+- Sinisterly (https://sinister.ly)
+-- Forum: Coding (https://sinister.ly/Forum-Coding)
+--- Forum: Python (https://sinister.ly/Forum-Python)
+--- Thread: Possibly the Next Generation of Artificial Intelligence. (/Thread-Possibly-the-Next-Generation-of-Artificial-Intelligence)
Pages:
1
2
RE: Possibly the Next Generation of Artificial Intelligence. - Misha- - 02-18-2015
Just dropping in to say that Python isn't the best language for doing this, since if it gets big, it's gonna get slow depending on the way you make it.
RE: Possibly the Next Generation of Artificial Intelligence. - Oni - 02-20-2015
(02-18-2015, 11:33 PM)Misha- Wrote: Just dropping in to say that Python isn't the best language for doing this, since if it gets big, it's gonna get slow depending on the way you make it.
Why would you be against Python? It seems like an okay fit, given the circumstances. Maybe not the best pick, but that is just our opinion.
RE: Possibly the Next Generation of Artificial Intelligence. - Equinox - 02-20-2015
(02-18-2015, 11:33 PM)Misha- Wrote: Just dropping in to say that Python isn't the best language for doing this, since if it gets big, it's gonna get slow depending on the way you make it.
(02-20-2015, 01:10 AM)Oni Wrote: Why would you be against Python? It seems like an okay fit, given the circumstances. Maybe not the best pick, but that is just our opinion.
Oni is right. Python is a great language (with given circumstances). If speed is ever a problem, multithread or multiprocess. Don't have enough memory on your computer to do so? Run it on a VPS. 9 times out of 10, a bot won't run on a home computer, because it can be "needy", if not programmed to an extreme that it is memory efficient.
And, to OP;
I support this idea, though I wouldn't use text files, much less encrypted ones. It takes a bit more to process them, and to decrypt them is more work. Of course, Python is only limited to how fast it can interpret the syntax, so instead of encrypted text files, I recommend you use a database. If this is going to be web based as well, isolate your variables from your queries so that SQLi isn't an issue.
Example of isolating your variables using the question mark param-style (SQLite3 param-style):
Code:
import sqlite3
s = sqlite3.connect("mydb.db")
x = s.execute("SELECT * FROM tablename WHERE id=?", (1))Isolating the variables requires a tuple, at least for the SQLite3 database type. I'm not sure of any others, since I've only messed with SQLite3, but I recommend using another db type, not to say SQLite3 is bad, it's just not the best.
RE: Possibly the Next Generation of Artificial Intelligence. - Misha- - 02-20-2015
(02-20-2015, 01:10 AM)Oni Wrote: Why would you be against Python? It seems like an okay fit, given the circumstances. Maybe not the best pick, but that is just our opinion.
I personally haven't made any large projects in Python, so I'm not exactly sure, but i've heard people say it's bad with big projects.
Just my opinion.
RE: Possibly the Next Generation of Artificial Intelligence. - Eclipse - 02-20-2015
(02-20-2015, 07:25 PM)Misha- Wrote: I personally haven't made any large projects in Python, so I'm not exactly sure, but i've heard people say it's bad with big projects.
Just my opinion.
“Most people are other people. Their thoughts are someone else's opinions, their lives a mimicry, their passions a quotation.”
― Oscar Wilde
RE: Possibly the Next Generation of Artificial Intelligence. - Misha- - 02-20-2015
(02-20-2015, 07:31 PM)Eclipse Wrote: “Most people are other people. Their thoughts are someone else's opinions, their lives a mimicry, their passions a quotation.”
― Oscar Wilde
>Their passion's a quotation.
So you decided to quote Oscar Wilde? o.o
Nice quote anyway.
RE: Possibly the Next Generation of Artificial Intelligence. - Eclipse - 02-20-2015
(02-20-2015, 07:32 PM)Misha- Wrote: >Their passion's a quotation.
So you decided to quote Oscar Wilde? o.o
Nice quote anyway.
You misunderstood the point of the quotation. And misunderstood the quotation.
RE: Possibly the Next Generation of Artificial Intelligence. - Misha- - 02-20-2015
(02-20-2015, 07:53 PM)Eclipse Wrote: You misunderstood the point of the quotation. And misunderstood the quotation.
I understood it, just wanted to point it out.
Anyway, let's not go too offtopic.
RE: Possibly the Next Generation of Artificial Intelligence. - Windows65 - 03-27-2015
I took a break from the AI for a while because I realized I was not going about it in the correct manner. I have read all of your comments and I will take your ideas into consideration, especially using something other than text files. I have been brainstorming AI since I started my break from the project and concluded this.
Why start with the speech based language? Why start from something that is difficult to grow upon? Also, why should AI mimic human behavior, something we know so little about?
Those three questions have helped me determine that I was on the wrong path for long term development. Yes, its cool to have a bot talk to you, but that's been done, and it has been done well. I want to make something with a practical purpose. I want to make something that is fun to make and will keep me and whoever else s working on it interested.
Here is what I want to build.
I want to build a Quad-copter that is controlled by AI, responds to situations, etc. Yes, this has been done but I am talking about a completely different level of complexity.
Benefits of this would be, sensory inputs/code has already been created and are open sourced, it is a very salable project, it is a fun and unique project.
The main downfall are expenses. Quadcopter parts are Very Expensive... Ranging from hundreds to even thousands. I am on a very small budget.
If you think this is a good idea I am willing to make a thread disusing in more depth. I will also let you know what I want to achieve by doing this project and the endless possibilities it will bring the the world. Let me know what you think.
[Also posted as an update on the main thread,]