can remove aligner

3 files changed, 15 insertions, 2 deletions

diff --git a/become_yukarin/config.py b/become_yukarin/config.py
index 07f35fd..0efbf04 100644
--- a/become_yukarin/config.py
+++ b/become_yukarin/config.py
@@ -34,6 +34,7 @@ class ModelConfig(NamedTuple):
     out_size: int
     aligner_out_time_length: int
     disable_last_rnn: bool
+    enable_aligner: bool
 
 
 class LossConfig(NamedTuple):
@@ -104,6 +105,7 @@ def create_from_json(s: Union[str, Path]):
             out_size=d['model']['out_size'],
             aligner_out_time_length=d['model']['aligner_out_time_length'],
             disable_last_rnn=d['model']['disable_last_rnn'],
+            enable_aligner=d['model']['enable_aligner'],
         ),
         loss=LossConfig(
             l1=d['loss']['l1'],
diff --git a/become_yukarin/loss.py b/become_yukarin/loss.py
index c59747a..b2b03fc 100644
--- a/become_yukarin/loss.py
+++ b/become_yukarin/loss.py
@@ -7,7 +7,7 @@ from .model import Predictor
 
 
 class Loss(chainer.link.Chain):
-    def __init__(self, config: LossConfig, predictor: Predictor, aligner: Aligner):
+    def __init__(self, config: LossConfig, predictor: Predictor, aligner: Aligner = None):
         super().__init__()
         self.config = config
 
@@ -21,7 +21,8 @@ class Loss(chainer.link.Chain):
         mask = chainer.as_variable(mask)
 
         h = input
-        h = self.aligner(h)
+        if self.aligner is not None:
+            h = self.aligner(h)
         y = self.predictor(h)
 
         loss = chainer.functions.sum(chainer.functions.absolute_error(y, target) * mask)
diff --git a/become_yukarin/model.py b/become_yukarin/model.py
index 3b5102e..c475685 100644
--- a/become_yukarin/model.py
+++ b/become_yukarin/model.py
@@ -212,8 +212,18 @@ def create_predictor(config: ModelConfig):
 
 
 def create_aligner(config: ModelConfig):
+    assert config.enable_aligner
     aligner = Aligner(
         in_size=config.in_channels,
         out_time_length=config.aligner_out_time_length,
     )
     return aligner
+
+
+def create(config: ModelConfig):
+    predictor = create_predictor(config)
+    if config.enable_aligner:
+        aligner = create_aligner(config)
+    else:
+        aligner = None
+    return predictor, aligner