core.units.mlip_unit

core.units.mlip_unit#

Copyright (c) Meta Platforms, Inc. and affiliates.

This source code is licensed under the MIT license found in the LICENSE file in the root directory of this source tree.

Submodules#

Functions#

load_predict_unit(...)

Load a MLIPPredictUnit from a checkpoint file.

Package Contents#

core.units.mlip_unit.load_predict_unit(path: str | pathlib.Path, inference_settings: fairchem.core.units.mlip_unit.api.inference.InferenceSettings | str = 'default', overrides: dict | None = None, device: str = 'cuda') → fairchem.core.units.mlip_unit.mlip_unit.MLIPPredictUnit#

Load a MLIPPredictUnit from a checkpoint file.

Parameters:

  • path – Path to the checkpoint file

  • inference_settings – Settings for inference. Can be “default” (general purpose) or “turbo” (optimized for speed but requires fixed atomic composition). Advanced use cases can use a custom InferenceSettings object.

  • overrides – Optional dictionary of settings to override default inference settings.

  • device – Optional torch device to load the model onto.

Returns:

A MLIPPredictUnit instance ready for inference

Contents