MLCLI: Machine Learning Job Configuration CLI

Overview

MLCLI is a powerful command-line interface (CLI) tool designed for managing, tracking, searching, comparing, and exporting machine learning job configurations from Elastic (ESS). It provides version control, fuzzy search, color-coded diffing, and structured output for better usability.

Features

✅ Import job configurations (only if changed, skipping duplicates) ✅ Search jobs (with exact and fuzzy matching) ✅ Compare job versions (color-coded diffing, summarized/full) ✅ Export job data (JSON, CSV, Markdown) ✅ Configurable output settings (custom columns, structured table display)

Installation

Prerequisites

• Node.js 14+
• npm or yarn

Install Globally

npm install -g mlcli

Run in Local Project

git clone https://github.com/yourusername/mlcli.git
cd mlcli
npm install

Usage

General Command Syntax

mlcli <command> [options]

1️⃣ Import Jobs (Only If Changed)

mlcli import <file>

Imports ML job configurations from a JSON file only if there are changes compared to the latest version.

Example:

mlcli import jobs.json

2️⃣ Search Jobs (Fuzzy & Structured Output)

mlcli search [options]

Options:

• --job-id <job_id> → Filter by exact job ID.
• --fuzzy <text> → Perform fuzzy search across multiple fields.

Example:

mlcli search --fuzzy anomaly

Output is displayed in a structured table.

3️⃣ Compare Job Versions

mlcli compare [job_id] [options]

Options:

• --full → Show full detailed diff.
• --all → Compare all jobs.

Example:

mlcli compare job_123

Displays a color-coded summary of differences between the latest two versions of a job.

4️⃣ Export Jobs

mlcli export --format <json|csv|md> [--settings]

Options:

• --format json → Exports all job data in JSON format.
• --format csv → Exports job data in CSV format.
• --format md → Exports job data as a Markdown table.
• --settings → Uses user-defined column settings for CSV and Markdown exports.

Example:

mlcli export --format csv

5️⃣ Configure Visible Columns

mlcli settings --columns <column1,column2,...>

Allows users to configure which columns should be displayed in search results and exports.

Example:

mlcli settings --columns job_id,description,groups

To view available columns:

mlcli settings

6️⃣ Interactive Mode (REPL)

mlcli

Runs MLCLI in an interactive shell where you can enter commands continuously without re-running mlcli each time.

Example:

mlcli> search --fuzzy anomaly
mlcli> compare job_123
mlcli> export --format json
mlcli> exit

Example Workflow

mlcli import jobs.json
mlcli search --fuzzy anomaly
mlcli compare job_123 --full
mlcli export --format csv

Contributing

We welcome contributions! Please submit issues or pull requests via GitHub.

License

This project is licensed under the MIT License.