Which machine learning models can I train?

The skill supports training popular transformer models including distilbert-base-uncased, roberta-base, and xlm-roberta-base for multilingual support.

Can I classify multiple categories for a single piece of text?

Yes, by enabling the multi-label option, the classifier can return all labels that meet the confidence threshold for a single input.

What is zero-shot classification in LlamaFarm?

Zero-shot classification allows you to categorize text into any label you define on the fly, without requiring a pre-trained dataset for those specific labels.

How many training examples do I need for a custom classifier?

While it varies by task, it is recommended to provide at least 10-20 examples per class to achieve basic accuracy for a custom model.

Do I need to run a local server for this skill?

Yes, this skill interacts with the LlamaFarm ML runtime, typically accessible at localhost:14345. You may need to run /llamafarm:start first.

LlamaFarm Text Classification

Name: LlamaFarm Text Classification
Author: llama-farm

byllama-farm

0•

데이터 과학 및 ML

Classifies text data using zero-shot labeling or custom-trained machine learning models directly within your terminal.

The LlamaFarm Classify skill provides a comprehensive workflow for organizing and categorizing text data. It enables users to perform immediate zero-shot classification without any prior training, or to train specialized models like DistilBERT and RoBERTa on custom datasets for domain-specific accuracy. This tool is ideal for developers and data scientists who need to automate sentiment analysis, intent recognition, or document tagging while maintaining a seamless workflow inside Claude Code.

주요 기능

01Batch prediction capabilities for processing large volumes of text

02Multi-label support for assigning multiple categories to a single input

03Zero-shot classification for immediate labeling without training data

040 GitHub stars

05Custom model training with support for BERT, RoBERTa, and XLM-RoBERTa

06Integrated model management to list, evaluate, and switch between classifiers

사용 사례

01Automating the categorization of customer support tickets and feedback

02Organizing large sets of documentation or code comments into functional tags

03Training a custom sentiment analysis model on domain-specific project logs

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add llama-farm/claude-code-marketplace classify

For use in Claude.ai and ChatGPT

Download Skill