5.0 / 5
We will cover the following tasks in 49 minutes:
- A look at what we will create.
- A look at the development environment.
Importing TensorFlow and the Dataset
- Importing Tensorflow and Keras.
- Importing the Boston Housing dataset.
- Splitting the data in training and test sets.
- A look at the shape of the data.
- Shuffling the training data randomly.
- A quick description of some of the features.
- Creating a dataframe from the training set using pandas.
- A look at the training labels.
- Normalisation of data to scale various features.
- A look at the normalised data.
Creating the Model
- Creating a Sequential model with Keras.
- Adding layers to our neural network model.
- Compiling the model.
- Looking at the model summary.
Training the Model
- Training by fitting the training set to the model.
- Model evaluation on the test set.
- Plotting Mean Absolute Error and Validation Mean Absolute Error values from the training process.
- Training a new model with the “Early Stop” callback from Keras.
- Plotting Mean Absolute Error and Validation Mean Absolute Error values from the training process on the new model.
- Evaluating the new model on the test set.
- Getting predictions from the model on the test set.
- Creating a scatter plot of actual labels vs predictions.
- Creating a histogram of the prediction error for the test set.
About the Host
I have been writing code since 1993, when I was 11, and one of the first things I remember I wrote was a database management software for a local hospital. More recently, I wrote an award winning Chatbot for a multi-billion-revenue company! In the last one year, I've been learning Machine Learning, NLP and Deep Learning and I'm very excited to share my learnings with you with the help of Rhyme.com !