This tutorial is an introduction to time series forecasting using TensorFlow. It builds a few different styles of models including Convolutional and Recurrent Neural Networks (CNNs and RNNs).
To use this data with a machine learning model, it is often useful to aggregate it into time series, where the data is sampled uniformly over time. For example, we could aggregate the sales weekly, or calculate the total sales in the last week for each day.
Distributed training Distribute your model training across multiple GPUs, multiple machines or TPUs. The Advanced section has many instructive notebooks examples, including Neural machine translation, Transformers, and CycleGAN.
For sequences other than time series (e.g. text), it is often the case that a RNN model can perform better if it not only processes sequence from start to end, but also backwards.
Note: This dataset has been collected and analysed during a research collaboration of Worldline and the Machine Learning Group of ULB (Université Libre de Bruxelles) on big data mining and fraud detection.
This document is the first in a two-part series that explores the topic of data engineering and feature engineering for machine learning (ML), with a focus on supervised learning tasks. This first part discusses the best practices for preprocessing data in an ML pipeline on Google Cloud.
For a time-series, the output for a time-step is calculated from the entire history instead of only the inputs and current hidden-state. This may be less efficient.
Learning how to deal with overfitting is important. Although it's often possible to achieve high accuracy on the training set, what you really want is to develop models that generalize well to a testing set (or data they haven't seen before).
