... | ... | @@ -3,7 +3,7 @@ The optimization parameters file specifies additional options for the run. It is |
|
|
- **max-iterations**: the maximum number of iterations to run. Default is 100.
|
|
|
- **max-line-search-iterations**: number of line search iterations for the GradientAscent estimator.
|
|
|
- **convergence-tolerance**: a tolerance threshold for premature stopping of the estimation. If the relative decrease of the loss function during an iteration of the estimation of the model is below this threshold, the computations stops.
|
|
|
- **gpu_mode**: can take the value 'auto', 'none', 'full' or 'kernel'. 'full' will run all computations on a gpu whereas 'kernel' will only run the kernel convolutions on gpu.
|
|
|
- **gpu-mode**: can take the value 'auto', 'none', 'full' or 'kernel'. 'full' will run all computations on a gpu whereas 'kernel' will only run the kernel convolutions on gpu.
|
|
|
- **number-of-processes**: the number of processes to use. Multiprocessing is only supported for the deterministic atlas and longitudinal model.
|
|
|
- **downsampling-factor**: a factor by which to downsample internally the images, to speed-up image deformations.
|
|
|
- **save-every-n-iters**: when to periodically save the output. Default is 100. Beware, for large meshes, saving the whole output of a model can be expensive.
|
... | ... | |