chemicalchecker.tool.targetmate.base.StackedModel

class StackedModel(n_components=None, **kwargs)[source]

Bases: SignaturedModel

Stacked TargetMate model.

Initialize the stacked model.

Methods

`array_on_disk`
`check_array_from_disk`
`compress_models`	Store model in compressed format for persistance
`cpu_count`
`create_models_path`
`directory_tree`
`explain`
`explain_stack`
`featurizer`
`find_base_mod`	Select a pipeline, for example, using HyperOpt.
`fit`	Fit the stacked model.
`fit_stack`
`func_hpc`	Execute the any method on the configured HPC.
`get_Xy_from_data`	Given data and certain idxs, get X and y (if available)
`get_data_fit`
`get_data_predict`
`get_datasets`
`get_destination_dir`
`get_master_idxs`
`load`	Load previously stored TargetMate instance.
`load_base_mod`	Load base model
`load_base_model`	Load a base model
`load_data`
`load_explanations`
`load_predictions`
`master_key_type`	Check master key types
`master_mapping`
`metric_calc`	Calculate metric.
`model_iterator`
`pca_fit`
`pca_transform`
`predict`
`predict_stack`
`prepare_data`
`prepare_for_ml`
`read_data`
`read_signatures`	Read signatures
`read_signatures_ensemble`	Return signatures as an ensemble
`read_signatures_prestacked`	Return signatures in a stacked form from an already prestacked file
`read_signatures_stacked`	Return signatures in a stacked form
`repath_bases_by_fold`	Redefine path of a TargetMate instance.
`repath_predictions_by_fold`	Redefine path of a TargetMate instance.
`repath_predictions_by_fold_and_set`
`repath_predictions_by_set`	Redefine path of a TargetMate instance.
`reset_path_bases`
`reset_path_predictions`	Reset predictions path
`sampler`
`save`	Save TargetMate instance
`save_data`
`signaturize`
`waiter`	Wait for jobs to finish
`wipe`	Delete temporary data

compress_models(): Store model in compressed format for persistance

find_base_mod(X, y, smiles, destination_dir): Select a pipeline, for example, using HyperOpt.

fit(data, idxs=None, is_tmp=False, wait=True, scramble=False)[source]

Fit the stacked model.

Parameters:

data (InputData) – Input data.
idxs (array) – Indices to use for the signatures. If none specified, all are used (default=None).
is_tmp (bool) – Save in the temporary directory or in the models directory.

func_hpc(func_name, *args, **kwargs)

Execute the any method on the configured HPC.

Parameters:

args (tuple) – the arguments for of the function method
kwargs (dict) – arguments for the HPC method.

get_Xy_from_data(data, idxs, scramble=False): Given data and certain idxs, get X and y (if available)

static load(models_path): Load previously stored TargetMate instance.

load_base_mod(): Load base model

load_base_model(destination_dir, append_pipe=False): Load a base model

master_key_type(): Check master key types

metric_calc(y_true, y_pred, metric=None): Calculate metric. Returns (value, weight) tuple.

read_signatures(datasets=None, idxs=None, smiles=None, inchikeys=None, is_tmp=None, sign_folder=None): Read signatures

read_signatures_ensemble(datasets, smiles, inchikeys, idxs, is_tmp, sign_folder): Return signatures as an ensemble

read_signatures_prestacked(mask, datasets, smiles, inchikeys, idxs, is_tmp, sign_folder): Return signatures in a stacked form from an already prestacked file

read_signatures_stacked(datasets, smiles, inchikeys, idxs, is_tmp, sign_folder): Return signatures in a stacked form

repath_bases_by_fold(fold_number, is_tmp=True, reset=True, only_train=False): Redefine path of a TargetMate instance. Used by the Validation class.

repath_predictions_by_fold(fold_number, is_tmp=True, reset=True): Redefine path of a TargetMate instance. Used by the Validation class.

repath_predictions_by_set(is_train, is_tmp=True, reset=True): Redefine path of a TargetMate instance. Used by the Validation class.

reset_path_predictions(is_tmp=True): Reset predictions path

save(): Save TargetMate instance

waiter(jobs, secs=3): Wait for jobs to finish

wipe(): Delete temporary data