EstimateZernikesAiDonutConfig#

class lsst.ts.wep.task.EstimateZernikesAiDonutConfig(*args, **kw)#

Bases: EstimateZernikesBaseConfig

AiDonut-specific configuration parameters for Zernike estimation.

Attributes Summary

`device`	Device to run the model on ('cpu' or 'cuda').
`instConfigFile`	' the path will be understood as relative to the ts_wep policy directory.
`modelPath`	Path to the AiDonut model file.
`nollIndices`	Noll indices for which you wish to estimate Zernike coefficients.
`returnWfDev`	If True, returns wavefront deviation.
`saveHistory`	Whether to save the algorithm history in the task metadata.
`startWithIntrinsic`	Whether to start Zernike estimation from the intrinsic Zernikes.
`temperature`	Temperature for softmax weighting of predictions based on model uncertainty.
`timeout`	Timeout in seconds for the multiprocessing pool.

Attributes Documentation

device: Field#: Device to run the model on (‘cpu’ or ‘cuda’). (str, default 'cpu')

instConfigFile: pexConfig.Field#

‘ the path will be understood as relative to the ts_wep policy directory. If not provided, the default instrument for the camera will be loaded. (str, default None)

Type:: Path to a instrument configuration file to override the instrument configuration. If begins with ‘policy

modelPath: Field#: Path to the AiDonut model file. (str, default '')

nollIndices: pexConfig.Field#: Noll indices for which you wish to estimate Zernike coefficients. Note these values must be unique, ascending, >= 4, and azimuthal pairs must be complete. For example, if nollIndices contains 5, it must also contain 6 (because 5 and 6 are the azimuthal pairs for astigmatism). (List, default (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))

returnWfDev: pexConfig.Field#: If True, returns wavefront deviation. If False, returns full OPD. (bool, default False)

saveHistory: pexConfig.Field#: Whether to save the algorithm history in the task metadata. Depending on the algorithm, saving the history might slow down estimation, but doing so will provide intermediate products from the estimation process. (bool, default False)

startWithIntrinsic: pexConfig.Field#: Whether to start Zernike estimation from the intrinsic Zernikes. (bool, default True)

temperature: Field#: Temperature for softmax weighting of predictions based on model uncertainty. Lower values put greater weight on lower-uncertainty predictions. (float, default 0.005)

timeout: pexConfig.Field#: Timeout in seconds for the multiprocessing pool. If the pool does not complete within this time, an empty results list is returned. (int, default 600)