mlspm.models

class mlspm.models.PosNet

Alias of mlspm.graph.models.PosNet

class mlspm.models.GraphImgNet

Alias of mlspm.graph.models.GraphImgNet

class mlspm.models.GraphImgNetIce

Alias of mlspm.graph.models.GraphImgNetIce

class mlspm.models.ASDAFMNet

Alias of mlspm.image.models.ASDAFMNet

class mlspm.models.AttentionUNet

Alias of mlspm.image.models.AttentionUNet

class mlspm.models.EDAFMNet

Alias of mlspm.image.models.EDAFMNet

mlspm.models.download_weights(weights_name: str, target_path: PathLike | None = None) → PathLike

Download pretrained weights for models.

The following weights are available:

• 'graph-ice-cu111': PosNet trained on ice clusters on Cu(111). (https://doi.org/10.5281/zenodo.10054348)

• 'graph-ice-au111-monolayer': PosNet trained on monolayer ice clusters on Au(111). (https://doi.org/10.5281/zenodo.10054348)

• 'graph-ice-au111-bilayer': PosNet trained on bilayer ice clusters on Au(111). (https://doi.org/10.5281/zenodo.10054348)

• 'asdafm-light': ASDAFMNet trained on molecules containing the elements H, C, N, O, and F. (https://doi.org/10.5281/zenodo.10514470)

• 'asdafm-heavy': ASDAFMNet trained on molecules additionally containing Si, P, S, Cl, and Br. (https://doi.org/10.5281/zenodo.10514470)

• 'edafm-base': EDAFMNet used for all predictions in the main ED-AFM paper and used for comparison in the various tests in the supplementary information of the paper. (https://doi.org/10.5281/zenodo.10606273)

• 'edafm-single-channel': EDAFMNet trained on only a single CO-tip AFM input. (https://doi.org/10.5281/zenodo.10606273)

• 'edafm-CO-Cl': EDAFMNet trained on alternative tip combination of CO and Cl. (https://doi.org/10.5281/zenodo.10606273)

• 'edafm-Xe-Cl': EDAFMNet trained on alternative tip combination of Xe and Cl. (https://doi.org/10.5281/zenodo.10606273)

• 'edafm-constant-noise': EDAFMNet trained using constant noise amplitude instead of normally distributed amplitude. (https://doi.org/10.5281/zenodo.10606273)

• 'edafm-uniform-noise': EDAFMNet trained using uniform random noise amplitude instead of normally distributed amplitude. (https://doi.org/10.5281/zenodo.10606273)

• 'edafm-no-gradient': EDAFMNet trained without background-gradient augmentation. (https://doi.org/10.5281/zenodo.10606273)

• 'edafm-matched-tips': EDAFMNet trained on data with matched tip distance between CO and Xe, instead of independently randomized distances. (https://doi.org/10.5281/zenodo.10606273)

Parameters:

• weights_name – Name of weights to download.

• target_path – Path where the weights file will be saved. If specified, the parent directory for the file has to exists. If not specified, a location in a cache directory is chosen. If the target file already exists, the download is skipped

Returns:

Path where the weights were saved.