From the course: Python for Time Series Forecasting

How to use Codespaces - Python Tutorial

From the course: Python for Time Series Forecasting

Start my 1-month free trial Buy for my team

How to use Codespaces

“

- [Instructor] To access the materials and understand the structure of the notebooks that you will work with during this course, I'll show you in this tutorial. First, we launch a Codespace, which is a virtual computer with all of the Python packages installed, ready for you to execute. Click on Code, create a new codespace. Now, let's wait until the virtual computer installs all of the programs. To view the progress, you can click on the bell and Building codespace to see how the programs are downloaded and installed. Wait for this command to finish. Now that everything is up to date, let's close this. Activate to get the icons with the folders in the bell, activate the extension vscode-icons. We can close this README, and the notebooks that will work during this course are in the folder notebooks. There are three main blocks, which are the first for foundation, loading, pre-processing, and the composition of time series data. Then we have block 2 for modeling, with the many ones you see here: ARIMA, SARIMA, Smoothing, and Prophet, as well as key concepts such as data stationarity and the metrics to evaluate which one of these is the best model. And finally, the validation, to simulate real-world scenarios on how time series can be tested over time on running Windows. Inside each one of these folders, you will find the notebook with the codes already written. So that you can go to the cells of codes, press Shift + Enter and execute. At the beginning, you will see to choose Python Environments. You can select the global env here and keep executing with Shift + Enter to see that the results are matching with what you learn in the video. Now for the assignments, which is the most important part because you apply what you have learned, there is one on each block. Here, you have the Assignment, which follows the approach of having a template file containing the general steps for you to substitute on the uppercase variable based on the type of data you'd like to work with, and execute the steps following the changing of variables, as they are in uppercase, to validate the results and get your time series forecasted into the future. Now, let's load the batteries because you're going to learn step by step how to apply the Python codes on time series to forecast and plan for the future.

Contents