How do I draw the following Keras model?

Question

I am new to deep learning models, and I am trying to figure out the best way to draw a visual representation of the following model. I found a few tools that allow you to draw a model, but I am confused about what the 'best' way to draw this particular model would be.

Any suggestions or recommendations would be great.

Thanks!

Answer 1

You should Go along with Tensorboard, it looks more professional than the rest and it is easier to understand the model.

May I ask your exact purpose for the visualization? You could check others material with the similar purpose and go along with the general choice! Many would go for draw_convnet or PlotNeuralNet if the final use-case was to publish a blog. It also depends on which libraries and softwares you are using!

Answer 2

You can use tf.keras.utils.plot_model. For instance this example given in the tensorflow doc:

import tensorflow as tf

input = tf.keras.Input(shape=(100,), dtype='int32', name='input')
x = tf.keras.layers.Embedding(
    output_dim=512, input_dim=10000, input_length=100)(input)
x = tf.keras.layers.LSTM(32)(x)
x = tf.keras.layers.Dense(64, activation='relu')(x)
x = tf.keras.layers.Dense(64, activation='relu')(x)
x = tf.keras.layers.Dense(64, activation='relu')(x)
output = tf.keras.layers.Dense(1, activation='sigmoid', name='output')(x)
model = tf.keras.Model(inputs=[input], outputs=[output])
dot_img_file = 'model_1.png'
tf.keras.utils.plot_model(model, to_file=dot_img_file, show_shapes=True)

you get the following:

For this function to work you need to install the dependencies:

• pydot (pip3 install pydot)
• an graphviz (install will depend on your OS) have a look at this link