deepmd.utils.finetune

Attributes

log

Classes

FinetuneRuleItem

Functions

get_index_between_two_maps(old_map, new_map)	Returns the mapping index of types in new_map to those in the old_map.
map_atom_exclude_types(atom_exclude_types, remap_index)	Return the remapped atom_exclude_types according to remap_index.
map_pair_exclude_types(pair_exclude_types, remap_index)	Return the remapped atom_exclude_types according to remap_index.

Module Contents

deepmd.utils.finetune.log

class deepmd.utils.finetune.FinetuneRuleItem(p_type_map: list[str], type_map: list[str], model_branch: str = 'Default', random_fitting: bool = False, resuming: bool = False)

p_type_map

type_map

model_branch = 'Default'

random_fitting = False

resuming = False

update_type

get_index_mapping()

Returns the mapping index of newly defined types to those in the pretrained model.

get_has_new_type()

Returns whether there are unseen types in the new type_map.

get_model_branch()

Returns the chosen model branch.

get_random_fitting()

Returns whether to use random fitting.

get_resuming()

Returns whether to only do resuming.

get_update_type()

Returns whether to update the type related params when loading from pretrained model with redundant types.

get_pretrained_tmap()

Returns the type map in the pretrained model.

get_finetune_tmap()

Returns the type map in the fine-tuned model.

deepmd.utils.finetune.get_index_between_two_maps(old_map: list[str], new_map: list[str])

Returns the mapping index of types in new_map to those in the old_map.

Parameters:

old_maplist[str]

The old list of atom type names.

new_maplist[str]

The new list of atom type names.

Returns:

index_map: list[int]

List contains len(new_map) indices, where index_map[i] is the index of new_map[i] in old_map. If new_map[i] is not in the old_map, the index will be i - len(new_map).

has_new_type: bool

Whether there are unseen types in the new type_map. If True, some type related params in the model, such as statistics, need to be extended to have a length of len(old_map) + len(new_map) in the type related dimension. Then positive indices from 0 to len(old_map) - 1 will select old params of types in old_map, while negative indices from -len(new_map) to -1 will select new params of types in new_map.

deepmd.utils.finetune.map_atom_exclude_types(atom_exclude_types: list[int], remap_index: list[int])

Return the remapped atom_exclude_types according to remap_index.

Parameters:

atom_exclude_typeslist[int]

Exclude the atomic contribution of the given types.

remap_indexlist[int]

The indices in the old type list that correspond to the types in the new type list.

Returns:

remapped_atom_exclude_types: list[int]

Remapped atom_exclude_types that only keeps the types in the new type list.

deepmd.utils.finetune.map_pair_exclude_types(pair_exclude_types: list[tuple[int, int]], remap_index: list[int])

Return the remapped atom_exclude_types according to remap_index.

Parameters:

pair_exclude_typeslist[tuple[int, int]]

Exclude the pair of atoms of the given types from computing the output of the atomic model.

remap_indexlist[int]

The indices in the old type list that correspond to the types in the new type list.

Returns:

remapped_pair_exclude_typess: list[tuple[int, int]]

Remapped pair_exclude_types that only keeps the types in the new type list.