skills.rest
mirakui
mirakuiCOMMUNITY
Explore all skills by this author →
Website𝕏@mirakuiGitHub

querying-mlflow-metrics

Fetch aggregated trace metrics from MLflow tracking servers.

Updated Sep 9, 2017
Search Tags:#monitoring#metrics#trace#latency#mlflow#token-usage
One-click install
npx skills add https://github.com/mirakui/dotfiles --skill querying-mlflow-metrics-mirakui
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: querying-mlflow-metrics
Source: https://github.com/mirakui/dotfiles/tree/main/claude/skills/querying-mlflow-metrics
Command: npx skills add https://github.com/mirakui/dotfiles --skill querying-mlflow-metrics-mirakui

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Fetch aggregated trace metrics from MLflow tracking servers to help monitor usage patterns, model costs, and performance across experiments, reducing manual querying and ad-hoc reporting.

Core Features & Use Cases

  • Fetch aggregated metrics such as total_tokens, latency, trace_count, and quality evaluations across experiments.
  • Support flexible time windows and grouping by trace_name or trace_status, with outputs in table or json formats.
  • Use Case: Compare token usage and latency across experiments to optimize deployment strategies and cost management.

Quick Start

Query the MLflow server for token usage, latency, and trace statistics for the specified experiments over a given time window.

Frequently Asked Questions about querying-mlflow-metrics

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I fetch aggregated MLflow trace metrics for token usage and latency?▼

To fetch aggregated MLflow trace metrics, provide the MLflow server URL, experiment IDs, target metric, aggregations list, and time interval to return token usage and latency statistics in table or JSON format.

Can I group MLflow trace metrics by trace name or status across experiments?▼

You can group MLflow trace metrics by trace_name or trace_status using optional dimensions, allowing you to compare token usage and latency trends across different experiments.

What do I need to query MLflow metrics for monitoring model costs and performance?▼

Querying MLflow metrics requires an accessible MLflow tracking server URL, a set of experiment IDs, a metric, an aggregations list, and a time interval to monitor usage patterns and model costs.

What is the best way to compare token usage across MLflow experiments?▼

To compare token usage across MLflow experiments, fetch aggregated trace metrics with optional grouping and output the results in table or JSON format for analysis.

Does querying MLflow metrics support quality evaluations and trace counts?▼

Querying MLflow metrics supports fetching aggregated metrics including total_tokens, latency, trace_count, and quality evaluations across specified experiments over a given time window.

Related Skills

Explore More Data & Analytics Skills→
BlackkadderBlackkadder

querying-mlflow-metrics

Fetch aggregated trace metrics from MLflow tracking servers.

Community
Intermediate
teegin-gteegin-g

querying-mlflow-metrics

Fetch aggregated MLflow tracking metrics from servers in table or JSON formats.

Community
Intermediate
terrylicaterrylica
62

mlflow-python

Log MLflow metrics, parameters, and artifacts using the Python API.

Community
Advanced
mkgs-databricks-demosmkgs-databricks-demos
1

querying-mlflow-metrics

Fetch aggregated MLflow metrics with configurable server URL, experiment IDs, and aggregations.

Official
Intermediate