A ultra-lightweight 3D renderer of the Tensorflow/Keras neural network architectures. This Library is working on Matplotlib visualization for now. In future the visualization can be moved to plotly for a more interactive visual of the neural network architecture.
Note: For now the rendering is working in Jupyter only Google Colab support is in works.
For more details visit NetPlot
pip install netplot
from netplot import ModelPlot import tensorflow as tf import numpy as np
%matplotlib notebook
X_input = tf.keras.layers.Input(shape=(32,32,3)) X = tf.keras.layers.Conv2D(4, 3, activation='relu')(X_input) X = tf.keras.layers.MaxPool2D(2,2)(X) X = tf.keras.layers.Conv2D(16, 3, activation='relu')(X) X = tf.keras.layers.MaxPool2D(2,2)(X) X = tf.keras.layers.Conv2D(8, 3, activation='relu')(X) X = tf.keras.layers.MaxPool2D(2,2)(X) X = tf.keras.layers.Flatten()(X) X = tf.keras.layers.Dense(10, activation='relu')(X) X = tf.keras.layers.Dense(2, activation='softmax')(X) model = tf.keras.models.Model(inputs=X_input, outputs=X)
modelplot = ModelPlot(model=model, grid=True, connection=True, linewidth=0.1) modelplot.show()
