-
Notifications
You must be signed in to change notification settings - Fork 2
/
README.m~
69 lines (59 loc) · 1.98 KB
/
README.m~
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
%% Scripts
% ~WORKFLOW SCRIPTS~
% GetModelOutput: Rework MEMLS runs into a
% GenerateObs: Choose random realizations of all parameters, and generate a
% set of observations along the flight lines
% EstimateT: Calculate MCMC estimate of temperature parameters (~4 minute
% run-time per location). Saves dat/Expi.mat, where i=0,1,2...
% PostProcess: Explores the MCMC output via figures, including the final
% estimated temperature profiles
%
% ~UTILITY SCRIPTS~
% GetSensorData
% ObsModel: Interpolates between model data as a way of exploring parameter
% space
<<<<<<< HEAD
=======
% ~ Add By Yuna ~
%ACF_k: calculate the autocorrelation of the Markov chain
%psrf: calculate the Gelman and Rubin Potential Scale Reduction Factor
>>>>>>> pr/3
%% Data
% dat/ includes the MEMLS database of model runs
%% Experiments
%dat/Exp0 ~ x=15, N=1E4
%dat/Exp1 ~ x=15, N=1E5
%dat/Exp2 ~ x=20, N=1E5
<<<<<<< HEAD
=======
>>>>>>> pr/3
%---------experiment of Yuna-----------------
%All experiments use newly generated TbMEMLS, which fixed the wrong
%temperature profile
%dat/Exp5 ~ x=15, N=1E4, Elapsed time is 192.280114 seconds.
<<<<<<< HEAD
%The result changes everytime. Maybe the MCMC does not reach to a steady
%distribution
=======
%The result varies hugely everytime. The MCMC does not seem to reach to a
%steady-state distribution.
%dat/Exp51 ~ x=15, N=1E4
%dat/Exp52 ~ x=15, N=1E4
>>>>>>> pr/3
%dat/Exp6 ~ x=15, N=1E5, Elapsed time is 1989.567482 seconds.
%dat/Exp7 ~ x=15, N=1E5
<<<<<<< HEAD
%Repeat the Exp6 to see if the result of Exp6 is reproduceable
=======
%Repeat the Exp6 to see if the result of Exp6 is reprodutive.
%% Modified on April 1st(No Kidding!!!)
% Add lines in EstimateT & PostProcess to racking likehood chain and dTs by Mike
% dat/Exp8 ~ x=15, N=2e4
%jmprho=9, about 35% acceptance rate for rho
% dat/Exp9 ~ x=15, N=2e4
% dat/Exp10 ~ x=15, N=5e4
% dat/Exp11 ~ x=15, N=5e4
%jmprho=9,jmpdT=0.3,jmpB=4
%rho 43%, dT 50%, B 35%
% dat/Exp12 ~ x=15, N=5e4, bad mixing
>>>>>>> pr/3