FinML Toolkit

A utility toolkit for logging, tracking, and managing financial ML operations. Each command records timestamped entries to its own log file for auditing and review.

Commands

Core Operations

Utility Commands

Data Storage

All data is stored in ~/.local/share/finml-toolkit/:

• Each command writes to its own log file (e.g., run.log, check.log, analyze.log)
• All actions are also recorded in history.log with timestamps
• Export files are written to the same directory as export.json, export.csv, or export.txt
• Log format: YYYY-MM-DD HH:MM|\x3Cinput> (pipe-delimited)

Requirements

• Bash (no external dependencies)
• Works on Linux and macOS

When to Use

• When you need to log and track financial ML operations over time
• To maintain an audit trail of run, check, convert, analyze, or generate actions
• When you want to search or export historical operation records
• For batch tracking of ML processing pipelines
• To compare and report on financial data processing tasks
• When managing configurations for finml workflows

Examples

# Log operations
finml-toolkit run "backtest strategy alpha-3"
finml-toolkit check "validate portfolio weights"
finml-toolkit convert "csv to parquet format"
finml-toolkit analyze "correlation matrix on sector data"
finml-toolkit generate "monthly performance report"
finml-toolkit batch "process all Q4 earnings files"
finml-toolkit compare "strategy A vs strategy B returns"
finml-toolkit config "set risk_threshold=0.05"

# View recent entries for a command (no args)
finml-toolkit run
finml-toolkit analyze

# Search and export
finml-toolkit search "portfolio"
finml-toolkit export json
finml-toolkit stats
finml-toolkit recent
finml-toolkit status

Configuration

Set FINML_TOOLKIT_DIR environment variable to change the data directory. Default: ~/.local/share/finml-toolkit/

Output

All commands output to stdout. Redirect with finml-toolkit run > output.txt.

Powered by BytesAgain | bytesagain.com | [email protected]