generate template configuration file
validate-config	Validate configuration file
list-models	Display available ML models
info	Show version and citation information

Main Workflow

1. Nested Cross-Validation (Standard)

# Create configuration
respredai create-config my_config.ini

# Validate configuration
respredai validate-config my_config.ini --check-data

# Run pipeline
respredai run --config my_config.ini

# Extract feature importance
respredai feature-importance --output ./output --model LR --target Target1

2. Cross-Dataset Validation

# Train on dataset A
respredai train --config config_A.ini --output ./trained/

# Evaluate on dataset B
respredai evaluate --models-dir ./trained/trained_models --data dataset_B.csv --output ./eval/

Global Options

Option	Description
--help	Show help message
--version	Show version number

Quick Examples

# Show available models
respredai list-models

# Show version and citation
respredai info

# Run pipeline quietly
respredai run --config config.ini --quiet

# Train specific models only
respredai train --config config.ini --models LR,RF