Keras is a high-level deep learning API written in Python. It provides an easy-to-use interface for building, training, and deploying deep learning models. Keras is now part of the TensorFlow library and is often used as a user-friendly frontend for creating neural networks.

Here’s a basic overview of how to use Keras to create a simple neural network:

1. Install Keras: First, you need to install Keras along with TensorFlow or another supported backend like Theano or CNTK. Since Keras is now part of TensorFlow, you can install it as follows:

2. Import Keras: Import the necessary modules from Keras to build your neural network:

3. Build the Model: Define the architecture of your neural network by stacking layers. You can use different types of layers like Dense (fully connected), Conv2D (convolutional), LSTM (recurrent), etc.

4. Compile the Model: Before training, you need to configure the model with loss function, optimizer, and optional metrics.

5. Training: Train your model on the training data using the fit function:

6. Prediction: After training, you can use the model to make predictions on new data:

This is just a brief overview of using Keras for building a simple neural network. Keras offers much more flexibility and functionality, including support for custom layers, loss functions, and optimizers.

Remember that Keras is now a part of TensorFlow, so if you have TensorFlow installed, you can directly use Keras without any additional installation.