ok

test_local.py

@@ -1,103 +0,0 @@
-import time
-from model_service import LocalEmbeddingService
-
-# Configuration
-LOCAL_MODEL_PATH = './models/bge-base-en-v1.5'
-
-def test_single_text():
-    """Test embedding generation for a single text."""
-    service = LocalEmbeddingService(LOCAL_MODEL_PATH)
-    
-    text = "Ginger was also a smart giraffe. She knew what was wrong."
-    
-    print(f"\n{'='*60}")
-    print("Testing single text embedding")
-    print(f"{'='*60}")
-    print(f"Text: '{text}'")
-    
-    start_time = time.time()
-    vector = service.generate_embedding(text)
-    end_time = time.time()
-    
-    print(f"\n✅ Embedding generated in {end_time - start_time:.4f} seconds")
-    print(f"Dimensions: {len(vector)}")
-    print(f"First 10 values: {vector[:10]}")
-    print(f"Vector norm (should be ~1.0): {sum(x**2 for x in vector)**0.5:.4f}")
-
-def test_batch_texts():
-    """Test embedding generation for multiple texts."""
-    service = LocalEmbeddingService(LOCAL_MODEL_PATH)
-    
-    texts = [
-        "The quick brown fox jumps over the lazy dog.",
-        "Machine learning is transforming technology.",
-        "Embeddings capture semantic meaning of text."
-    ]
-    
-    print(f"\n{'='*60}")
-    print("Testing batch text embeddings")
-    print(f"{'='*60}")
-    print(f"Number of texts: {len(texts)}")
-    
-    start_time = time.time()
-    vectors = service.generate_embedding(texts)
-    end_time = time.time()
-    
-    print(f"\n✅ {len(vectors)} embeddings generated in {end_time - start_time:.4f} seconds")
-    print(f"Average time per text: {(end_time - start_time) / len(texts):.4f} seconds")
-    print(f"Each embedding dimension: {len(vectors[0])}")
-    
-    # Show first embedding sample
-    print(f"\nFirst embedding (first 10 values): {vectors[0][:10]}")
-
-def test_similarity():
-    """Test cosine similarity between embeddings."""
-    service = LocalEmbeddingService(LOCAL_MODEL_PATH)
-    
-    texts = [
-        "The cat sits on the mat.",
-        "A feline rests on the rug.",  # Similar meaning
-        "Python is a programming language."  # Different meaning
-    ]
-    
-    print(f"\n{'='*60}")
-    print("Testing semantic similarity")
-    print(f"{'='*60}")
-    
-    vectors = service.generate_embedding(texts)
-    
-    # Calculate cosine similarities (vectors are already normalized)
-    def cosine_sim(v1, v2):
-        return sum(a * b for a, b in zip(v1, v2))
-    
-    sim_01 = cosine_sim(vectors[0], vectors[1])
-    sim_02 = cosine_sim(vectors[0], vectors[2])
-    
-    print(f"\nText 1: '{texts[0]}'")
-    print(f"Text 2: '{texts[1]}'")
-    print(f"Similarity: {sim_01:.4f} (similar meaning)")
-    
-    print(f"\nText 1: '{texts[0]}'")
-    print(f"Text 3: '{texts[2]}'")
-    print(f"Similarity: {sim_02:.4f} (different meaning)")
-    
-    print(f"\n✅ As expected, similar texts have higher similarity!")
-
-def main():
-    """Run all tests."""
-    try:
-        test_single_text()
-        test_batch_texts()
-        test_similarity()
-        
-        print(f"\n{'='*60}")
-        print("✅ All tests completed successfully!")
-        print(f"{'='*60}\n")
-        
-    except FileNotFoundError:
-        print("\n❌ Model not found. Please run download_model.py first.")
-    except Exception as e:
-        print(f"\n❌ An error occurred: {e}")
-
-if __name__ == "__main__":
-    main()