Nn_builder

Overview:

The nn_builder is a tool that allows users to build neural networks with less boilerplate code. It simplifies the process of creating neural networks by specifying the type of network desired and automatically building it. The tool supports PyTorch and TensorFlow 2.0 frameworks and provides examples and usage instructions in a colab notebook.

Features:

• Simple network creation: Build PyTorch and TensorFlow neural networks with just one line of code using the nn_builder.
• Supported networks: Currently supports three types of networks - NN, CNN, and RNN.
• Flexible network configuration: Specify various parameters for the network, such as input dimension, layers, activation functions, dropout probability, initializer, batch norm, embeddings, output range, and random seed.

Summary:

The nn_builder is a convenient tool that simplifies the process of building neural networks by reducing the amount of boilerplate code required. It supports PyTorch and TensorFlow frameworks and provides a simple and efficient way to create various types of networks. With its flexible configuration options, users can easily customize the network according to their requirements. Overall, the nn_builder is a valuable resource for developers looking to streamline their neural network development process.