Forums Tyrant

Automatic deck optimization - leftylink's thread page 7 (locked)

216 posts

Flag Post
Originally posted by thror13:
Originally posted by aquaworld:

Hi, I had the same Necrogeddon error as the person above, and I fixed it. However, whenever I attempt to use tyrant_optimize, I get this:

C:\MinGW\tyrant_optimize>tyrant_optimize.exe ogre-spam “objective” climb 10000 -
t 3 -s -o

C:\MinGW\tyrant_optimize>tyrant_optimize.exe ogre-spam" “objective” climb 10000
-t 3 -s -o

C:\MinGW\tyrant_optimize>tyrant_optimize.exe “ogre-spam” “objective” climb 10000
-t 3 -s -o

Once I press Enter, it says that tyrant_optimize.exe has stopped working and it crashes.
I have compiled my tyrant_optimize using catepillar’s guide on the first page, and I removed the entire unit for the Necrogeddon issue. I just copied and pasted the Custom.txt and ownedcards.txt from my old tyrant_optimize version (Commonkey’s) into the newly compiled tyrant_optimize. Also in the newly compiled tyrant_optimize, I still have all the H files and CPP files, I don’t know if they’re supposed to be there. Any help would be appreciated.

try
C:\MinGW\tyrant_optimize>tyrant_optimize.exe ogre-spam “Mission 100” climb 1000

if it works, the .exe is fine…. then add flags one by one, and recheck your txt’s and xmls
if it doesn’t work, recompile or get the already compiled .exe from the net.
if you recompile, make sure you get the latest version since there are a couple obsolete ones out there…

Originally posted by hunterhogan:
  • your third example is the only one that could possible work because both deck names must have quotes around them
  • not true…. deck names with no spaces work without quotes… therefore ogre-spam works, ogre spam does not, but “ogre spam” does…

    Thank you very much, I managed to get it to work. It’s a bit weird since I also tried my previous commands that crashed the program and they also worked, I guess it was just a bit buggy.

     
    Flag Post
    Originally posted by Scaler_:
    Originally posted by Puppenmaedchen:

    I couldn’t find a post on how to merge the updates, that’s what I was really asking, even if I probably phrased it wrong. Sorry ’bout that.

    First post: “https://github.com/leftylink/tyrant_optimize/archive/leftylink-integration.zip – This is the source code distribution that contains all my changes. It is automatically updated at the same URL whenever I update the leftylink-integration branch.”

    Really, just trust me that I actually can read and checked the thread and the links. Download the file you linked and check the date. That is the Dec. 03 release. Everything that was fixed/updated/implemented later (Dec. 19 release) is not in there – that is why I am asking how to get the updates merged. /sigh

     
    Flag Post

    Is there a command, which you can use to set not to optimize a slot in a deck (e.g. what you can use if your want to build around a specific card)?

     
    Flag Post
    Originally posted by riot99:

    Is there a command, which you can use to set not to optimize a slot in a deck (e.g. what you can use if your want to build around a specific card)?

    No.

     
    Flag Post
    Originally posted by Puppenmaedchen:

    The error message is this:

    The deck Step 1 was not found. Available decks:
    [then it lists all Mission, Raid and Custom decks]

    It is the same for all Steps.

    Edit: just to make sure, this is the command line that I use:

    tyrant_optimize “timesurgeopak” “Step 1” -o -r -q -t 1 climb 100000 > TimeSurgeOpak.txt (with correct version of the " of course)

    I’m having the exact same issue. Optimizer won’t recognize quests. I did exactly what catepillar’s instructions said. Optimizer seems to be fine otherwise.

     
    Flag Post

    tbh, i have once tried to recompile, i can confirm that from the front page you get the older version of the optimizer.
    but, bf i started searching for the correct files, i got the exe from a friend, and that was that….

    anyway here’s the .exe i’m using:
    http://www.2shared.com/file/AggUzdW1/tyrant_optimize.html

    you can get it from there and be happy…

     
    Flag Post

    I’m sure I’m going to sound so green but can someone please explain what “hill-climbing” means ?

     
    Flag Post
    Originally posted by ElvisTR:

    I’m sure I’m going to sound so green but can someone please explain what “hill-climbing” means ?

    You don’t sound green, but perhaps you sound a little LMGTFY

    First two sentences of the wikipedia article with a few words changed:

    In computer science, hill climbing is a mathematical optimization technique which belongs to the family of local search. It is an iterative algorithm that starts with an arbitrary solution (your deck) to a problem (a defense deck), then attempts to find a better solution by incrementally changing a single card at a time.

     
    Flag Post
    Originally posted by ElvisTR:

    I’m sure I’m going to sound so green but can someone please explain what “hill-climbing” means ?

    The optimizer changes one card for another, then checks if the win rate improved. If so, it keeps that change. This is relatively quick since it doesn’t test all possible combinations (which would take ages). But the method has its limitations.

     
    Flag Post

    Isn’t it true that against some defdecks, the offense deck with the highest winning % will have fewer than 10 cards? If so, doesn’t that limit the optimizer?

     
    Flag Post
    Originally posted by defklown:

    Isn’t it true that against some defdecks, the offense deck with the highest winning % will have fewer than 10 cards? If so, doesn’t that limit the optimizer?

    I think the optimizer can be asked to generate decks with < 10 cards.

     
    Flag Post
    Originally posted by defklown:

    Isn’t it true that against some defdecks, the offense deck with the highest winning % will have fewer than 10 cards? If so, doesn’t that limit the optimizer?

    Originally posted by purei:

    I think the optimizer can be asked to generate decks with < 10 cards.

    both quite true.
    to do it either try brute force mode, or try a smaller deck as attack with the climb algorithm.
    that way the resulting deck will be as large as its startup deck.

     
    Flag Post

    Yes, but it doesn’t check whether removing a card will improve the deck. So if you start with a 10 card deck, you will end up with a 10 card deck, even when a 3 card deck is better.

     
    Flag Post

    I have followed catepillar´s instructions in the first page. It compiled sucessfully, but it’s not working with hashes.
    Does anyone can give me a tip to what to do?

     
    Flag Post

    read the last page(s) of the thread?…

     
    Flag Post
    Originally posted by aboardbr:

    I have followed catepillar´s instructions in the first page. It compiled sucessfully, but it’s not working with hashes.
    Does anyone can give me a tip to what to do?

    hashes only work on command line

     
    Flag Post

    Thank you everyone for the quick reply!
    I am already used to the optimizer. An old version that I have works perfectly with hashes.
    Here is an exemple of command line that do not work:
    tyrant_optimize PoAB+k “Mission 1” climb 1000

    Here is a screenshot of what I am talking about:

     
    Flag Post

    Image and video hosting by TinyPic

     
    Flag Post

    Maybe we need a -v that shows version.

     
    Flag Post

    So what am I doing wrong? Maybe outdated source code?

     
    Flag Post
    Originally posted by aboardbr:

    So what am I doing wrong? Maybe outdated source code?

    You are running an outdated version. Are you sure you ran git checkout leftylink-integration with no errors?

     
    Flag Post
    Originally posted by catepillar:
    Originally posted by aboardbr:

    So what am I doing wrong? Maybe outdated source code?

    You are running an outdated version. Are you sure you ran git checkout leftylink-integration with no errors?

    When I run the command I receive this message:
    Branch leftylink-integration set up to track remote branch leftylink-integration from origin.
    Switched to a new branch ‘leftylink-integration’

    Is it right?

     
    Flag Post

    I’m also unable to use hashes (I’m using “windoze” and followed catepillar’s guide on page 1). I got the same thing as aboardbr did the first time I ran git checkout leftylink-integration . I ran it again to make sure and now I get:
    M Makefile
    Already on ‘leftylink-integration’

     
    Flag Post

    Having trouble? No warranties or guaranties. Use at your own risk.

    Originally posted by hunterhogan:
    Originally posted by DarkBlood1:

    Can anyone upload the latest version?

    I’m sure you, DarkBlood1, will not make me regret this, and I hope I don’t regret it for other reasons.

    I have no idea if this will work Compiled on Microsoft Windows 7 Home Premium.

    http://www.hunterthinks.com/solo/games/T_O_2012_12_19_Win_Compiled.zip

     
    Flag Post

    Every time I try to use this thing, I get the same error message every time:

    Any ideas?