Centre LSTM Model image

projects/06_Generative_Music.py

@@ -122,7 +122,9 @@ st.markdown("The data that our model will be trained on will consist of piano MI
 st.markdown("The sequence of notes and chords from the MIDI files are broken down into increments of 100, which are used to predict the next note or chord.")
 st.markdown("#### Model")
 st.markdown("For this project we will use a network consisting of three LSTM layers, three Dropout layers, two Dense layers and one activation layer.")
-st.image(lstm, caption = 'Music Generation Model', width = 250)
+left_col, cent_col, right_col = st.columns(3)
+with cent_col:
+    st.image(lstm, caption = 'Music Generation Model', width = 250)
 st.markdown("It may be possible to improve this model by playing around with the the structure of the network, or adding new categories (e.g. varying note duration, rest periods between notes, etc). However, to achieve satisfying results with more classes we would also have to increase the depth of the LSTM network.")
 st.markdown("*This is based off the tutorial by Sigurður Skúli [How to Generate Music using a LSTM Neural Network in Keras](https://towardsdatascience.com/how-to-generate-music-using-a-lstm-neural-network-in-keras-68786834d4c5)*")
 st.divider()