File size: 2,750 Bytes
60e4d0e |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 |
import json
import os
import datetime
from qa_loader import load_qa_and_create_vectorstore
from rag_chain import generate_response
from rapidfuzz import fuzz # Benzerlik oranı hesaplamak için
# Log klasörünü hazırla
os.makedirs("logs", exist_ok=True)
# Zaman damgalı log dosyası
timestamp = datetime.datetime.now().strftime("%Y-%m-%d_%H-%M-%S")
log_file = f"logs/auto_test_results_{timestamp}.txt"
# Vektör veritabanını yükle
retriever = load_qa_and_create_vectorstore()
# Q&A dosyasını oku
with open("Q&A_cleaned.json", "r", encoding="utf-8") as f:
qa_data = json.load(f)
# Performans istatistikleri
total_questions = len(qa_data)
correct_answers = 0
incorrect_answers = 0
# Minimum kabul edilebilir benzerlik oranı
SIMILARITY_THRESHOLD = 60 # %60 eşleşme
# Log dosyasını aç ve başlık ekle
with open(log_file, "w", encoding="utf-8") as log:
log.write(f"Auto Test Run - {timestamp}\n")
log.write("=" * 80 + "\n")
for idx, item in enumerate(qa_data, start=1):
question = item['QUESTION']
expected_answer = item['ANSWER']
print(f"{idx}/{total_questions} Asking: {question}")
ai_response = generate_response(retriever, question)
# Benzerlik oranını hesapla
similarity_score = fuzz.ratio(expected_answer.lower(), ai_response.lower())
if similarity_score >= SIMILARITY_THRESHOLD:
result = f"✅ Correct (Similarity: {similarity_score:.2f}%)"
correct_answers += 1
else:
result = f"❌ Incorrect (Similarity: {similarity_score:.2f}%)"
incorrect_answers += 1
# Log'a yaz
log.write(f"Question {idx}/{total_questions}:\n")
log.write(f"Q: {question}\n")
log.write(f"Expected Answer: {expected_answer}\n")
log.write(f"AI Response: {ai_response}\n")
log.write(f"Similarity: {similarity_score:.2f}%\n")
log.write(f"Result: {result}\n")
log.write("-" * 80 + "\n")
print(f"🔎 {result} - Logged")
# Test sonrası performans özeti
accuracy = (correct_answers / total_questions) * 100
log.write("\nTEST SUMMARY\n")
log.write("=" * 80 + "\n")
log.write(f"Total Questions: {total_questions}\n")
log.write(f"Correct Answers: {correct_answers}\n")
log.write(f"Incorrect Answers: {incorrect_answers}\n")
log.write(f"Accuracy: {accuracy:.2f}%\n")
log.write("=" * 80 + "\n")
# Sonuç özeti terminale yazdır
print("\n🔔 TEST COMPLETED")
print(f"✅ Correct: {correct_answers}")
print(f"❌ Incorrect: {incorrect_answers}")
print(f"📊 Accuracy: {accuracy:.2f}%")
print(f"📂 Detailed log saved to: {log_file}")
|