Add MDFT (WIP)

[itsdfish]

No description provided.

[github-actions]

Benchmark Results

	master	ec79709...	master/ec797093c1d047...
logpdf/("SequentialSamplingModels.DDM", 10)	1.66 ± 0.11 μs	1.66 ± 0.11 μs	1
logpdf/("SequentialSamplingModels.DDM", 100)	17.3 ± 1.1 μs	17.2 ± 1.2 μs	1
logpdf/("SequentialSamplingModels.LBA", 10)	2.16 ± 0.14 μs	2.16 ± 0.15 μs	1
logpdf/("SequentialSamplingModels.LBA", 100)	20.4 ± 0.5 μs	20.4 ± 0.47 μs	1
logpdf/("SequentialSamplingModels.LNR", 10)	0.958 ± 0.15 μs	0.967 ± 0.14 μs	0.991
logpdf/("SequentialSamplingModels.LNR", 100)	8.2 ± 0.26 μs	8.23 ± 0.26 μs	0.997
logpdf/("SequentialSamplingModels.RDM", 10)	2.28 ± 0.13 μs	2.27 ± 0.13 μs	1
logpdf/("SequentialSamplingModels.RDM", 100)	21.9 ± 0.47 μs	21.8 ± 0.51 μs	1
logpdf/("SequentialSamplingModels.Wald", 10)	0.227 ± 0.16 μs	0.23 ± 0.17 μs	0.987
logpdf/("SequentialSamplingModels.Wald", 100)	2 ± 0.068 μs	2 ± 0.06 μs	1
logpdf/("SequentialSamplingModels.WaldMixture", 10)	0.924 ± 0.16 μs	0.926 ± 0.16 μs	0.998
logpdf/("SequentialSamplingModels.WaldMixture", 100)	8.94 ± 0.16 μs	8.94 ± 0.16 μs	1
rand/("SequentialSamplingModels.DDM", 10)	3.94 ± 0.57 μs	3.92 ± 0.54 μs	1.01
rand/("SequentialSamplingModels.DDM", 100)	0.0384 ± 0.002 ms	0.0384 ± 0.002 ms	0.998
rand/("SequentialSamplingModels.LBA", 10)	3.22 ± 0.89 μs	3.23 ± 1 μs	0.996
rand/("SequentialSamplingModels.LBA", 100)	16.3 ± 0.51 μs	16.4 ± 1 μs	0.992
rand/("SequentialSamplingModels.LCA", 10)	0.597 ± 0.26 ms	0.589 ± 0.26 ms	1.01
rand/("SequentialSamplingModels.LCA", 100)	6.62 ± 0.31 ms	6.52 ± 0.29 ms	1.02
rand/("SequentialSamplingModels.LNR", 10)	1.07 ± 0.21 μs	1.08 ± 0.2 μs	0.987
rand/("SequentialSamplingModels.LNR", 100)	7.51 ± 3.8 μs	7.55 ± 3.8 μs	0.995
rand/("SequentialSamplingModels.RDM", 10)	1.34 ± 0.18 μs	1.36 ± 0.19 μs	0.988
rand/("SequentialSamplingModels.RDM", 100)	11.3 ± 3.8 μs	11.1 ± 3.7 μs	1.01
rand/("SequentialSamplingModels.Wald", 10)	0.463 ± 0.16 μs	0.462 ± 0.17 μs	1
rand/("SequentialSamplingModels.Wald", 100)	2.89 ± 0.19 μs	2.89 ± 0.2 μs	1
rand/("SequentialSamplingModels.WaldMixture", 10)	1.18 ± 0.15 μs	1.18 ± 0.15 μs	0.995
rand/("SequentialSamplingModels.WaldMixture", 100)	11.3 ± 0.2 μs	11.3 ± 0.2 μs	1
simulate/SequentialSamplingModels.DDM	3.72 ± 1.6 μs	3.74 ± 1.4 μs	0.994
simulate/SequentialSamplingModels.LBA	2.25 ± 0.072 μs	2.19 ± 0.079 μs	1.03
simulate/SequentialSamplingModels.LCA	0.0833 ± 0.017 ms	0.0815 ± 0.016 ms	1.02
simulate/SequentialSamplingModels.RDM	0.0997 ± 0.032 ms	0.101 ± 0.033 ms	0.99
simulate/SequentialSamplingModels.Wald	9.01 ± 4.3 μs	9.07 ± 4.5 μs	0.994
simulate/SequentialSamplingModels.WaldMixture	4.02 ± 1.4 μs	4.02 ± 1.6 μs	0.999
simulate/mdft	0.185 ± 0.069 ms	0.187 ± 0.065 ms	0.991
time_to_load	3.7 ± 0.017 s	3.72 ± 0.031 s	0.996

Benchmark Plots

A plot of the benchmark results have been uploaded as an artifact to the workflow run for this PR.
Go to "Actions"->"Benchmark a pull request"->[the most recent run]->"Artifacts" (at the bottom).

[kiante-fernandez]

Was the equation you were thinking of implementing this one?

$$V(t+1, i) = \theta V(t, i) + (1 - \theta) \left( \sum_{j=1}^n W_{ij} \phi_{ij}(t) C_j(t) \right)$$

where

• $V(t, i)$: represents the preference state for response option ( i ) at time ( t ).

• $\theta$: leakage parameter

• $W_{ij}$: The weight matrix element indicating the influence of option ( j ) on option ( i ).

• $\phi_{ij}(t)$: valence function at time ( t ), indicating the relative attractiveness of comparing option ( i ) to option ( j )

• $C_j(t)$: perceptual input for option ( j ) at time ( t )

[itsdfish]

Hi @kiante-fernandez. Yes. I am implementing something similar to the equation above. It is from Equation 2 in Multialternative Decision Field Theory: A Dynamic Connectionist Model of Decision Making. What is committed has not been run or tested, and is almost certainly incorrect. I made some changes and wanted to see whether there were performance issues, and decided to run it remotely because my computer is sensitive to heat.

[codecov-commenter]

Codecov Report

Attention: Patch coverage is 84.00000% with 32 lines in your changes missing coverage. Please review.

Project coverage is 89.57%. Comparing base (2dadbf5) to head (afb0821).
Report is 1 commits behind head on master.

❗ Current head afb0821 differs from pull request most recent head d12a2d9

Please upload reports for the commit d12a2d9 to get more accurate results.

Files	Patch %	Lines
src/ClassicMDFT.jl	70.90%	16 Missing ⚠️
src/MDFT.jl	91.07%	10 Missing ⚠️
src/type_system.jl	54.54%	5 Missing ⚠️
src/LCA.jl	92.85%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master      #73      +/-   ##
==========================================
- Coverage   90.27%   89.57%   -0.70%     
==========================================
  Files          24       26       +2     
  Lines        1480     1650     +170     
==========================================
+ Hits         1336     1478     +142     
- Misses        144      172      +28     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.