Yaya-Cout
169fb7404e
Fix spelling ( #128 )
...
* Fix spelling in .cpp files
* Fix spelling in all files
2022-01-20 17:21:35 +01:00
Arthur Camouseigt
018dd91ca1
[Probability] Changed distribution parameter's type to double
...
Plots are still rendered in float but computations are now in double
Change-Id: I7e0a38effb780861b1443ee92a097cd319de3bc8
2020-11-04 14:45:35 +01:00
Léa Saviot
9b2f91f5e1
[apps/probability] Geometric distribution default computed value is 1
2020-06-04 14:50:07 +02:00
Léa Saviot
a9c47a9468
[apps/probability] Change geometric distribution definition
...
THe distribution now represents the number of trials neded before a
success, so is defined for k in {1, 2, 3, ...}
2020-06-04 14:50:07 +02:00
Léa Saviot
d200688f20
[apps/proba] Geometric distibution should return 0, not nan for x < 0
2020-06-04 14:50:07 +02:00
Léa Saviot
503227b49b
[apps/proba] Fix float/integer comparison
2020-06-04 14:50:07 +02:00
Romain Goyet
84768472bd
Explicit std::min/max template usage
...
This is unfortunately required in several cases:
- Sometimes when we use either float and double (this should be changed)
- Because KDCoordinate is not an int, so any arithmemtic promotes it to
an int
- Because we mix pointer differences and ints
2020-04-14 09:31:02 -04:00
Romain Goyet
950862f8d0
Use std::min and std::max
2020-04-14 09:31:02 -04:00
Neven Sajko
c0c73accef
[probability] Prevent soft-float in StudentDistribution::lnCoefficient
2020-04-09 13:58:56 +02:00
Neven Sajko
c92b770112
Performance fixes relating to floating point constants
...
This replaces unnecessary double-precision soft-float operations with
single-precision floating-point operations, mainly by casting.
In a couple places I also replace a function call with a constant.
2020-04-09 13:58:56 +02:00
Léa Saviot
29b4c75227
[apps/probability] Assert discrete distributions override evaluation
2020-03-16 11:40:40 +01:00
Léa Saviot
5c70fdc7a6
[apps/probability] Handle a == b case in finite integral computation
...
For non continuous distributions, P(a <= X <= a) is not necessarily
null.
2020-03-16 11:40:40 +01:00
Léa Saviot
0260e2ffb4
[apps/proba] Fisher distribution authorized parameter values
2020-02-12 15:13:21 +01:00
Léa Saviot
8c801392a5
[apps/proba] Fisher distribution zoom values
2020-02-12 15:13:21 +01:00
Léa Saviot
55a4e3d94c
[apps/proba] Implement Fisher formulae
2020-02-12 15:13:21 +01:00
Léa Saviot
93c96186c4
[apps/probability] Initial Fisher distribution
2020-02-12 15:13:21 +01:00
Léa Saviot
7406fcae98
[apps/proba] Fix Student distribution
2019-09-09 15:11:20 +02:00
Léa Saviot
29c3e3a206
[apps/distribution] remove falty assertion
2019-09-09 15:10:50 +02:00
Léa Saviot
aa45adb49a
[apps/proba] Handle infinite xMax in exponential distribution
2019-09-09 11:25:20 +02:00
Léa Saviot
dd97b23624
[apps/proba] k limit for Chi2 law
...
After the limit, P(X<k) returns undef
2019-09-06 10:40:59 +02:00
Léa Saviot
80f3c1fd0a
[apps/proba] Use DBL_EPSILON in cnditions
2019-09-06 10:23:32 +02:00
Léa Saviot
d9e49cb369
[apps/proba/dist] More laxist assert
2019-09-06 10:23:32 +02:00
Émilie Feral
de2f79b915
[probability] Distribution: loosen constraints on
...
cumulativeDistributiveInverseForProbabilityUsingIncreasing... precision
2019-09-05 13:59:30 +02:00
Émilie Feral
db4cfb1763
[poincare] Solver: increase IncreasingFunctionRoot precision
2019-09-05 13:27:18 +02:00
Émilie Feral
22d49f9a14
[apps/probability] Fix typo
2019-09-05 12:06:48 +02:00
Léa Saviot
1b169e0836
[apps/solver] Fix bad assertion
2019-09-04 17:34:50 +02:00
Léa Saviot
1c34eab71e
[poincare/solver] Fix precision handling in IncreasingFunctionRoot
...
Scenario: ChiSquared distribution, P(X<?) = 0.1
2019-09-04 16:55:39 +02:00
Léa Saviot
b712338a18
[apps/proba] Fix dome distribution behaviours for x = infinity
2019-09-03 10:10:23 +02:00
Léa Saviot
7f2e52f6a6
[apps/proba] Fix result consistency
...
Scenario: Poisson distribution, lambda = 4, P(x < ?) = 1.3E-112
2019-09-02 17:56:56 +02:00
Léa Saviot
c71ef7ff79
[apps/graph] Evaluation methods renaming
2019-08-29 11:07:22 +02:00
Léa Saviot
7ea7ecd3c2
[poincare/coordinate2D] Rename member values to x and y
2019-08-29 11:07:22 +02:00
Léa Saviot
a15bc53e06
[poincare/Coordinate2D] Template for double or float
2019-08-29 11:07:22 +02:00
Léa Saviot
3dd5112a0b
[poincare] Use reg incomplete beta function in binomial distribution
2019-08-27 11:09:16 +02:00
Léa Saviot
a8a5ecfd4e
[apps/proba] Factorize code
2019-08-26 16:04:08 +02:00
Léa Saviot
1f2f5f4c28
[apps/proba] BinomCDF
2019-08-26 15:04:27 +02:00
Léa Saviot
ff0105182d
[apps/proba] P(X<?) = 0 gives 0 for ChiSquaredDistribution
2019-08-26 10:14:38 +02:00
Léa Saviot
294def02bd
[apps/proba] Improve Distribution::finiteIntegralBetweenAbscissas
2019-08-23 15:27:53 +02:00
Léa Saviot
0840ca7a1e
[apps/poincare] Move NormalDistribution to Poincare
2019-08-23 13:47:52 +02:00
Léa Saviot
a1107b3f00
[apps/probability] Remove TODOs
2019-08-22 10:48:40 +02:00
Léa Saviot
4e8a7bbedc
[apps/proba] Increase accuracy check in tests
2019-08-21 14:54:55 +02:00
Léa Saviot
cfc0a7d6cf
[apps/proba] Fix Student computations
2019-08-21 14:52:47 +02:00
Léa Saviot
c248363b34
[apps/proba] Better handling of NAN values
...
For instance, chi squared law woth 1E8 degrees of freedom, comppute
P(X<?) = 0.4 will give P(X<undef) = undef
2019-08-21 14:47:17 +02:00
Léa Saviot
83fda9a587
[apps/proba] Fix Chi Square computations
...
Because the values can be very small or very big, computations should
not be made sequentially, to prevent rounding errors.
For instace, for degrees of freedom = 70, coefficient() would return 0
event though the cumulativeDistributiveFunctionAtAbscissa was not 0.
2019-08-21 14:45:27 +02:00
Léa Saviot
04b0df9a72
[apps/proba] Fix Student precision for computation
2019-08-21 12:18:32 +02:00
Léa Saviot
06b3b2c892
[apps/proba] Student parameter maximum is 200
2019-08-21 12:18:08 +02:00
Léa Saviot
bbf506ddb7
[apps/proba] Refactor law -> distribution
2019-08-20 17:16:19 +02:00
