updated MDN RNN TD example

cpmpercussion · Oct 26, 2018 · b5d7468 · b5d7468
1 parent ec9955d
commit b5d7468
Showing 1 changed file with 2 additions and 2 deletions.
diff --git a/notebooks/MDN-RNN-time-distributed-MDN-training.ipynb b/notebooks/MDN-RNN-time-distributed-MDN-training.ipynb
@@ -110,7 +110,7 @@
     "inputs = keras.layers.Input(shape=(SEQ_LEN,OUTPUT_DIMENSION), name='inputs')\n",
     "lstm1_out = keras.layers.LSTM(HIDDEN_UNITS, name='lstm1', return_sequences=True)(inputs)\n",
     "lstm2_out = keras.layers.LSTM(HIDDEN_UNITS, name='lstm2', return_sequences=True)(lstm1_out)\n",
-    "mdn_out = mdn.MDN(OUTPUT_DIMENSION, NUMBER_MIXTURES, name='mdn_outputs')(lstm2_out)\n",
+    "mdn_out = keras.layers.TimeDistributed(mdn.MDN(OUTPUT_DIMENSION, NUMBER_MIXTURES, name='mdn_outputs'), name='td_mdn')(lstm2_out)\n",
     "\n",
     "model = keras.models.Model(inputs=inputs, outputs=mdn_out)\n",
     "model.compile(loss=mdn.get_mixture_loss_func(OUTPUT_DIMENSION,NUMBER_MIXTURES), optimizer='adam')\n",
@@ -204,7 +204,7 @@
    "source": [
     "# Fit the model\n",
     "filepath=\"kanji_mdnrnn-{epoch:02d}-{val_acc:.2f}.hdf5\"\n",
-    "checkpoint = keras.callbacks.ModelCheckpoint(filepath, monitor='val_acc', verbose=1, save_best_only=True, mode='max')\n",
+    "checkpoint = keras.callbacks.ModelCheckpoint(filepath, monitor='val_loss', verbose=1, save_best_only=True, mode='min')\n",
     "callbacks = [keras.callbacks.TerminateOnNaN(), checkpoint]\n",
     "\n",
     "history = model.fit(X, y, batch_size=BATCH_SIZE, epochs=EPOCHS, callbacks=callbacks, validation_data=(Xval,yval))\n",