Merge pull request #123 from cbrom/master

Refer to issue #77 
fix #77

train.py

@@ -90,9 +90,9 @@ def _pad(seq, max_len, constant_values=0):
                   mode='constant', constant_values=constant_values)
 
 
-def _pad_2d(x, max_len, b_pad=0):
+def _pad_2d(x, max_len, b_pad=0, constant_values=0):
     x = np.pad(x, [(b_pad, max_len - len(x) - b_pad), (0, 0)],
-               mode="constant", constant_values=0)
+               mode="constant", constant_values=constant_values)
     return x
 
 
@@ -417,17 +417,19 @@ def collate_fn(batch):
     # (B, T, C)
     # pad for time-axis
     if is_mulaw_quantize(hparams.input_type):
+        padding_value = P.mulaw_quantize(0, mu=hparams.quantize_channels)
         x_batch = np.array([_pad_2d(np_utils.to_categorical(
             x[0], num_classes=hparams.quantize_channels),
-            max_input_len) for x in batch], dtype=np.float32)
+            max_input_len, padding_value) for x in batch], dtype=np.float32)
     else:
         x_batch = np.array([_pad_2d(x[0].reshape(-1, 1), max_input_len)
                             for x in batch], dtype=np.float32)
     assert len(x_batch.shape) == 3
 
     # (B, T)
     if is_mulaw_quantize(hparams.input_type):
-        y_batch = np.array([_pad(x[0], max_input_len) for x in batch], dtype=np.int)
+        padding_value = P.mulaw_quantize(0, mu=hparams.quantize_channels)
+        y_batch = np.array([_pad(x[0], max_input_len, constant_values=padding_value) for x in batch], dtype=np.int)
     else:
         y_batch = np.array([_pad(x[0], max_input_len) for x in batch], dtype=np.float32)
     assert len(y_batch.shape) == 2