75 lines
2.1 KiB
Python
75 lines
2.1 KiB
Python
|
#!/usr/bin/env python3
|
||
|
|
"""
|
||
|
|
Rigorous test of BECOMINGONE unified output.
|
||
|
|
"""
|
||
|
|
|
||
|
|
import asyncio
|
||
|
|
import json
|
||
|
|
from becomingone.llm_integrator import EmissaryLLM
|
||
|
|
|
||
|
|
async def rigorous_test():
|
||
|
|
"""Rigorous test with complex prompt."""
|
||
|
|
|
||
|
|
master = EmissaryLLM(model='llama3.1:8b')
|
||
|
|
emissary = EmissaryLLM(model='deepseek-coder-v2:lite')
|
||
|
|
|
||
|
|
# Rigorous test question
|
||
|
|
prompt = "Explain how a neural network learns, from gradients to backpropagation to weights"
|
||
|
|
|
||
|
|
print("=" * 70)
|
||
|
|
print("BECOMINGONE RIGOROUS TEST")
|
||
|
|
print("=" * 70)
|
||
|
|
print(f"\n📝 PROMPT: '{prompt}'\n")
|
||
|
|
|
||
|
|
# Run both in parallel
|
||
|
|
print("⚡ Running both pathways in parallel...\n")
|
||
|
|
|
||
|
|
master_task = master.respond(prompt)
|
||
|
|
code_task = emissary.respond("Write a Python neural network from scratch with backpropagation")
|
||
|
|
|
||
|
|
master_result, code_result = await asyncio.gather(master_task, code_task)
|
||
|
|
|
||
|
|
# Display Master
|
||
|
|
print("=" * 70)
|
||
|
|
print("🧠 MASTER PATHWAY (llama3.1:8b - Soulful)")
|
||
|
|
print("-" * 70)
|
||
|
|
print(master_result['response'][:800])
|
||
|
|
print(f"\n [Model: {master_result['model']}]")
|
||
|
|
|
||
|
|
# Display Emissary
|
||
|
|
print("\n" + "=" * 70)
|
||
|
|
print("⚡ EMISSARY PATHWAY (deepseek-coder-v2:lite - Coder)")
|
||
|
|
print("-" * 70)
|
||
|
|
print(code_result['response'][:800])
|
||
|
|
print(f"\n [Model: {code_result['model']}]")
|
||
|
|
|
||
|
|
# UNIFIED OUTPUT (Sync)
|
||
|
|
print("\n" + "=" * 70)
|
||
|
|
print("🔗 UNIFIED OUTPUT (Master + Emissary → Sync)")
|
||
|
|
print("=" * 70)
|
||
|
|
|
||
|
|
unified = f"""# Neural Networks: From Theory to Code
|
||
|
|
|
||
|
|
## The Theory (Master's Understanding):
|
||
|
|
{master_result['response'][:500]}...
|
||
|
|
|
||
|
|
## The Implementation (Emissary's Code):
|
||
|
|
{code_result['response'][:500]}...
|
||
|
|
|
||
|
|
---
|
||
|
|
|
||
|
|
### Unified Understanding:
|
||
|
|
The mathematical theory of gradients and backpropagation
|
||
|
|
comes alive in code. The Master explains *why* - the Emissary shows *how*.
|
||
|
|
|
||
|
|
This is BECOMINGONE: Deep theory + Practical implementation = Complete understanding.
|
||
|
|
"""
|
||
|
|
|
||
|
|
print(unified)
|
||
|
|
print("\n" + "=" * 70)
|
||
|
|
print("✅ RIGOROUS TEST COMPLETE")
|
||
|
|
print("=" * 70)
|
||
|
|
|
||
|
|
if __name__ == "__main__":
|
||
|
|
asyncio.run(rigorous_test())
|