From lefterissoulas@gmail.com Sat Dec 29 13:35:41 2012 Received: from mx.cims.nyu.edu (mx.cims.nyu.edu [128.122.49.99]) by mail.cims.nyu.edu (8.14.5+Sun/8.14.5) with ESMTP id qBTIZeiV002610 for ; Sat, 29 Dec 2012 13:35:40 -0500 (EST) Received: from mail-wg0-f53.google.com (mail-wg0-f53.google.com [74.125.82.53]) by mx.cims.nyu.edu (8.14.4+Sun/8.14.4) with ESMTP id qBTIZbm1016412 for ; Sat, 29 Dec 2012 13:35:37 -0500 (EST) Received: by mail-wg0-f53.google.com with SMTP id ei8so5404811wgb.32 for ; Sat, 29 Dec 2012 10:35:32 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:date:message-id:subject:from:to:content-type; bh=TrOXdQoIJgJ58BepaRDjzN/ouWUOukYbaM01B/Ke0Fo=; b=hEeMaG8s2EJgwrCl8++93LRR1ZdQhT5hpyky1umrxDymndYNn24ruLu/G42mMpMXkQ 8VMYb3d9dJXX7HgOhuaZ7hyFVKzIqHJg0pFL1B+YNiaKtL1cpGxl4OhnCgg4iENoF7mW YesScL3XnRRdMkE6Cx+ibL7q3EStm3bPjVnRAkcp4nKAqZVK1w7/awJti/WK44pi9L/N tRbGU/3YA3/HhCQnxiOJcR5OsvVxiUKe1OAt9HWRj4JfihoKu0/uz9Wfa0TtSWPocNDA AkbHgf8v0cAjw0rrUSjq3ZFnpwKFKwmBMb+xT6HnXNJVFbcmCjZGulquAK2j/SPRDqaO PT+A== MIME-Version: 1.0 Received: by 10.194.235.6 with SMTP id ui6mr59046941wjc.12.1356806131940; Sat, 29 Dec 2012 10:35:31 -0800 (PST) Received: by 10.216.172.147 with HTTP; Sat, 29 Dec 2012 10:35:31 -0800 (PST) Date: Sat, 29 Dec 2012 13:35:31 -0500 Message-ID: Subject: installation script and execution instructions From: Lefteris Soulas To: dennis shasha , Dennis Shasha Content-Type: multipart/alternative; boundary=089e01493bd085fd2c04d2020b96 X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.4.1 (mx.cims.nyu.edu [128.122.49.99]); Sat, 29 Dec 2012 13:35:37 -0500 (EST) X-Virus-Scanned: clamav-milter 0.97.4 at mx X-Virus-Status: Clean X-Scanned-By: MIMEDefang 2.73 on 128.122.49.99 Status: RO Content-Length: 6890 --089e01493bd085fd2c04d2020b96 Content-Type: text/plain; charset=ISO-8859-1 Dear Dennis, here is the installation script which installs the packages that are needed for the programs( and some more python packages that I find useful for future use). https://www.dropbox.com/s/wchg4qhxbgkcklu/installationScript those commands work in linux environment so I would suggest you to make a virtual machine or use a pure ubuntu distribution. For the SketchStatstream -------------------------------------- please download this new sketch statstream version from here: https://www.dropbox.com/s/kxep0j93q0lml6t/SketchStatstream.zip you may unzip it and enter the pack folder. then type "python main.py" a gui will pop up prompting you to enter the options you want to pass as arguments. specifically the proper way to press the buttons is 1) press "input stream file" , then select as input file this one https://www.dropbox.com/s/rdebr2lkegrducn/fi1.csv (after you save it somewhere in your computer) 2) press "output stream file", then select where you want the csv file with the correlation pairs to be saved 3) press "proceed" 4) enter 3 as the number of streams (I will do it as you told me so it can automatically read the number of the streams but I want to search it a bit more because python can let me read a csv from the beginning after I read certain lines and I must close it and reopen it which I don't find pretty -- anyway I will fix it the way you said) 5) enter 160 as the size of the streams 6) enter 10 as the basic window (you can enter whatever you want) 7) enter 50 as the sliding window (you can enter whatever you want) 8) enter 0.8 as the correlation threshold (you can enter whatever you want) Just wait and the program will finish and report the correlation in the csv output file and open the gui where you will see the results graphically. For the LearnStream -------------------------------------- please download this new learnStream version from here: https://www.dropbox.com/s/zbnh1n88ker757c/learnStream.rar you may unzip it and enter the pack folder. download and unpack the folder which has the data from Timir. https://www.dropbox.com/s/3u0xgfad926rno3/2012_05_27.rar now you have to edit line 8 of testWithRealData.py you must put the path where you have extracted the folder with the data I told you to download above.. presently its "/home/lefteris/Desktop/".. change it to whatever fits your path. save the file and then run this command inside the folder python testWithRealData.py the program will execute (the numbers you see is the round because it takes a while to finish and I print those number to keep me company :) ) after it finishes you will see the relative errors and the profits for all the three methods (golden method, plain sgd, our approach). I hope I haven't forgotten anything. I must go out for dinner but if you have any problems email me back and I will check it when I return. Hope you are doing well. How is your father's health? Best, Lefteris you must then type "python main.py"