Skip to main content

CODITECT NESTED LEARNING Processor

Extracts reusable patterns from sessions for knowledge accumulation. Implements workflow, decision, and code pattern recognition with knowledge graph construction and similarity scoring.

NESTED = Networked Extraction System for Transferable Experience and Decisions

Features:

  • Workflow pattern recognition (task sequences)
  • Decision pattern extraction (rationale capture)
  • Code pattern detection (reusable templates)
  • Knowledge graph construction (pattern relationships)
  • Similarity scoring (pattern matching)
  • Incremental learning (pattern evolution)

Usage: from nested_learning import NestedLearningProcessor, PatternType

processor = NestedLearningProcessor()

# Extract patterns from session
patterns = processor.extract_patterns(session_data)

# Find similar patterns
similar = processor.find_similar_patterns(query_pattern, threshold=0.7)

# Update pattern library
processor.update_pattern_library(new_patterns)

Author: AZ1.AI CODITECT Team Sprint: Sprint +1 - MEMORY-CONTEXT Implementation Day 4 Date: 2025-11-16

File: nested_learning.py

Classes​

NestedLearningError​

Base exception for NESTED LEARNING errors.

PatternExtractionError​

Raised when pattern extraction fails.

PatternStorageError​

Raised when pattern storage fails.

DatabaseError​

Raised when database operations fail.

ConfigurationError​

Raised when configuration is invalid.

PatternType​

Types of patterns that can be extracted.

Pattern​

Represents an extracted pattern.

WorkflowPattern​

Workflow-specific pattern with task sequence.

DecisionPattern​

Decision-specific pattern with options and rationale.

CodePattern​

Code-specific pattern with language and structure.

Functions​

main()​

Main entry point for testing.

extract_patterns(session_data)​

Extract patterns from session data.

store_patterns(patterns)​

Store patterns in database.

track_pattern_usage(pattern_id, success)​

Track pattern usage for incremental learning.

update_pattern_quality(pattern_id)​

Update pattern quality score based on usage metrics.

get_pattern_evolution(pattern_id)​

Get evolution history for a pattern.

deprecate_pattern(pattern_id, reason)​

Mark a pattern as deprecated.

recommend_patterns(context, pattern_type, limit, min_quality)​

Recommend patterns relevant to current context.

find_similar_patterns(query, pattern_type, threshold, limit)​

Find patterns similar to query.

get_pattern_statistics()​

Get statistics about pattern library.

extract(conversation, metadata)​

Extract workflow patterns from conversation.

extract(decisions, metadata)​

Extract decision patterns from decisions list.

extract(file_changes, metadata)​

Extract code patterns from file changes.

extract(conversation, file_changes, metadata)​

Extract error patterns from conversation and file changes.

extract(decisions, file_changes, metadata)​

Extract architecture patterns from decisions and file structure.

Usage​

python nested_learning.py