Add reinit text encoder and duration predictor parameter (#1562)

* Add reinit encoder and duration predictor option

* Add .data to prevent any overlooked autograd hook

TTS/tts/models/vits.py

@@ -41,6 +41,23 @@
 mel_basis = {}
 
 
+@torch.no_grad()
+def weights_reset(m: nn.Module):
+    # check if the current module has reset_parameters and if it is reset the weight
+    reset_parameters = getattr(m, "reset_parameters", None)
+    if callable(reset_parameters):
+        m.reset_parameters()
+
+
+def get_module_weights_sum(mdl: nn.Module):
+    dict_sums = {}
+    for name, w in mdl.named_parameters():
+        if "weight" in name:
+            value = w.data.sum().item()
+            dict_sums[name] = value
+    return dict_sums
+
+
 def load_audio(file_path):
     """Load the audio file normalized in [-1, 1]
 
@@ -528,6 +545,8 @@ class VitsArgs(Coqpit):
     freeze_waveform_decoder: bool = False
     encoder_sample_rate: int = None
     interpolate_z: bool = True
+    reinit_DP: bool = False
+    reinit_text_encoder: bool = False
 
 
 class Vits(BaseTTS):
@@ -744,6 +763,28 @@ def init_upsampling(self):
                 orig_freq=self.config.audio["sample_rate"], new_freq=self.args.encoder_sample_rate
             )  # pylint: disable=W0201
 
+    def on_init_end(self, trainer):  # pylint: disable=W0613
+        """Reinit layes if needed"""
+        if self.args.reinit_DP:
+            before_dict = get_module_weights_sum(self.duration_predictor)
+            # Applies weights_reset recursively to every submodule of the duration predictor
+            self.duration_predictor.apply(fn=weights_reset)
+            after_dict = get_module_weights_sum(self.duration_predictor)
+            for key, value in after_dict.items():
+                if value == before_dict[key]:
+                    raise RuntimeError(" [!] The weights of Duration Predictor was not reinit check it !")
+            print(" > Duration Predictor was reinit.")
+
+        if self.args.reinit_text_encoder:
+            before_dict = get_module_weights_sum(self.text_encoder)
+            # Applies weights_reset recursively to every submodule of the duration predictor
+            self.text_encoder.apply(fn=weights_reset)
+            after_dict = get_module_weights_sum(self.text_encoder)
+            for key, value in after_dict.items():
+                if value == before_dict[key]:
+                    raise RuntimeError(" [!] The weights of Text Encoder was not reinit check it !")
+            print(" > Text Encoder was reinit.")
+
     def get_aux_input(self, aux_input: Dict):
         sid, g, lid = self._set_cond_input(aux_input)
         return {"speaker_ids": sid, "style_wav": None, "d_vectors": g, "language_ids": lid}