Compare time series predictions of COVID-19 deaths

4.5
étoiles
12 évaluations
Offert par
Coursera Project Network
Dans ce projet guidé, vous :

Preprocess time series data for various machine learning models

Visualize time series data

Compare the time series predictions of 4 machine learning models

Clock2 hours
IntermediateIntermédiaire
CloudAucun téléchargement requis
VideoVidéo en écran partagé
Comment DotsAnglais
LaptopOrdinateur de bureau uniquement

In this 2-hour long project-based course, you will learn how to preprocess time series data, visualize time series data and compare the time series predictions of 4 machine learning models.You will create time series analysis models in the python programming language to predict the daily deaths due to SARS-CoV-19, or COVID-19. You will create and train the following models: SARIMAX, Prophet, neural networks and XGBOOST. You will visualize data using the matplotlib library, and extract features from a time series data set, perform data splitting and normalization. To successfully complete this project, learners should have prior Python programming experience, a basic understanding of machine learning, and a familiarity of the Pandas library. Note: This course works best for learners who are based in the North America region. We’re currently working on providing the same experience in other regions.

Les compétences que vous développerez

Time Series ForecastingMachine LearningFeature EngineeringPython ProgrammingTime Series Models

Apprendrez étape par étape

Votre enseignant(e) vous guidera étape par étape, grâce à une vidéo en écran partagé sur votre espace de travail :

  1. Preprocess the data using pandas to be ready for machine learning, and visualize the data using matplotlib

  2. Create a SARIMAX model, optimize the model hyperparameters, use the model for forecasting future COVID-19 deaths and visualize the results

  3. Create a prophet model and use the model for forecasting future COVID-19 deaths and visualize the results

  4. Create a function that extracts features for training the XGBOOST and a feedforward neural network models

  5. Split time series feature dataset into training and test datasets and perform data normalization

  6. Train an XGBOOST model and a feedforward neural network model, and finally compare the predictions of all the models covered in the project

Comment fonctionnent les projets guidés

Votre espace de travail est un bureau cloud situé dans votre navigateur, aucun téléchargement n'est requis.

Votre enseignant(e) vous guide étape par étape dans une vidéo en écran partagé

Foire Aux Questions

Foire Aux Questions

D'autres questions ? Visitez le Centre d'Aide pour les Etudiants.