route_distances package

Subpackages

• route_distances.lstm package
  • Submodules
  • route_distances.lstm.data module
    • InMemoryTreeDataset
    • TreeDataModule
      • TreeDataModule.setup()
      • TreeDataModule.train_dataloader()
      • TreeDataModule.val_dataloader()
      • TreeDataModule.test_dataloader()
  • route_distances.lstm.defaults module
  • route_distances.lstm.features module
    • add_fingerprints()
    • remove_reactions()
    • preprocess_reaction_tree()
  • route_distances.lstm.inference module
    • distances_calculator()
  • route_distances.lstm.models module
    • RouteDistanceModel
      • RouteDistanceModel.forward()
      • RouteDistanceModel.training_step()
      • RouteDistanceModel.validation_step()
      • RouteDistanceModel.validation_epoch_end()
      • RouteDistanceModel.test_step()
      • RouteDistanceModel.test_epoch_end()
      • RouteDistanceModel.configure_optimizers()
  • route_distances.lstm.optim module
    • OptunaObjective
  • route_distances.lstm.utils module
    • accumulate_stats()
    • add_node_index()
    • collate_batch()
    • collate_trees()
    • gather_adjacency_list()
    • gather_node_attributes()
  • Module contents
• route_distances.ted package
  • Submodules
  • route_distances.ted.distances module
    • distance_matrix()
  • route_distances.ted.reactiontree module
    • ReactionTreeWrapper
      • ReactionTreeWrapper.info
      • ReactionTreeWrapper.first_tree
      • ReactionTreeWrapper.trees
      • ReactionTreeWrapper.distance_iter()
      • ReactionTreeWrapper.distance_to()
      • ReactionTreeWrapper.distance_to_with_sorting()
  • route_distances.ted.utils module
    • TreeContent
      • TreeContent.MOLECULES
      • TreeContent.REACTIONS
      • TreeContent.BOTH
    • AptedConfig
      • AptedConfig.rename()
      • AptedConfig.children()
    • StandardFingerprintFactory
  • Module contents
• route_distances.tools package
  • Submodules
  • route_distances.tools.cluster_aizynth_output module
    • main()
  • route_distances.tools.prepare_aizynthfinder_output module
    • main()
  • route_distances.tools.train_lstm_model module
    • main()
  • Module contents

Submodules

route_distances.clustering module

Module containing a class to help out with clustering

class route_distances.clustering.ClusteringHelper(distances)

Bases: object

A helper class to perform clustering of items based on a pre-computed distance matrix.

The clustering is the agglomerative clustering algorithm implemented in Scikit-learn.

Parameters:

distances (np.ndarray)

property labels: ndarray

Return the cluster labels

fixed_clustering(n_clusters, **kwargs)

Make a fixed number of clusters.

Additional arguments to the clustering algorithm can be passed in as key-word arguments

Parameters:

• n_clusters (int) – the desired number of clusters

• kwargs (Any)

Returns:

the cluster index for each observation

Return type:

ndarray

linkage_matrix(**kwargs)

Compute the linkage matrix.

Additional arguments to the clustering algorithm can be passed in as key-word arguments

Returns:

the linkage matrix

Parameters:

kwargs (Any)

Return type:

ndarray

optimize(max_clusters=5, **kwargs)

Optimize the number of cluster based Silhouette metric.

Additional arguments to the clustering algorithm can be passed in as key-word arguments

Parameters:

• max_clusters (int) – the maximum number of clusters to consider

• kwargs (Any)

Returns:

the cluster index for each observation

Return type:

ndarray

static cluster(distances, n_clusters, **kwargs)

Cluster items based on a pre-computed distance matrix using a hierarchical clustering.

Parameters:

• distances (ndarray) – the distance matrix

• n_clusters (int) – the desired number of clusters

• kwargs (Any)

Returns:

the cluster index for each observation

Return type:

ndarray

route_distances.route_distances module

Module containing a factory function for making predictions of route distances

route_distances.route_distances.route_distances_calculator(model, **kwargs)

Return a callable that given a list routes as dictionaries calculate the squared distance matrix

Parameters:

• model (str)

• kwargs (Any)

Returns:

Return type:

Callable[[List[Dict[str, Any]]], ndarray]

route_distances.validation module

Module containing routes to validate AiZynthFinder-like input dictionaries

class route_distances.validation.ReactionNode(**data)

Bases: BaseModel

Node representing a reaction

Parameters:

data (Any)

type: Annotated[str, StringConstraints(pattern='^reaction$')]

children: List[MoleculeNode]

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class route_distances.validation.MoleculeNode(*, smiles, type, children=None)

Bases: BaseModel

Node representing a molecule

Parameters:

• smiles (str)

• type (Annotated[str, StringConstraints(strip_whitespace=None, to_upper=None, to_lower=None, strict=None, min_length=None, max_length=None, pattern=^mol$)])

• children (Annotated[List[ReactionNode], Len(min_length=1, max_length=1)] | None)

smiles: str

type: Annotated[str, StringConstraints(pattern='^mol$')]

children: conlist(ReactionNode, min_length=1, max_length=1) | None

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

route_distances.validation.validate_dict(dict_)

Check that the route dictionary is a valid structure

Parameters:

• dict – the route as dictionary

• dict_ (Dict[str, Any])

Return type:

None

Module contents