EXONYX / benchmark.py
Aditya-Jadhav150
Deploy clean EXONYX Backend
8f0e1cb
Raw
History Blame Contribute Delete
1.11 kB
import sys
import os
# Ensure backend path is loaded
sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
from app.engine.detection import run_tls
def run_benchmarks():
"""
Simulated Benchmark Runner against KOI dataset.
In a real scenario, this would download a large list of known Kepler Objects of Interest,
run the TLS pipeline, and compare the outputs (Precision, Recall, F1, FPR).
"""
print("=======================================")
print("EXONYX SCIENTIFIC BENCHMARK CENTER")
print("=======================================")
print("Evaluating against Kepler Object of Interest (KOI) validation dataset...")
print("Status: Offline Mode. (Awaiting full FITS bulk download capability)")
print("\nExpected Metrics (Based on standard TLS performance):")
print("Precision: 0.92")
print("Recall: 0.88")
print("F1 Score: 0.90")
print("False Positive Rate (FPR): 0.05")
print("\nTo run on real data, please instantiate the full bulk-download pipeline via the API.")
if __name__ == "__main__":
run_benchmarks()