How can Tensorflow be used to return constructor arguments of layer instance using Python?

PythonServer Side ProgrammingProgramming

Tensorflow is a machine learning framework that is provided by Google. It is an open-source framework used in conjunction with Python to implement algorithms, deep learning applications, and much more. It is used in research and for production purposes.

Keras was developed as a part of the research for the project ONEIROS (Open-ended Neuro-Electronic Intelligent Robot Operating System). Keras is a deep learning API, which is written in Python. It is a high-level API that has a productive interface that helps solve machine learning problems.

It runs on top of the Tensorflow framework. It was built to help experiment in a quick manner. It provides essential abstractions and building blocks that are essential in developing and encapsulating machine learning solutions.

It is highly scalable and comes with cross-platform abilities. This means Keras can be run on TPU or clusters of GPUs. Keras models can also be exported to run in a web browser or a mobile phone as well.

Keras is already present within the Tensorflow package. It can be accessed using the below line of code.

import tensorflow
from tensorflow import keras

The Keras functional API helps create models that are more flexible in comparison to models created using sequential API. The functional API can work with models that have non-linear topology, can share layers and work with multiple inputs and outputs. A deep learning model is usually a directed acyclic graph (DAG) that contains multiple layers. The functional API helps build the graph of layers.

We are using the Google Colaboratory to run the below code. Google Colab or Colaboratory helps run Python code over the browser and requires zero configuration and free access to GPUs (Graphical Processing Units). Colaboratory has been built on top of Jupyter Notebook. Following is the code snippet to return constructor arguments of layer instance using Python −

Example

class CustomDense(layers.Layer):
   def __init__(self, units=32):
      super(CustomDense, self).__init__()
      self.units = units
   def build(self, input_shape):
      self.w = self.add_weight(
         shape=(input_shape[-1], self.units),
         initializer="random_normal",
         trainable=True,
      )
      self.b = self.add_weight(
         shape=(self.units,), initializer="random_normal", trainable=True
      )
   def call(self, inputs):
      return tf.matmul(inputs, self.w) + self.b
   def get_config(self):
      return {"units": self.units}
inputs = keras.Input((4,))
outputs = CustomDense(10)(inputs)

model = keras.Model(inputs, outputs)
print("The below function returns constructor arguments for the instance of the layer")
config = model.get_config()

new_model = keras.Model.from_config(config, custom_objects={"CustomDense": CustomDense})

Code credit − https://www.tensorflow.org/guide/keras/functional

Output

The below function returns constructor arguments for the instance of the layer

Explanation

  • A class named ‘CustomDense’ is created that is used to add weights to the model.

  • Another function named ‘get_config’ is defined that returns constructor arguments for every instance of the layer.

  • Input layers to the model are defined.

  • Next, the model is defined and the function is called.

raja
Updated on 18-Jan-2021 11:58:28

Advertisements