example_search.py

#!/usr/bin/env python3
"""
Standalone semantic search for ComparEdge SaaS Product Matcher.
Install: pip install sentence-transformers numpy huggingface_hub torch
Usage:   python example_search.py "best CRM for small team"
"""
import sys, json
import numpy as np
import torch
from sentence_transformers import SentenceTransformer
from sentence_transformers.util import cos_sim
from huggingface_hub import hf_hub_download

MODEL_ID = "ComparEdge/saas-product-matcher"

def main():
    query = " ".join(sys.argv[1:]) if len(sys.argv) > 1 else "best project management tool"
    print(f"Loading {MODEL_ID}...")
    model = SentenceTransformer(MODEL_ID)

    emb_path = hf_hub_download(MODEL_ID, "product_embeddings.npy")
    idx_path  = hf_hub_download(MODEL_ID, "products_index.json")

    embeddings = np.load(emb_path)
    with open(idx_path) as f:
        products = json.load(f)

    q_emb  = model.encode(query, normalize_embeddings=True)
    scores = cos_sim(torch.tensor(q_emb), torch.tensor(embeddings))[0]
    top_idx = scores.argsort(descending=True)[:5]

    print(f"\nTop matches for: \"{query}\"")
    print("-" * 60)
    for i, idx in enumerate(top_idx):
        p = products[idx]
        print(f"{i+1}. {p['name']} ({p['category']})")
        print(f"   Score : {scores[idx]:.3f}")
        print(f"   About : {p['description']}")
        print(f"   Link  : https://comparedge.com/tools/{p['slug']}")
        print()

if __name__ == "__main__":
    main()