What is the ADAPA Control Center and how can I access it?

The ADAPA Control Center gives you the power to control all your ADAPA instances in Amazon EC2 (Amazon Elastic Compute Cloud) from a single location.



In order to access the ADAPA Control Center (as well as the ADAPA Console), you first need to purchase ADAPA through Amazon. In order to do that, you will need an account with Amazon. Once you purchase ADAPA, you will also be required to subscribed to the Amazon Elastic Compute Cloud (Amazon EC2) service (by creating an Amazon Web Services account). Once that process is complete, you will be directed to the ADAPA Control Center (see blog on how to sign in for the paid version of ADAPA).

Note that you only need to create your Amazon Web Services (AWS) account once.

How do I get the access keys required to run the paid version of ADAPA in Amazon?

Follow the link below:

http://aws.amazon.com

Log in to your account (using your Amazon ID and password), move your cursor to the top of the yellow button "Your Web Services Account" on the top right corner of the AWS welcome page and select "AWS Acess Identifiers." You will be faced with a page that looks like this:



Note that we have covered the Access Key ID in the picture, but you will be able to see your own key in the box named "Your Access Key ID:" located on the right side of the page. Underneath it, you will find the box "Your Secret Access Key:" where you can retrieve your secret access key by clicking on "Show." You can also generate a new secret access key by clicking on the yellow button "Generate." As you probably already found out, Access Key ID and Secret Access Key are required by the ADAPA Control Center in order for you to be able to launch ADAPA instances. Copy the keys from this AWS page to the ADAPA Control Center.

How do I manage my Amazon Web Services account?

The first time you subscribed to ADAPA Predictive Analytics Edition, you were required to create an Amazon Web Services (AWS) account. You can manage this account and view all the charges you have incurred when using ADAPA (once you log in). For that, follow the hyper link below:

http://aws.amazon.com

The page you will encounter looks like this:



This is the welcome page to Amazon Web Services. The most important thing here is yellow button on the top right corner of page that says "Your Web Services Account." Once you move your cursor to the top of this button a menu will be shown with the following options:

1) AWS Account Activity
2) AWS Access Identifiers
3) Payment Method
4) AWS Usage Reports
5) Your AWS Profile

If you select the first option "AWS Account Activity," you will be faced with a page that looks like this:



This is a very important page, since it is in here that you will be able to see how much you own Amazon for using ADAPA. You can view details of your EC2 usage by clicking on the hyper link provided "View/Edit Service."

If you select the second option "AWS Access Identifiers," you will be faced with a page that looks like this:



This is also a very important page since it contains the access identifiers that you will need to create ADAPA instances in the ADAPA Control Center. Noticed that on the right of the page you will find the Amazon Access Key ID and the Access Secret Key (click on the + sign to see this one).

Feel free to browse through the other pages (available in the drop-down menu). They contain more information about EC2, such as your profile and payment method. Go to these pages to modify or view your personal information.

How do I manage my ADAPA subscription with Amazon?

Once you successfully subscribe to ADAPA through Amazon AWS (Amazon Web Services), you can manage your subscription by clicking once more on the hyper link provided at the "How to Buy" page of the Zementis website.

Once you sign-in (and considering that your ADAPA subscription is valid), you will be shown the following header followed by the ADAPA Control Center.



Note on the header the hyper link to view or change your ADAPA subscription. Click on it. Once you do, you will be re-directed to a page that looks like this:



This pages (named "Application Billing") shows pricing information about your ADAPA subscription. It also allows you to cancel your subscription as well as transaction and account history. Note that you will not be charged unless you actually use ADAPA. In case you select to cancel, click on the the link available on the bottom of the page and follow the instructions (for more on how to cancel your subscription, see information at the bottom of this blog).

Note that there are two other tabs you can select on top of this page: "Payment Method" and "Application Activation." If you select "Payment Method," you will be faced with a page that looks like this:



Note that here you can view your current payment method or enter a new one. If you decided to do that, click "Continue."

If you select tab "Application Activation," you will be faced with a page that looks like this:



This page mainly describes the process to get an Amazon Activation Key. However, you do NOT need an activation key to execute ADAPA. So, this page can be disregarded.

If you decided to cancel your ADAPA subscription, go back to "Application Billing" and click on the hyper link available on the bottom of the page. Once you do that, you will be faced with a page that looks like this:



Note that as the message in this page says, if you choose to cancel your ADAPA subscription, you will no longer be able to use it. Which is obvious, but still worth mentioning in case you are wondering (just kidding!). Click on "Cancel this Application" to go ahead with the canceling or "Keep this Application" to keep it as is. If you go ahead with canceling, you will be faced with a confirmation page which looks like this:



To subscribe again, go back to the "How to Buy" page of the Zementis website and click on the hyper link provided in there for the Amazon service.

ADAPA's computed value is not the same as the expected value, why is that?

There are many reasons for the validation test to fail. I can think of two main reasons.

1) The model ADAPA loaded and executed may be different than the model you built in your development environment. This may reflect a problem with ADAPA or see below.

2) It may be the case that the PMML file you got out of your model development environment does not really represent all aspects of the model or is problematic semantically speaking.

In both cases, you can try to follow ADAPA's decisions by clicking on the computed value which is a hyperlink and follow through its log of computations which are displayed as a text file. This may be very helpful in determining why ADAPA generated the value(s) it did.

Also, the problem may have to do with your data validation file itself. It may be the case that you generated your model in SPSS, for example, exported it as a PMML, converted it using the iGoogle converter and uploaded it into ADAPA. So far so good, but how about the data? If you saved your data in SPSS as well, you have to make sure you saved the expected value or prediction with the correct name. SPSS usually calls this value "PRE_1." You will need to change the name of this variable to the name of the predicted variable defined in the PMML file. Also, if your data contains the original target used to build the model, you will need to rename it to something different than the predicted variable. Your new predicted variable now should be the predicted result you got out of SPSS or any model development environment you used to score the data in the first place.

When uploading PMML example files into ADAPA, I get computed and expected values. What do they mean?

Each of the examples listed in the Zementis website is composed of two files: a PMML model file (.xml) and a validation file (.csv).

Given that the models have been built in a tool other than ADAPA, we want to make sure that both development tool and ADAPA produce the same results. This is done by supplying ADAPA with the expected results for a number of input records. When this happens, ADAPA will automatically compare the given expected value with its own computed value. If the both values match for all records (and given enough validation records), we can feel confident that ADAPA has uploaded the model correctly. When this happens, there is no longer the need to supply ADAPA with the expected results, since all we really want from now on is to get the computed results back.