Getting Started
The following steps will guide you through the process of signing up and getting access to a free trial of the Epistemix Platform. Access to the Platform IDE is free for all users, and trial users benefit from full access to our US 2010 Synthetic Population for one month. Once you've signed up, don't forget to explore the Quickstart Guide to start learning about what you can do with the Platform.
Signing Up¶
If you do not already have credentials, you will need to sign up to access the platform.
Navigate to www.epistemix.com/platform, click the sign up button at the bottom of the page, and submit the form.
Within a few minutes of submitting the sign up form, you will receive an email with your credentials and further instructions.
Platform Hub¶
Once you have your credentials, using your web browser, navigate to platform.epistemix.com.
Enter your username and password.
Congratulations, you have successfully signed into the Platform Hub!
From here, the Platform Integrated Development Environment (IDE) can be launched, and you can get start building agent-based models!
You will also find links to the language guide, reference pages, our blog, and our community forum.
At the bottom of the Platform Hub page, you will find information regarding which synthetic populations are available to you, as well as the number of Compute Credits that you have remaining.
Platform IDE¶
When you click the "Launch IDE" button within the Platform Hub, you will be taken to a web-based integrated development environment (IDE). When our cluster has finished allocating resources, you will see an interactive Jupyter notebook containing the synthetic population that is available to your account and the FRED software used to run agent-based simulations using that population.
For next steps, see the welcome text presented in the Platform IDE (also available here). In particular, be sure to look at the Introductory Models and Quickstart Guide. The Quickstart Guide can be accessed by clicking the Quickstart Guide folder on the left, then lesson 001-..., and lastly, the file ending in .ipynb.
Synthetic Populations¶
During your free trial, you will have access to the full US synthetic population on FRED Cloud. Additionally, a limited set of counties from the US synthetic population is available for use in simulations within the Platform IDE itself. These counties are listed in the table below. We are working on providing access to the full US synthetic population within the Platform IDE, and this will be available very soon!
| County | State |
|---|---|
| Park | Colorado |
| Clarke | Georgia |
| Butte | Idaho |
| New York | New York |
| Lenoir | North Carolina |
| Erie | Pennsylvania |
| Jefferson | Pennsylvania |
| Loving | Texas |
| Grand Isle | Vermont |
| Dane | Wisconsin |
| Kewaunee | Wisconsin |
Once you have finished your free trial, you will lose access to the US synthetic population. See synthetic population pricing for subscription options to retain access to the synthetic population.
FRED Cloud¶
The Platform IDE includes all the tools and resources you need to develop, test, and analyze the results of agent-based simulations models. Once you have developed a model, the next step is often to run it many times using different parameter values or input data to explore the effect of different simulated scenarios on the model outputs. This is where FRED Cloud comes in.
FRED Cloud provides the ability to submit batches of simulations to run on Cloud infrastructure that is independent of the Platform IDE environment. This makes it possible to run large numbers of simulations in parallel, as well as making it convenient to run simulations that are expected to take more than a few minutes to complete. Additionally, FRED Cloud makes it possible to run simulations with populations much larger than is possible within the Platform IDE. With FRED Cloud, you can specify the simulations you want to run from within the Platform IDE, and then log off to go and do something else. When you log back in to the Platform IDE after the simulations have finished running, you can easily retrieve the results of the simulations that were run on FRED Cloud, ready for subsequent analysis. A typical workflow is to develop the a model and quickly run it on a small population within the Platform IDE, and then run it 'at scale' on FRED Cloud using the larger population that is of interest to your project.
Usage of FRED Cloud is charged against pre-paid Compute Credits. As part of your free trial, you will be provided with 100 Credits to help you get started. Once you have used these, you can purchase more through the Platform Hub. See compute credit pricing for details of how Compute Credits are charged against simulation runs on FRED Cloud. For practical examples of how to use FRED Cloud, see this tutorial in the Quickstart Guide.
If you would like to use FRED Cloud in your project but are having problems, we want to help in any way we can. Please schedule a support call with one of our developers so we can talk through your problem.