OpenSkillIndex← back to search
claude-codedevelopment

Babysitter spacy-ner

spaCy NER model training and entity extraction for conversational AI

install
source · Clone the upstream repo
git clone https://github.com/a5c-ai/babysitter
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/a5c-ai/babysitter "$T" && mkdir -p ~/.claude/skills && cp -r "$T/library/specializations/ai-agents-conversational/skills/spacy-ner" ~/.claude/skills/a5c-ai-babysitter-spacy-ner && rm -rf "$T"
manifest: library/specializations/ai-agents-conversational/skills/spacy-ner/SKILL.md
tags
#ner-training#entity-extraction#spacy#chatbot#annotation#model-deployment
source content

spaCy NER Skill

Capabilities

  • Train custom spaCy NER models
  • Configure entity extraction pipelines
  • Design annotation schemas
  • Implement entity linking
  • Set up model evaluation
  • Deploy efficient NER inference

Target Processes

  • entity-extraction-slot-filling
  • chatbot-design-implementation

Implementation Details

spaCy Components

  1. NER: Named Entity Recognition
  2. EntityLinker: Link to knowledge bases
  3. EntityRuler: Rule-based matching
  4. SpanCategorizer: Overlapping entities

Training Configuration

  • config.cfg setup
  • Training data format (spaCy v3)
  • Augmentation strategies
  • Evaluation metrics

Configuration Options

  • Base model selection (en_core_web_*)
  • Custom entity types
  • Training parameters
  • GPU acceleration
  • Model packaging

Best Practices

  • Quality annotation data
  • Balance entity types
  • Use prodigy for annotation
  • Regular model evaluation

Dependencies

  • spacy
  • spacy-transformers (optional)