mammos_ai#

Pre-trained AI models.

Functions

`Hc_Mr_BHmax_from_Ms_A_K`(Ms, A, K1[, model])	Predict Hc, Mr and BHmax from micromagnetic properties Ms, A and K1.
`Hc_Mr_BHmax_from_Ms_A_K_metadata`([model])	Get metadata for the specified Hc, Mr, BHmax prediction model.
`is_hard_magnet_from_Ms_A_K`(Ms, A, K1[, model])	Classify material as soft or hard magnetic from micromagnetic parameters.
`is_hard_magnet_from_Ms_A_K_metadata`([model])	Get metadata for the specified classification model.

mammos_ai.Hc_Mr_BHmax_from_Ms_A_K(Ms, A, K1, model='cube50_singlegrain_random_forest_v0.1')[source]#

Predict Hc, Mr and BHmax from micromagnetic properties Ms, A and K1.

This function predicts extrinsic properties coercive field Hc, remanent magnetization Mr and maximum energy product BHmax given a set of micromagnetic material parameters.

The following models are available for the prediction:

cube50_singlegrain_random_forest_v0.1: Random forest model trained on simulated data for single grain cubic particles with 50 nm edge length with the external field applied parallel to the anisotropy axis. These are both aligned along an edge of the cube. Further details on the training data and model can be found in the model repository.

Parameters:

Ms (Entity | Quantity | ArrayLike) – Spontaneous magnetization.
A (Entity | Quantity | ArrayLike) – Exchange stiffness constant.
K1 (Entity | Quantity | ArrayLike) – Uniaxial anisotropy constant.
model (str) – AI model used for the prediction

Returns:

An object containing extrinsic properties Hc, Mr, BHmax

Return type:

ExtrinsicProperties

Examples: >>> import mammos_ai >>> import mammos_entity as me >>> mammos_ai.Hc_Mr_BHmax_from_Ms_A_K(me.Ms(1e6), me.A(1e-12), me.Ku(1e6)) ExtrinsicProperties(Hc=…, Mr=…, BHmax=…)

mammos_ai.Hc_Mr_BHmax_from_Ms_A_K_metadata(model='cube50_singlegrain_random_forest_v0.1')[source]#

Get metadata for the specified Hc, Mr, BHmax prediction model.

Parameters:: model (str) – AI model used for the prediction
Return type:: dict

mammos_ai.is_hard_magnet_from_Ms_A_K(Ms, A, K1, model='cube50_singlegrain_random_forest_v0.1')[source]#

Classify material as soft or hard magnetic from micromagnetic parameters.

This function classifies a magnetic material as either soft or hard magnetic based on its micromagnetic parameters spontaneous magnetization Ms, exchange stiffness constant A and uniaxial anisotropy constant K1. The shape of the input parameters needs to be the same. If single values are provided, a single classification is returned. If arrays are provided, a numpy array with the same shape is returned.

The following models are available for the prediction:

cube50_singlegrain_random_forest_v0.1: Random forest model trained on simulated data for single grain cubic particles with 50 nm edge length with the external field applied parallel to the anisotropy axis. These are both aligned along an edge of the cube. Further details on the training data and model can be found in the model repository.

Parameters:

Ms (Entity | Quantity | ArrayLike) – Spontaneous magnetization.
A (Entity | Quantity | ArrayLike) – Exchange stiffness constant.
K1 (Entity | Quantity | ArrayLike) – Uniaxial anisotropy constant.
model (str) – AI model used for the classification

Returns:

Classification as False (soft) or True (hard). Returns a boolean for scalar inputs, or a numpy array with the same shape as the input for array inputs.

Return type:

bool | np.ndarray

Examples: >>> import mammos_ai >>> import mammos_entity as me >>> mammos_ai.is_hard_magnet_from_Ms_A_K(me.Ms(1e6), me.A(1e-12), me.Ku(1e6)) True

mammos_ai.is_hard_magnet_from_Ms_A_K_metadata(model='cube50_singlegrain_random_forest_v0.1')[source]#

Get metadata for the specified classification model.

Parameters:: model (str) – AI model used for the classification
Return type:: dict