Revision 41e684800fbdd0bf24b91ff51f61f60818cc9bc0 authored by Konrad Werys on 14 November 2019, 09:31:46 UTC, committed by Konrad Werys on 14 November 2019, 09:31:46 UTC
1 parent 8e25865
issues.md
# NOTES
## Different results from the configurations
parameter_to_map: T1_SHMOLLI
functions_type: **ModelT1ThreeParam/ModelT1Shmolli**
fitting_method: LevMarVnl
max_function_evals: 4000
fTolerance: 1e-12
sign_calc_method: MagPhase
start_point_calc_method: Default
ModelT1ThreeParam: 10s and seemingly correct results
ModelT1Shmolli: 2.2s and wrong results
parameter_to_map: T1_SHMOLLI
functions_type: ModelT1Shmolli
fitting_method: LevMarVnl
max_function_evals: 4000
fTolerance: 1e-12
sign_calc_method: MagPhase
start_point_calc_method: **Default/StartPointSHMOLLI**
Default: 2.2s and wrong results
StartPointSHMOLLI: 10s and seemingly correct results
## 5 vs 7 samples
5 samples are calculated way slower. Maybe playing with fmax and maxIter would reduce the time?
Computing file changes ...