incremental backup

model.py

@@ -15,7 +15,7 @@ from loss import CTCLoss
 #  tf.config.experimental.set_memory_growth(i, True)
 
 
-def model(input_dim, output_dim, rnn_layers = 3, rnn_units = 72):
+def model(input_dim, output_dim, rnn_layers = 5, rnn_units = 128):
     li = kl.Input((None, input_dim))
     l1 = kl.Reshape((-1, input_dim, 1))(li)
 
@@ -58,7 +58,7 @@ def model(input_dim, output_dim, rnn_layers = 3, rnn_units = 72):
     lo = kl.Dense(output_dim + 1, activation = 'softmax')(lc2)
 
     m = keras.Model(li, lo)
-    m.compile(optimizer = ko.Lion(0.0004),
+    m.compile(optimizer = ko.Adam(0.0001),
               loss = CTCLoss)
 
     return m

train.py

@@ -22,13 +22,13 @@ valid_ds = to_dataset(valid_data, batch_size = bs)
 m = model(input_dim = fft_length // 2 + 1,
           output_dim = char_to_num.vocabulary_size())
 
-m.load_weights('model23-latest.keras')
-ckpt1 = kc.ModelCheckpoint('model24-latest.keras',
+m.load_weights('model40-latest.keras')
+ckpt1 = kc.ModelCheckpoint('model41-latest.keras',
                            monitor = 'val_loss',
                            save_best_only = False,
                            verbose = 1)
 
-ckpt2 = kc.ModelCheckpoint('model24-best.keras',
+ckpt2 = kc.ModelCheckpoint('model41-best.keras',
                            monitor = 'val_loss',
                            save_best_only = True,
                            verbose = 1)
@@ -36,6 +36,6 @@ ckpt2 = kc.ModelCheckpoint('model24-best.keras',
 ce1 = ce(valid_ds, m)
 
 m.fit(train_ds,
-      epochs = 80,
+      epochs = 40,
       validation_data = valid_ds,
       callbacks = [ckpt1, ckpt2, ce1])