add noise

2 files changed, 34 insertions, 0 deletions

diff --git a/become_yukarin/config.py b/become_yukarin/config.py
index 864de4e..05b0790 100644
--- a/become_yukarin/config.py
+++ b/become_yukarin/config.py
@@ -17,6 +17,8 @@ class DatasetConfig(NamedTuple):
     target_var_path: Path
     features: List[str]
     train_crop_size: int
+    global_noise: float
+    local_noise: float
     seed: int
     num_test: int
 
@@ -84,6 +86,8 @@ def create_from_json(s: Union[str, Path]):
             target_var_path=Path(d['dataset']['target_var_path']).expanduser(),
             features=d['dataset']['features'],
             train_crop_size=d['dataset']['train_crop_size'],
+            global_noise=d['dataset']['global_noise'],
+            local_noise=d['dataset']['local_noise'],
             seed=d['dataset']['seed'],
             num_test=d['dataset']['num_test'],
         ),
diff --git a/become_yukarin/dataset/dataset.py b/become_yukarin/dataset/dataset.py
index a6550b7..83936b1 100644
--- a/become_yukarin/dataset/dataset.py
+++ b/become_yukarin/dataset/dataset.py
@@ -270,6 +270,21 @@ class CropProcess(BaseDataProcess):
         return numpy.split(data, [start, start + self._crop_size], axis=self._time_axis)[1]
 
 
+class AddNoiseProcess(BaseDataProcess):
+    def __init__(self, p_global: float = None, p_local: float = None):
+        assert p_global is None or 0 <= p_global
+        assert p_local is None or 0 <= p_local
+        self._p_global = p_global
+        self._p_local = p_local
+
+    def __call__(self, data: numpy.ndarray, test):
+        assert not test
+
+        g = numpy.random.randn() * self._p_global
+        l = numpy.random.randn(*data.shape).astype(data.dtype) * self._p_local
+        return data + g + l
+
+
 class DataProcessDataset(chainer.dataset.DatasetMixin):
     def __init__(self, data: typing.List, data_process: BaseDataProcess):
         self._data = data
@@ -342,6 +357,21 @@ def create(config: DatasetConfig):
             )),
         ]))
 
+    # add noise
+    data_process_train.append(SplitProcess(dict(
+        input=ChainProcess([
+            LambdaProcess(lambda d, test: d['input']),
+            AddNoiseProcess(p_global=config.global_noise, p_local=config.local_noise),
+        ]),
+        target=ChainProcess([
+            LambdaProcess(lambda d, test: d['target']),
+            AddNoiseProcess(p_global=config.global_noise, p_local=config.local_noise),
+        ]),
+        mask=ChainProcess([
+            LambdaProcess(lambda d, test: d['mask']),
+        ]),
+    )))
+
     data_process_test = data_process_base
 
     num_test = config.num_test