Retrieve W&B or local metrics and generate markdown comparison reports.
npx skills add https://github.com/hao-ai-lab/FastVideo --skill summarize-run
Please help me install this Agent Skill. Skill: summarize-run Source: https://github.com/hao-ai-lab/FastVideo/tree/main/.agents/skills/summarize-run Command: npx skills add https://github.com/hao-ai-lab/FastVideo --skill summarize-run
This Skill simplifies the process of retrieving and summarizing key metrics from W&B run data, enabling faster analysis of machine learning experiments.
Summarize the latest training metrics from the output directory 'outputs/wan_finetune' using a reference summary file to generate a report.
High-intent search queries and answers about installing and using this skill.
To extract W&B run metrics for a machine learning experiment summary, this Skill retrieves data like train loss, step time, and validation results directly from W&B APIs or local JSON summaries to evaluate training progress.
Yes, you can generate a training report from local JSON summaries. This Skill works offline by retrieving metrics from stored JSON files and producing structured markdown reports to document performance without requiring online W&B access.
You can compare current training metrics against a reference run by using a reference summary file. This Skill generates detailed markdown reports comparing your latest results against the reference data to monitor progress and identify issues.
A W&B experiment summary report includes metrics such as train loss, step time, gradient norm, and validation results. These are retrieved from your training data and structured to help evaluate machine learning training progress.
Yes, this tool works with offline W&B training outputs. It can summarize the latest training metrics from an output directory using local JSON summary files, allowing you to evaluate experiment progress without an active W&B connection.
hao-ai-labMonitor W&B training runs to detect anomalies and generate alerts.
founderphantomTrack ML experiments and manage hyperparameter sweeps with Weights & Biases.
dongzhuoyaoLog batch and epoch metrics, configs, and system telemetry to Weights & Biases.
gztcodeAutomate ML experiment tracking and collaboration with Weights & Biases.
