model-log

name: model-log description: Model development log - Track runs, log decisions, generate development report

Model Development Log

Usage

/model-log init              # Initialize new log
/model-log log <run>         # Log a run
/model-log decision <run>    # Log decision for a run
/model-log view              # View current log
/model-log report            # Generate development report
/model-log scan              # Scan and add all runs

Actions

init

Initialize model_development_log.yaml in project root:

library(aipharma)
log <- init_model_log(project_dir = ".")

log

Log a completed run with auto-parsing:

log <- log_run(log, run_number, phase = "base_structural|covariate|final",
               parent_run = "...", changes = "description")

decision

Record accept/reject decision:

log <- log_decision(log, run_number, decision = "accepted|rejected",
                    rationale = "reason for decision")

view

Display current log summary:

print_log_summary(log)

report

Generate markdown development report:

generate_log_report(log, format = "markdown", file = "reports/model_development_report.md")

scan

Scan all runs and add to log:

log <- scan_and_log_runs(log, phase = "...")