Estimators setup

Significance test for differences in mean square error (MSE) using Wilcoxon signed-rank test

Reservoir TE distribution

 

 

AIS

 

 

 

Zdrojaky:
LorenzTEsig.py
LorenzTEnormalizedSig.py
LorenzTEOGSig_n.py
LorenzTEONSig_n.py

Estimators setup

 

 

Significance test for differences in mean square error (MSE) using Wilcoxon signed-rank test

Reservoir TE distribution

AIS

 

 

Zdrojaky:
MGTESig.py
MGTEnormalizedSig.py
MGTEOGSig_n.py
MGTEONSig_n.py

Estimators setup

 

Reservoir TE distribution

 

Zdrojaky:
NARMATEsig.py
NARMATEnormalizedSig.py

Estimators setup

 

Reservoir TE distribution

Zdrojaky:
LorenzTEsig.py
LorenzTEnormalizedSig.py

Relative entropy

 

Limiting density of discrete points

 

Doplnennie 24 a 23 .

Initial network setup

Spectral radius

 

argmin NRMSE=0.2, min NRMSE=0.00001525

Win

 

argmin NRMSE=0.1,  min NRMSE=0.00155

Orthogonalisation

 

argmin NRMSE=0,  min NRMSE=0.001396

Orthonormalisation

 

argmin NRMSE=5,  min NRMSE=0.01466

Všetko spolu

OG: argmin NRMSE=(0,0.2),  min NRMSE=0.00001164

ON: argmin NRMSE=(10,0.125),  min NRMSE=0.00001281

Prezentácia: pdf

Prieskum entropie rezervoirových TE (18, 19, 20 a 28) pomocou Kozachenko-Leonenko entropy estimatora tj.

H(TE):

TE a predikcie X(t) zložky Lorenzov systému.

NRMSE

Mean TE

Mean TE reservoir-output

TE Input/Output

Zdrojaky:
LorenzTE.py
LorenzTEnormalized.py
LorenzTEnormalizedOG.py
LorenzTEnormalizedON.py

Optimálne nastavenia parametrov embedding vektora podľa Ragwitz-Kantz kritéria pre reservoir. Tj.

Initial network setup

History

Tau

Frequencies

 

Zdrojaky:

embedding.py
embeddingMC.py