NN_server/server.py

@@ -64,34 +64,18 @@ def get_required_drone_streak(freq):
         return 1
 
 
-def get_required_drone_prob(freq):
+def update_drone_streak(freq, prediction):
-    raw_value = config.get(f"DRONE_PROB_THRESHOLD_{freq}", config.get("DRONE_PROB_THRESHOLD_DEFAULT", "0"))
+    if prediction == "drone":
-    try:
-        value = float(raw_value)
-        return min(1.0, max(0.0, value))
-    except (TypeError, ValueError):
-        logging.warning("Invalid DRONE_PROB_THRESHOLD for %s=%r, falling back to 0.0", freq, raw_value)
-        return 0.0
-
-
-def update_drone_streak(freq, prediction, drone_probability):
-    required_prob = get_required_drone_prob(freq)
-    drone_probability = 0.0 if drone_probability is None else float(drone_probability)
-    passes_prob_gate = prediction == "drone" and drone_probability >= required_prob
-
-    if passes_prob_gate:
         drone_streaks[freq] = drone_streaks.get(freq, 0) + 1
     else:
         drone_streaks[freq] = 0
 
     required = get_required_drone_streak(freq)
-    triggered = passes_prob_gate and drone_streaks[freq] >= required
+    triggered = prediction == "drone" and drone_streaks[freq] >= required
     logging.info(
-        "NN alarm gate freq=%s prediction=%s drone_probability=%.3f threshold=%.3f streak=%s/%s triggered=%s",
+        "NN alarm gate freq=%s prediction=%s streak=%s/%s triggered=%s",
         freq,
         prediction,
-        drone_probability,
-        required_prob,
         drone_streaks[freq],
         required,
         triggered,
@@ -180,21 +164,15 @@ def receive_data():
                 print('-' * 100)
                 print(str(model))
                 result_msg[str(model.get_model_name())] = {'freq': freq}
-                inference_result = model.get_inference([np.asarray(data['data_real'], dtype=np.float32), np.asarray(data['data_imag'], dtype=np.float32)])
+                prediction, probability = model.get_inference([np.asarray(data['data_real'], dtype=np.float32), np.asarray(data['data_imag'], dtype=np.float32)])
-                if inference_result is None:
-                    raise RuntimeError(f"Inference failed for {model.get_model_name()}")
-                prediction, probability = inference_result[:2]
-                drone_probability = float(probability) if prediction == "drone" else 0.0
                 result_msg[str(model.get_model_name())]['prediction'] = prediction
                 result_msg[str(model.get_model_name())]['probability'] = str(probability)
-                result_msg[str(model.get_model_name())]['drone_probability'] = str(drone_probability)
-                result_msg[str(model.get_model_name())]['drone_threshold'] = str(get_required_drone_prob(freq))
                 prediction_list.append(prediction)
                 print('-' * 100)
                 print()
                 
                 try:
-                    result = update_drone_streak(freq, prediction, drone_probability)
+                    result = update_drone_streak(freq, prediction_list[0])
                     data_to_send={
                         'freq': str(freq),
                         'amplitude': result