Posts Mentioning RSS Toggle Comment Threads | Keyboard Shortcuts

  • Varchas 4:18 pm on December 25, 2009 Permalink | Log in to leave a Comment  

    SVM Classifier Released On Amazon Ec2 

    We are very excited to announce a Support Vector Machine (SVM) classification algorithm available on our free public cloud machine as well as free public AMI. It is a 2-class classifier which implements the SMO optimization algorithm. The salient features are as follows:

    • Implements 2 class Sequential Minimal Optimization (SMO)
    • Uses 1305 file formant and thus can handle categorical and sparse data formats
    • Generated models can be saved and used for classification at any time
    • Offers 2nd order working set selection and caching techniques for speed

    Our benchmarks beat the popular LibSVM and SVM-Light tools for standard problems from the UCI machine learning datasets repository. Please find documentation and samples on how to run the SVM here.

     
  • aToF 6:20 pm on December 15, 2009 Permalink | Log in to leave a Comment  

    How to launch an AMI, a video example 

    If you find the instructions hard you can always take a look at the following videos about how to launch an Amazon EC2 AMI and most important, how to TERMINATE  it once you are done, otherwise you will get unwanted charges.

    Launching an Amazon EC2 AMI

    Terminating an Amazon Ec2 AMI

     
  • aToF 1:24 pm on December 8, 2009 Permalink | Log in to leave a Comment  

    Amazon EC2 AMI 

    The current public AMI is:

    ami-d70fe1be

    There is no “root” access to our AMI’s. The user name is “a1305″ and you can use your own Amazon Ec2 account’s private key to ssh. Commonly the command will look like:

    ssh -i <path to private key> a1305@<Public DNS of launched cloud machine>

    Currently we only support 32-bit configurations. Please use the m1.small type machine.
    (More …)

     
    • Nawwar 3:56 pm on March 28, 2010 Permalink

      First of all, very nice job. Your library and AMI will be very helpful for me. So thank you.
      But I wonder why there’s no root access? is there a reason for that?
      I need to mount an EBS volume, and that requires root!

  • aToF 12:50 pm on December 8, 2009 Permalink | Log in to leave a Comment  

    Online cloud machine for testing 

    Since some of you might want to quickly test the 1305 library, we have launched an Amazon EC2 machine, so that you can connect and test our algorithms. The ssh command is:

    ssh a1305@ec2-67-202-56-237.compute-1.amazonaws.com

    and use sixtyfive as the password.

    Notice that this is a public machine that many people might be using. So make sure you don’t upload sensitive data. It is also recommended that you create a folder within the home directory “/home/a1305/<your unique directory name here>” and work from within this folder. You might also experience slow performance because other people might be submitting jobs. If you need a machine that you have exclusive access follow these directions.

    For help and support either send emails to cloud@analytics1305.com or post your question on the forums.

     
    • aToF 7:15 am on May 28, 2010 Permalink

      UPDATE: We have bought this machine down since we are now providing binaries you can download. Thanks to those who used it!

      - Analytics1305 Developer Team

  • aToF 4:35 pm on December 7, 2009 Permalink | Log in to leave a Comment  

    Quick & Easy- Get Started on Amazon Cloud 

    The Amazon Ec2 cloud service is great. Even though I have access to a few rather large boxes the idea of having as many compute servers as I like within minutes is very exciting. Amazon also recently announced the high memory instances which give you 68 GB of memory, I shifted in my chair as I wrote that. Still one of the issues is sifting through the Amazon documentation and multitude of tools to get up and running to the point where you have ssh’ed into your machine and are ready to have it serve your will. Here I present a quick way to get from having just heard of the Amazon Ec2 service to being at the console of your first Ec2 instance. This article assumes you are running some form of linux.

    What is Amazon Ec2?



    The Elastics Compute Cloud (Ec2) is an Amazon Web Services (AWS) offering which let’s you start remote computers and have them do whatever you want done.

    What is an AMI (Amazon Machine Image)?

    An AMI is a snapshot of a machine. Imagine taking check-pointing everything in your machine right now and then being able to load it exactly as you left it on multiple computers. Thus, when you start a machine on Ec2 you actually load an AMI, the snapshot of some system taken by someone (either Amazon, or some other trusted source).

    What is an Instance?

    An instance is a running machine. There are different types of instances and instances differ in the hardware available like memory, cpu’s etc.

    Now let’s dive into getting up and running on the Amazon Cloud!

    Get an amazon Ec2 account

    You will need:

    1. A credit card – to pay for the machines you start up. The machine we will start here costs between 8 and 10 cents per hour. There are also some data transfer charges but for the purpose of this article your cost will be in the same order.

    2. An email address and telephone- if you have ever ordered through Amazon you can use the same email address. They will verify some details and you’re done.

    Go here and click “Sign up for Amazon Ec2.”.

    Launch the machine

    1. Go to the AWS console (click here) and click “Launch Instances”.

    2. Choose AMI- On the pop up screen click on the “Community AMI’s” and in the text box enter the AMI you want to launch. The Analytics1305 AMI can be found here and when you enter it you will see our AMI (Notice that the ami id is subject to change so always look at this page to find out the latest one). Then click “Select”.

    3. Create Key Pair- In the next step you will be asked to enter the name of your key pair. If you have never created one click on “Create and Download Key Pair”. Let us name the key pair “gsg-keypair”, put that in the text box. You will be asked to download a file called “gsg-keypair.pem”, Download it and keep it somewhere safe. You will only need this file when you actually ssh. Also, you should change the permissions on the file.

    The command “chmod 0600 gsg-keypair.pem” should do the trick.

    4. Configure Firewall- In the next step you will be asked to name a security group. Click on skip this step at the bottom.

    5. Launch – Now you get to the final screen before launch. Put 1 in the “Number of Instances” box and select gsg-keypair in the “Key Pair Name” drop down. Also select and “Instance Type” as small (the default). Also select “Security Group” as “default”. Click launch.

    Click “Close” on the message screen.

    Now in the console click “0 Running Instances”. You should see the instance you have launched “starting” or “pending”. This means the machine is booting up. Give it a minute or so and it will go to “running” state. You can refresh your browser to see the current state.

    Click on the instance you just ran and in the information that shows up below look for the “Public Dns”. This will look something like ec2-133-122-188-111.compute-1.amazonaws.com. Note this down.

    SSH into your machine

    1. Now you need to enable ssh on the machine. Click on “Security Groups” on the left pane of the console. Click on the “default” security group.

    2. At the bottom there will be information displayed about the group. Click on the drop down box that says “Custom…” and select “SSH” and then click “Save”.

    3. Finally you can ssh into the machine. From the console type:

    ssh -i <path to gsg-keypair.pem> a1305@<Public Dns of machine>

    When it asks you to verify the RSA key finger print type yes and return. Congratulations. You are now logged into an Amazon Ec2 cloud machine with Analytics1305 libraries pre-installed.

    To terminate the machines. Select the machine in the console by clicking the checkbox next to the machine. Then click “Instance Actions” and then choose “Terminate” from the drop down menu. Click “Yes, Terminate”. The machine’s status in the conslole will go to “shutting-down” and eventually “terminated”.

    For questions and support send and email to cloud@analytics1305.com or post on the forum.

     
  • aToF 10:44 pm on December 5, 2009 Permalink | Log in to leave a Comment  

    Cloud 1305 goes live !! 

    Dear Friends.

    The first version of our software library 1305 goes live on the Amazon EC2 cloud. For the moment 3 algorithms are available

    1. Neighbor search
    2. Kernel Density Estimation
    3. K-means clustering

    For more information and instructions visit http://www.analytics1305.com/documentation/

     
c
compose new post
j
next post/next comment
k
previous post/previous comment
r
reply
e
edit
o
show/hide comments
t
go to top
l
go to login
h
show/hide help
esc
cancel