Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[expwnent]

It's the correlation coefficient, but due to weirdness, sometimes it's more than 1. Not sure quite what's going on there. I think it has to do with weirdness with discrete vs continous variables? Not sure.

[lorb]

If that's supposed to be the correlation coefficient (covariance divided by product of standard deviations) something went awry because that is always in the interval [-1,1] whereas here we have values above two and once even -57 (at status.current_soul.preferences.9.item_subtype)

If you don't mind sharing the source i don't mind taking a look at it for the error. (depending on what it is written in of course .. i am not going to debug brainfuck code or other such weirdness)

[expwnent]

I fixed it.

Thanks for the offer, but right now the code is horribly illegible. And I fixed it anyway.

[lorb]

That's great.

By the way. Do we know what this is? Because if the kid inherits this in next to all cases from father _and_ mother that means both parents have next to always the same value here. That leads to the conclusion that either there is only one possible value and all the dwarfs in the data have that (is the data from just one or multiple fortesses? is it from just one or multiple genned worlds?) Or there are multiple values possible but only those that have the same value can become a couple.
 
appearance.colors.3
    mother: 0.999994
    father: 0.999994

edit: besides that i do not know what the data looks like but maybe something like a chiČ-Test or Goodmans lambda would give us a better clue whats going on?

[expwnent]

I am told that within a civilization, all dwarves have the same hair and eye color.

There are three independent saves mixed in to the data, with several dozen to several hundred dwarves in each. Three isn't very many, so it's possible there's some interesting stuff there that isn't showing up.

[lorb]

I am told that within a civilization, all dwarves have the same hair and eye color.

There are three independent saves mixed in to the data, with several dozen to several hundred dwarves in each. Three isn't very many, so it's possible there's some interesting stuff there that isn't showing up.

How independent? If they are from different worlds the dwarfs can't be from the same civilization.

[expwnent]

Code: [Select]

appearance.colors.3
    mother: 1
    father: 1
    random: 0.38512
Probability of it being the same as your parents = 1

Probability of it being the same as a random dwarf, across all saves is about 1/3, what it should be.

Indicates that it's the same within a save.

[lorb]

Nice one
So we are able to compile a list of traits that will be the same for every dwarf in a given civilization/world/fortress?

[expwnent]

The following traits are always the same within a given world:

Code: [Select]

appearance.colors.1
appearance.colors.2
appearance.colors.3
appearance.genes.colors.2
appearance.genes.colors.3
appearance.genes.colors.4
appearance.genes.colors.5
appearance.genes.colors.6
appearance.genes.colors.7
status.current_soul.preferences.0.color_id
status.current_soul.preferences.0.creature_id
status.current_soul.preferences.0.item_subtype
status.current_soul.preferences.0.item_type
status.current_soul.preferences.0.mattype
status.current_soul.preferences.0.plant_id
status.current_soul.preferences.0.shape_id
status.current_soul.preferences.0.type
status.current_soul.preferences.1.color_id
status.current_soul.preferences.1.creature_id
status.current_soul.preferences.1.item_subtype
status.current_soul.preferences.1.item_type
status.current_soul.preferences.1.mattype
status.current_soul.preferences.1.plant_id
status.current_soul.preferences.1.shape_id
status.current_soul.preferences.1.type
status.misc_traits.0.id
The following are always the same, period (note that I only have three worlds: very possible false positives):

Code: [Select]

appearance.colors.1
appearance.colors.2
appearance.genes.colors.2
appearance.genes.colors.3
appearance.genes.colors.4
appearance.genes.colors.5
status.current_soul.preferences.0.color_id
status.current_soul.preferences.0.creature_id
status.current_soul.preferences.0.item_subtype
status.current_soul.preferences.0.item_type
status.current_soul.preferences.0.mattype
status.current_soul.preferences.0.plant_id
status.current_soul.preferences.0.shape_id
status.current_soul.preferences.0.type
status.current_soul.preferences.1.color_id
status.current_soul.preferences.1.creature_id
status.current_soul.preferences.1.item_subtype
status.current_soul.preferences.1.item_type
status.current_soul.preferences.1.mattype
status.current_soul.preferences.1.plant_id
status.current_soul.preferences.1.shape_id
status.current_soul.preferences.1.type
status.misc_traits.0.id
With only three worlds I'm not ready to call that significant yet. And I bet a lot of them are only the same because they're set to -1 or some form of "does not apply".

[slink]

I am told that within a civilization, all dwarves have the same hair and eye color.

There are three independent saves mixed in to the data, with several dozen to several hundred dwarves in each. Three isn't very many, so it's possible there's some interesting stuff there that isn't showing up.

My fortress of 256 Dwarves demonstrated that hair and eye color is not always the same within a civilization.  It's a vanilla fortress except for renaming the sizes of prepared meals, and I use no graphical tileset so that's not a secret source of modding either.