Jump to content
Science Forums


  • Content Count

  • Joined

  • Last visited

  • Days Won


Poppins last won the day on December 12 2014

Poppins had the most liked content!

About Poppins

  • Rank
    Advanced Member

Profile Information

  • Gender
    Not Telling
  • Interests
    Computational Neuroscience
  1. Thanks for the help. I wish I knew a little more math. They did try to explain it a little bit in the url that I gave, but I'm a low level mathematician. I stick with arithmetic and basic algebra because that is what I know. I have been teaching myself Linear Algebra lately though.
  2. If anyone can explain this stuff in English to me I would be extremely grateful.
  3. I'm working on learning machine-learning as it is, not as I believe it to be, but I'm running into equations like the following. L1 norm: , which leads to sparse solutions.Elastic Net: , a convex combination of L2 and L1, where is given by 1 - l1_ratio. I have no idea what any of this means. I haven't taken math since my first year in college which was a Computer Science class type overview of Excel and Excel functions. You can find these equations here. I found an absolutely amazing resource on the internet just recently which led me to this alien scripture, you can find that reso
  4. If you tried to access the webpage yesterday it probably took forever to load. I'm using IP addresses to localize the inputs to states and ISP areas right now. I entered an IP from Norway and it took a while to check the database. Once it finally printed out the results, everything said "none". So, if you're from Norway, then in the eyes of my program you are from the city of None and the state of None, but I do think that you can still vote for the USA, which I will have to fix. I'm going to do that now.
  5. The program has been updated for local, statewide, and national inputs (although I think that it only works for USA right now).
  6. I don't understand what you mean here by tuples and n-tuples. I've already posted the algorithm here before as well. Here it is, essentially, but spiced up so that it's fun to look at if you put it in the python shell. while True: poi = raw_input('DNA> ') npoi = '' while len(poi + npoi) != 0: if len(poi) > len(npoi): print ' ' * len(npoi), '0=' + '=' * (len(poi)-len(npoi)) + '=0' else: print ' ' * len(poi), '0=' + '=' * (len(npoi)-len(poi)) + '=0' if len(poi) != 0: npoi = poi[-1] + npoi if len(poi) == 1:
  7. I tried one method to remove the fragments, but it ended up doing something like this. I need money -> became a fragment of "I need money now" -> became a fragment of "I need money now really bad". I guess I could tokenize the fragments and get rid of some of them that way, but it's still so early in its development that I haven't even thought about it. It's basically just a pattern recognizer, but the cool thing is that once it's where I want it to be it's going to be global. It will recognize any pattern, in any orthography, in any location. I'm going to be taking IP addresses an
  8. I left work yesterday and forgot to put it up overnight :l. I wish I remembered. It's back up now... for the time being.
  9. Here is the link to the page. I'm going to have it up for a little while. Maybe a few days as long as my computer doesn't slow down too much. I need testers for this. This is the first version of it, and it's not pretty but hopefully it does what it's supposed to do. So please, go to that link and enter some information into the text area and hit submit. Let's see what ends up happening. The goal is to get enough information from every zip code entered into the system so I can send it along to the politicians who are in office or running for office. I think that they would like to see what t
  10. If you guys got python try this one out, it's pretty fun to look at. poi = raw_input('>') npoi = '' while len(poi + npoi) != 0: if len(poi) > len(npoi): print ' ' * len(npoi), '||', '=' * (len(poi)-len(npoi)), '||' else: print ' ' * len(poi), '||', '=' * (len(npoi)-len(poi)), '||' if len(poi) != 0: npoi = poi[-1] + npoi if len(poi) == 1: poi = '' else: poi = poi[:-1] if len(poi) == 0: poi = npoi[1:] npoi = '' || ================== || || ================ || || ============== || || ============ || || ========== || || ======== || || ====== ||
  11. The generic code you posted looks almost exactly like R. I'm not familiar with the diff algorithm but I'm about to check it out. All I'm doing today is some fun database maintenance and making sure that my programs are gathering data correctly. I've gathered over 100,000 records over the weekend without supervision, pretty amazing huh?
  12. I made it print spaces instead of character for the poi and the npoi. while len(poi + npoi) != 0: print ' ' * len(poi), '||' print ' ' * len(npoi), '||' ... || || || || || || || || || || || || || || || || || || || || || || || || || || || || || || || || || || || || || || || || || || || || || || || || |
  13. Here is and example of the problem that I was trying to solve: The search was domestic violence and every search produces 10 results. The results would be something like this- 1. Burglary 2. Criminal Sexual Conduct 1st degree 3. Criminal Sexual Conduct 2nd degree 4. Criminal Sexual Conduct 3rd degree 5. Criminal Sexual Conduct 1st degree (Multiple variables) 6. Domestic Violence 7. Breaking & Entering 8. Operating a chop shop 9. Larceny from the person > $250 < $1000 10. Indecent Exposure What I wanted was a better list for all of the Domestic Violence charges that can be assoc
  14. To make this more complicated (and since I have the time), I'll calculate the example above. input = "Hi how are ya?" matching pattern = "Hi, how are you?" "Hi" is in the matching pattern, "how are y" is there too, and "a" and "?" is there which gives us the sum exponent of 4 + 387420489 + 1 + 1 = 387420495 in the bad pattern ("Hello ivan, are you ready to have some fun?") we only get "H"(1), "i"(1), "h"(1), "o"(4), "are y"(3125), "a"(3), "?"(1). This comes to a sum of 3136. Obviously not a good match if you're using exponents which therefor gives you a relevant response.
  • Create New...