Revision a763b81ee81810f5fa544165d5053bde478b2dbd authored by Konrad Werys on 28 November 2019, 13:47:50 UTC, committed by Konrad Werys on 28 November 2019, 13:47:50 UTC
1 parent e3f0532
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 ...