Add masks to avoid division through 0 (or taking the log of 0)

NuRadioReco/detector/RNO_G/rnog_detector.py

@@ -1168,7 +1168,14 @@ def __call__(self, freq):
         response = np.ones_like(freq, dtype=np.complex128)
 
         for gain, phase, weight in zip(self.__gains, self.__phases, self.__weights):
-            response *= (gain(freq / units.GHz) * np.exp(1j * phase(freq / units.GHz))) ** weight
+
+            _gain = gain(freq / units.GHz)
+
+            # to avoid RunTime warning and NANs in total reponse
+            if weight == -1:
+                _gain = np.where(_gain > 0, _gain, 1e-6)
+
+            response *= (_gain * np.exp(1j * phase(freq / units.GHz))) ** weight
 
         return response
 
@@ -1212,7 +1219,7 @@ def __rmul__(self, other):
     def __str__(self):
         ampl = 20 * np.log10(np.abs(self(0.5 * units.GHz)))
         return "Response of " + ", ".join([f"{name} ({weight})" for name, weight in zip(self.get_names(), self.__weights)]) \
-            + f": |R(0.5 GHz)| = {ampl} dB (amplitude)"
+            + f": |R(0.5 GHz)| = {ampl:.2f} dB (amplitude)"
 
     def plot(self, show=False, in_dB=True):
         import matplotlib.pyplot as plt
@@ -1221,15 +1228,20 @@ def plot(self, show=False, in_dB=True):
 
         fig, ax = plt.subplots()
         for gain, weight, name in zip(self.__gains, self.__weights, self.__names):
+            _gain = gain(freqs)
+
             if in_dB:
-                ax.plot(freqs / units.MHz, weight * 20 * np.log10(gain(freqs)), label=name)
+                mask = _gain > 0  # to avoid RunTime warning
+                ax.plot(freqs[mask] / units.MHz, weight * 20 * np.log10(_gain[mask]), label=name)
             else:
-                ax.plot(freqs / units.MHz, gain(freqs), label=name)
+                ax.plot(freqs / units.MHz, _gain, label=name)
 
+        _gain = np.abs(self(freqs))
         if in_dB:
-            ax.plot(freqs / units.MHz, 20 * np.log10(np.abs(self(freqs))), color="k", label="total")
+            mask = _gain > 0  # to avoid RunTime warning
+            ax.plot(freqs[mask] / units.MHz, 20 * np.log10(_gain[mask]), color="k", label="total")
         else:
-            ax.plot(freqs / units.MHz, np.abs(self(freqs)), color="k", label="total")
+            ax.plot(freqs / units.MHz, _gain, color="k", label="total")
 
         ax.set_xlabel("frequency / MHz")
         if in_dB: