Loading deepchem/feat/material_featurizers/element_property_fingerprint.py +8 −6 Original line number Diff line number Diff line Loading @@ -50,13 +50,8 @@ class ElementPropertyFingerprint(MaterialCompositionFeaturizer): data_source: str of "matminer", "magpie" or "deml" (default "matminer") Source for element property data. """ try: from matminer.featurizers.composition import ElementProperty except ModuleNotFoundError: raise ImportError("This class requires matminer to be installed.") self.data_source = data_source self.ep_featurizer = ElementProperty.from_preset(self.data_source) self.ep_featurizer = None def _featurize(self, composition: PymatgenComposition) -> np.ndarray: """ Loading @@ -73,6 +68,13 @@ class ElementPropertyFingerprint(MaterialCompositionFeaturizer): Vector of properties and statistics derived from chemical stoichiometry. Some values may be NaN. """ if self.ep_featurizer is None: try: from matminer.featurizers.composition import ElementProperty self.ep_featurizer = ElementProperty.from_preset(self.data_source) except ModuleNotFoundError: raise ImportError("This class requires matminer to be installed.") try: feats = self.ep_featurizer.featurize(composition) except: Loading deepchem/feat/material_featurizers/sine_coulomb_matrix.py +8 −6 Original line number Diff line number Diff line Loading @@ -54,14 +54,9 @@ class SineCoulombMatrix(MaterialStructureFeaturizer): flatten: bool (default True) Return flattened vector of matrix eigenvalues. """ try: from matminer.featurizers.structure import SineCoulombMatrix as SCM except ModuleNotFoundError: raise ImportError("This class requires matminer to be installed.") self.max_atoms = max_atoms self.flatten = flatten self.scm = SCM(flatten=False) self.scm = None def _featurize(self, struct: PymatgenStructure) -> np.ndarray: """ Loading @@ -79,6 +74,13 @@ class SineCoulombMatrix(MaterialStructureFeaturizer): 2D sine Coulomb matrix with shape (max_atoms, max_atoms), or 1D matrix eigenvalues with shape (max_atoms,). """ if self.scm is None: try: from matminer.featurizers.structure import SineCoulombMatrix as SCM self.scm = SCM(flatten=False) except ModuleNotFoundError: raise ImportError("This class requires matminer to be installed.") # Get full N x N SCM sine_mat = self.scm.featurize(struct) Loading deepchem/molnet/load_function/material_datasets/load_bandgap.py +49 −146 Original line number Diff line number Diff line Loading @@ -2,60 +2,43 @@ Experimental bandgaps for inorganic crystals. """ import os import logging import deepchem as dc from deepchem.molnet.load_function.molnet_loader import TransformerGenerator, _MolnetLoader from deepchem.data import Dataset from typing import List, Optional, Tuple, Union import deepchem from deepchem.feat import MaterialCompositionFeaturizer from deepchem.splits.splitters import Splitter from deepchem.molnet.defaults import get_defaults from typing import List, Tuple, Dict, Optional, Any logger = logging.getLogger(__name__) DEFAULT_DIR = deepchem.utils.data_utils.get_data_dir() BANDGAP_URL = 'https://deepchemdata.s3-us-west-1.amazonaws.com/datasets/expt_gap.tar.gz' BANDGAP_TASKS = ['experimental_bandgap'] # dict of accepted featurizers for this dataset # modify the returned dicts for your dataset DEFAULT_FEATURIZERS = get_defaults("feat") # Names of supported featurizers featurizers = [ 'ElementPropertyFingerprint', ] DEFAULT_FEATURIZERS = {k: DEFAULT_FEATURIZERS[k] for k in featurizers} # dict of accepted transformers DEFAULT_TRANSFORMERS = get_defaults("trans") class _BandgapLoader(_MolnetLoader): # dict of accepted splitters DEFAULT_SPLITTERS = get_defaults("splits") # names of supported splitters splitters = ['RandomSplitter'] DEFAULT_SPLITTERS = {k: DEFAULT_SPLITTERS[k] for k in splitters} def create_dataset(self) -> Dataset: dataset_file = os.path.join(self.data_dir, 'expt_gap.json') targz_file = os.path.join(self.data_dir, 'expt_gap.tar.gz') if not os.path.exists(dataset_file): if not os.path.exists(targz_file): dc.utils.data_utils.download_url( url=BANDGAP_URL, dest_dir=self.data_dir) dc.utils.data_utils.untargz_file(targz_file, self.data_dir) loader = dc.data.JsonLoader( tasks=self.tasks, feature_field="composition", label_field="experimental_bandgap", featurizer=self.featurizer) return loader.create_dataset(dataset_file) def load_bandgap( featurizer=DEFAULT_FEATURIZERS['ElementPropertyFingerprint'], transformers: List = [DEFAULT_TRANSFORMERS['NormalizationTransformer']], splitter=DEFAULT_SPLITTERS['RandomSplitter'], featurizer: Union[dc.feat.Featurizer, str] = dc.feat.ElementPropertyFingerprint(), splitter: Union[dc.splits.Splitter, str, None] = 'random', transformers: List[Union[TransformerGenerator, str]] = ['normalization'], reload: bool = True, data_dir: Optional[str] = None, save_dir: Optional[str] = None, featurizer_kwargs: Dict[str, Any] = {}, splitter_kwargs: Dict[str, Any] = { 'frac_train': 0.8, 'frac_valid': 0.1, 'frac_test': 0.1 }, transformer_kwargs: Dict[str, Dict[str, Any]] = { 'NormalizationTransformer': { 'transform_X': True } }, **kwargs) -> Tuple[List, Optional[Tuple], List]: **kwargs ) -> Tuple[List[str], Tuple[Dataset, ...], List[dc.trans.Transformer]]: """Load band gap dataset. Contains 4604 experimentally measured band gaps for inorganic Loading @@ -68,27 +51,25 @@ def load_bandgap( Parameters ---------- featurizer : MaterialCompositionFeaturizer (default ElementPropertyFingerprint) A featurizer that inherits from deepchem.feat.Featurizer. transformers : List[Transformer] A transformer that inherits from deepchem.trans.Transformer. splitter : Splitter (default RandomSplitter) A splitter that inherits from deepchem.splits.splitters.Splitter. reload : bool (default True) Try to reload dataset from disk if already downloaded. Save to disk after featurizing. data_dir : str, optional (default None) Path to datasets. save_dir : str, optional (default None) Path to featurized datasets. featurizer_kwargs : Dict[str, Any] Specify parameters to featurizer, e.g. {"size": 1024} splitter_kwargs : Dict[str, Any] Specify parameters to splitter, e.g. {"seed": 42} transformer_kwargs : dict Maps transformer names to constructor arguments, e.g. {"BalancingTransformer": {"transform_x":True, "transform_y":False}} **kwargs : additional optional arguments. featurizer: Featurizer or str the featurizer to use for processing the data. Alternatively you can pass one of the names from dc.molnet.featurizers as a shortcut. splitter: Splitter or str the splitter to use for splitting the data into training, validation, and test sets. Alternatively you can pass one of the names from dc.molnet.splitters as a shortcut. If this is None, all the data will be included in a single dataset. transformers: list of TransformerGenerators or strings the Transformers to apply to the data. Each one is specified by a TransformerGenerator or, as a shortcut, one of the names from dc.molnet.transformers. reload: bool if True, the first call for a particular featurizer and splitter will cache the datasets to disk, and subsequent calls will reload the cached datasets. data_dir: str a directory to save the raw data in save_dir: str a directory to save the dataset in Returns ------- Loading @@ -112,91 +93,13 @@ def load_bandgap( Examples -------- >> import deepchem as dc >> tasks, datasets, transformers = dc.molnet.load_bandgap(reload=False) >> tasks, datasets, transformers = dc.molnet.load_bandgap() >> train_dataset, val_dataset, test_dataset = datasets >> n_tasks = len(tasks) >> n_features = train_dataset.get_data_shape()[0] >> model = dc.models.MultitaskRegressor(n_tasks, n_features) """ # Featurize logger.info("About to featurize band gap dataset.") my_tasks = ['experimental_bandgap'] # machine learning targets # Get DeepChem data directory if needed if data_dir is None: data_dir = DEFAULT_DIR if save_dir is None: save_dir = DEFAULT_DIR if issubclass(featurizer, MaterialCompositionFeaturizer): featurizer = featurizer(**featurizer_kwargs) else: raise TypeError( "featurizer must be a subclass of MaterialCompositionFeaturizer.") if issubclass(splitter, Splitter): splitter = splitter() else: raise TypeError("splitter must be a subclass of Splitter.") # Reload from disk if reload: featurizer_name = str(featurizer.__class__.__name__) splitter_name = str(splitter.__class__.__name__) save_folder = os.path.join(save_dir, "bandgap-featurized", featurizer_name, splitter_name) loaded, all_dataset, transformers = deepchem.utils.data_utils.load_dataset_from_disk( save_folder) if loaded: return my_tasks, all_dataset, transformers # First type of supported featurizers supported_featurizers: List[str] = ['ElementPropertyFingerprint'] # Load .tar.gz file if featurizer.__class__.__name__ in supported_featurizers: dataset_file = os.path.join(data_dir, 'expt_gap.json') if not os.path.exists(dataset_file): targz_file = os.path.join(data_dir, 'expt_gap.tar.gz') if not os.path.exists(targz_file): deepchem.utils.data_utils.download_url( url=BANDGAP_URL, dest_dir=data_dir) deepchem.utils.data_utils.untargz_file( os.path.join(data_dir, 'expt_gap.tar.gz'), data_dir) # Changer loader to match featurizer and data file type loader = deepchem.data.JsonLoader( tasks=my_tasks, feature_field="composition", label_field="experimental_bandgap", featurizer=featurizer) # Featurize dataset dataset = loader.create_dataset(dataset_file) train_dataset, valid_dataset, test_dataset = splitter.train_valid_test_split( dataset, **splitter_kwargs) # Initialize transformers transformers = [ DEFAULT_TRANSFORMERS[t](dataset=dataset, **transformer_kwargs[t]) if isinstance(t, str) else t( dataset=dataset, **transformer_kwargs[str(t.__name__)]) for t in transformers ] for transformer in transformers: train_dataset = transformer.transform(train_dataset) valid_dataset = transformer.transform(valid_dataset) test_dataset = transformer.transform(test_dataset) if reload: # save to disk deepchem.utils.data_utils.save_dataset_to_disk( save_folder, train_dataset, valid_dataset, test_dataset, transformers) return my_tasks, (train_dataset, valid_dataset, test_dataset), transformers loader = _BandgapLoader(featurizer, splitter, transformers, BANDGAP_TASKS, data_dir, save_dir, **kwargs) return loader.load_dataset('bandgap', reload) deepchem/molnet/load_function/material_datasets/load_mp_formation_energy.py +48 −148 Original line number Diff line number Diff line Loading @@ -2,60 +2,42 @@ Calculated formation energies for inorganic crystals from Materials Project. """ import os import logging import deepchem from deepchem.feat import MaterialStructureFeaturizer from deepchem.splits.splitters import Splitter from deepchem.molnet.defaults import get_defaults import deepchem as dc from deepchem.molnet.load_function.molnet_loader import TransformerGenerator, _MolnetLoader from deepchem.data import Dataset from typing import List, Optional, Tuple, Union from typing import List, Tuple, Dict, Optional, Any logger = logging.getLogger(__name__) DEFAULT_DIR = deepchem.utils.data_utils.get_data_dir() MPFORME_URL = 'https://deepchemdata.s3-us-west-1.amazonaws.com/datasets/mp_formation_energy.tar.gz' MPFORME_TASKS = ['formation_energy'] # dict of accepted featurizers for this dataset # modify the returned dicts for your dataset DEFAULT_FEATURIZERS = get_defaults("feat") # Names of supported featurizers featurizers = [ 'CGCNNFeaturizer', 'SineCoulombMatrix', ] DEFAULT_FEATURIZERS = {k: DEFAULT_FEATURIZERS[k] for k in featurizers} # dict of accepted transformers DEFAULT_TRANSFORMERS = get_defaults("trans") # dict of accepted splitters DEFAULT_SPLITTERS = get_defaults("splits") class _MPFormationLoader(_MolnetLoader): # names of supported splitters splitters = ['RandomSplitter'] DEFAULT_SPLITTERS = {k: DEFAULT_SPLITTERS[k] for k in splitters} def create_dataset(self) -> Dataset: dataset_file = os.path.join(self.data_dir, 'mp_formation_energy.json') targz_file = os.path.join(self.data_dir, 'mp_formation_energy.tar.gz') if not os.path.exists(dataset_file): if not os.path.exists(targz_file): dc.utils.data_utils.download_url( url=MPFORME_URL, dest_dir=self.data_dir) dc.utils.data_utils.untargz_file(targz_file, self.data_dir) loader = dc.data.JsonLoader( tasks=self.tasks, feature_field="structure", label_field="formation_energy", featurizer=self.featurizer) return loader.create_dataset(dataset_file) def load_mp_formation_energy( featurizer=DEFAULT_FEATURIZERS['SineCoulombMatrix'], transformers: List = [DEFAULT_TRANSFORMERS['NormalizationTransformer']], splitter=DEFAULT_SPLITTERS['RandomSplitter'], featurizer: Union[dc.feat.Featurizer, str] = dc.feat.SineCoulombMatrix(), splitter: Union[dc.splits.Splitter, str, None] = 'random', transformers: List[Union[TransformerGenerator, str]] = ['normalization'], reload: bool = True, data_dir: Optional[str] = None, save_dir: Optional[str] = None, featurizer_kwargs: Dict[str, Any] = {}, splitter_kwargs: Dict[str, Any] = { 'frac_train': 0.8, 'frac_valid': 0.1, 'frac_test': 0.1 }, transformer_kwargs: Dict[str, Dict[str, Any]] = { 'NormalizationTransformer': { 'transform_X': True } }, **kwargs) -> Tuple[List, Optional[Tuple], List]: **kwargs ) -> Tuple[List[str], Tuple[Dataset, ...], List[dc.trans.Transformer]]: """Load mp formation energy dataset. Contains 132752 calculated formation energies and inorganic Loading @@ -69,27 +51,25 @@ def load_mp_formation_energy( Parameters ---------- featurizer : MaterialStructureFeaturizer (default SineCoulombMatrix) A featurizer that inherits from deepchem.feat.Featurizer. transformers : List[Transformer] A transformer that inherits from deepchem.trans.Transformer. splitter : Splitter (default RandomSplitter) A splitter that inherits from deepchem.splits.splitters.Splitter. reload : bool (default True) Try to reload dataset from disk if already downloaded. Save to disk after featurizing. data_dir : str, optional (default None) Path to datasets. save_dir : str, optional (default None) Path to featurized datasets. featurizer_kwargs : Dict[str, Any] Specify parameters to featurizer, e.g. {"size": 1024} splitter_kwargs : Dict[str, Any] Specify parameters to splitter, e.g. {"seed": 42} transformer_kwargs : dict Maps transformer names to constructor arguments, e.g. {"BalancingTransformer": {"transform_X":True, "transform_y":False}} **kwargs : additional optional arguments. featurizer: Featurizer or str the featurizer to use for processing the data. Alternatively you can pass one of the names from dc.molnet.featurizers as a shortcut. splitter: Splitter or str the splitter to use for splitting the data into training, validation, and test sets. Alternatively you can pass one of the names from dc.molnet.splitters as a shortcut. If this is None, all the data will be included in a single dataset. transformers: list of TransformerGenerators or strings the Transformers to apply to the data. Each one is specified by a TransformerGenerator or, as a shortcut, one of the names from dc.molnet.transformers. reload: bool if True, the first call for a particular featurizer and splitter will cache the datasets to disk, and subsequent calls will reload the cached datasets. data_dir: str a directory to save the raw data in save_dir: str a directory to save the dataset in Returns ------- Loading @@ -114,93 +94,13 @@ def load_mp_formation_energy( Examples -------- >> import deepchem as dc >> tasks, datasets, transformers = dc.molnet.load_mp_formation_energy(reload=False) >> tasks, datasets, transformers = dc.molnet.load_mp_formation_energy() >> train_dataset, val_dataset, test_dataset = datasets >> n_tasks = len(tasks) >> n_features = train_dataset.get_data_shape()[0] >> model = dc.models.MultitaskRegressor(n_tasks, n_features) """ # Featurize logger.info("About to featurize mp formation energy dataset.") my_tasks = ['formation_energy'] # machine learning targets # Get DeepChem data directory if needed if data_dir is None: data_dir = DEFAULT_DIR if save_dir is None: save_dir = DEFAULT_DIR if issubclass(featurizer, MaterialStructureFeaturizer): featurizer = featurizer(**featurizer_kwargs) else: raise TypeError( "featurizer must be a subclass of MaterialStructureFeaturizer.") if issubclass(splitter, Splitter): splitter = splitter() else: raise TypeError("splitter must be a subclass of Splitter.") # Reload from disk if reload: featurizer_name = str(featurizer.__class__.__name__) splitter_name = str(splitter.__class__.__name__) save_folder = os.path.join(save_dir, "mp-forme-featurized", featurizer_name, splitter_name) loaded, all_dataset, transformers = deepchem.utils.data_utils.load_dataset_from_disk( save_folder) if loaded: return my_tasks, all_dataset, transformers # First type of supported featurizers supported_featurizers: List[str] = [ 'CGCNNFeaturizer', 'SineCoulombMatrix', ] # Load .tar.gz file if featurizer.__class__.__name__ in supported_featurizers: dataset_file = os.path.join(data_dir, 'mp_formation_energy.json') if not os.path.exists(dataset_file): targz_file = os.path.join(data_dir, 'mp_formation_energy.tar.gz') if not os.path.exists(targz_file): deepchem.utils.data_utils.download_url( url=MPFORME_URL, dest_dir=data_dir) deepchem.utils.data_utils.untargz_file( os.path.join(data_dir, 'mp_formation_energy.tar.gz'), data_dir) # Changer loader to match featurizer and data file type loader = deepchem.data.JsonLoader( tasks=my_tasks, feature_field="structure", label_field="formation_energy", featurizer=featurizer) # Featurize dataset dataset = loader.create_dataset(dataset_file) train_dataset, valid_dataset, test_dataset = splitter.train_valid_test_split( dataset, **splitter_kwargs) # Initialize transformers transformers = [ DEFAULT_TRANSFORMERS[t](dataset=dataset, **transformer_kwargs[t]) if isinstance(t, str) else t( dataset=dataset, **transformer_kwargs[str(t.__name__)]) for t in transformers ] for transformer in transformers: train_dataset = transformer.transform(train_dataset) valid_dataset = transformer.transform(valid_dataset) test_dataset = transformer.transform(test_dataset) if reload: # save to disk deepchem.utils.data_utils.save_dataset_to_disk( save_folder, train_dataset, valid_dataset, test_dataset, transformers) return my_tasks, (train_dataset, valid_dataset, test_dataset), transformers loader = _MPFormationLoader(featurizer, splitter, transformers, MPFORME_TASKS, data_dir, save_dir, **kwargs) return loader.load_dataset('mp-forme', reload) deepchem/molnet/load_function/material_datasets/load_mp_metallicity.py +48 −148 File changed.Preview size limit exceeded, changes collapsed. Show changes Loading
deepchem/feat/material_featurizers/element_property_fingerprint.py +8 −6 Original line number Diff line number Diff line Loading @@ -50,13 +50,8 @@ class ElementPropertyFingerprint(MaterialCompositionFeaturizer): data_source: str of "matminer", "magpie" or "deml" (default "matminer") Source for element property data. """ try: from matminer.featurizers.composition import ElementProperty except ModuleNotFoundError: raise ImportError("This class requires matminer to be installed.") self.data_source = data_source self.ep_featurizer = ElementProperty.from_preset(self.data_source) self.ep_featurizer = None def _featurize(self, composition: PymatgenComposition) -> np.ndarray: """ Loading @@ -73,6 +68,13 @@ class ElementPropertyFingerprint(MaterialCompositionFeaturizer): Vector of properties and statistics derived from chemical stoichiometry. Some values may be NaN. """ if self.ep_featurizer is None: try: from matminer.featurizers.composition import ElementProperty self.ep_featurizer = ElementProperty.from_preset(self.data_source) except ModuleNotFoundError: raise ImportError("This class requires matminer to be installed.") try: feats = self.ep_featurizer.featurize(composition) except: Loading
deepchem/feat/material_featurizers/sine_coulomb_matrix.py +8 −6 Original line number Diff line number Diff line Loading @@ -54,14 +54,9 @@ class SineCoulombMatrix(MaterialStructureFeaturizer): flatten: bool (default True) Return flattened vector of matrix eigenvalues. """ try: from matminer.featurizers.structure import SineCoulombMatrix as SCM except ModuleNotFoundError: raise ImportError("This class requires matminer to be installed.") self.max_atoms = max_atoms self.flatten = flatten self.scm = SCM(flatten=False) self.scm = None def _featurize(self, struct: PymatgenStructure) -> np.ndarray: """ Loading @@ -79,6 +74,13 @@ class SineCoulombMatrix(MaterialStructureFeaturizer): 2D sine Coulomb matrix with shape (max_atoms, max_atoms), or 1D matrix eigenvalues with shape (max_atoms,). """ if self.scm is None: try: from matminer.featurizers.structure import SineCoulombMatrix as SCM self.scm = SCM(flatten=False) except ModuleNotFoundError: raise ImportError("This class requires matminer to be installed.") # Get full N x N SCM sine_mat = self.scm.featurize(struct) Loading
deepchem/molnet/load_function/material_datasets/load_bandgap.py +49 −146 Original line number Diff line number Diff line Loading @@ -2,60 +2,43 @@ Experimental bandgaps for inorganic crystals. """ import os import logging import deepchem as dc from deepchem.molnet.load_function.molnet_loader import TransformerGenerator, _MolnetLoader from deepchem.data import Dataset from typing import List, Optional, Tuple, Union import deepchem from deepchem.feat import MaterialCompositionFeaturizer from deepchem.splits.splitters import Splitter from deepchem.molnet.defaults import get_defaults from typing import List, Tuple, Dict, Optional, Any logger = logging.getLogger(__name__) DEFAULT_DIR = deepchem.utils.data_utils.get_data_dir() BANDGAP_URL = 'https://deepchemdata.s3-us-west-1.amazonaws.com/datasets/expt_gap.tar.gz' BANDGAP_TASKS = ['experimental_bandgap'] # dict of accepted featurizers for this dataset # modify the returned dicts for your dataset DEFAULT_FEATURIZERS = get_defaults("feat") # Names of supported featurizers featurizers = [ 'ElementPropertyFingerprint', ] DEFAULT_FEATURIZERS = {k: DEFAULT_FEATURIZERS[k] for k in featurizers} # dict of accepted transformers DEFAULT_TRANSFORMERS = get_defaults("trans") class _BandgapLoader(_MolnetLoader): # dict of accepted splitters DEFAULT_SPLITTERS = get_defaults("splits") # names of supported splitters splitters = ['RandomSplitter'] DEFAULT_SPLITTERS = {k: DEFAULT_SPLITTERS[k] for k in splitters} def create_dataset(self) -> Dataset: dataset_file = os.path.join(self.data_dir, 'expt_gap.json') targz_file = os.path.join(self.data_dir, 'expt_gap.tar.gz') if not os.path.exists(dataset_file): if not os.path.exists(targz_file): dc.utils.data_utils.download_url( url=BANDGAP_URL, dest_dir=self.data_dir) dc.utils.data_utils.untargz_file(targz_file, self.data_dir) loader = dc.data.JsonLoader( tasks=self.tasks, feature_field="composition", label_field="experimental_bandgap", featurizer=self.featurizer) return loader.create_dataset(dataset_file) def load_bandgap( featurizer=DEFAULT_FEATURIZERS['ElementPropertyFingerprint'], transformers: List = [DEFAULT_TRANSFORMERS['NormalizationTransformer']], splitter=DEFAULT_SPLITTERS['RandomSplitter'], featurizer: Union[dc.feat.Featurizer, str] = dc.feat.ElementPropertyFingerprint(), splitter: Union[dc.splits.Splitter, str, None] = 'random', transformers: List[Union[TransformerGenerator, str]] = ['normalization'], reload: bool = True, data_dir: Optional[str] = None, save_dir: Optional[str] = None, featurizer_kwargs: Dict[str, Any] = {}, splitter_kwargs: Dict[str, Any] = { 'frac_train': 0.8, 'frac_valid': 0.1, 'frac_test': 0.1 }, transformer_kwargs: Dict[str, Dict[str, Any]] = { 'NormalizationTransformer': { 'transform_X': True } }, **kwargs) -> Tuple[List, Optional[Tuple], List]: **kwargs ) -> Tuple[List[str], Tuple[Dataset, ...], List[dc.trans.Transformer]]: """Load band gap dataset. Contains 4604 experimentally measured band gaps for inorganic Loading @@ -68,27 +51,25 @@ def load_bandgap( Parameters ---------- featurizer : MaterialCompositionFeaturizer (default ElementPropertyFingerprint) A featurizer that inherits from deepchem.feat.Featurizer. transformers : List[Transformer] A transformer that inherits from deepchem.trans.Transformer. splitter : Splitter (default RandomSplitter) A splitter that inherits from deepchem.splits.splitters.Splitter. reload : bool (default True) Try to reload dataset from disk if already downloaded. Save to disk after featurizing. data_dir : str, optional (default None) Path to datasets. save_dir : str, optional (default None) Path to featurized datasets. featurizer_kwargs : Dict[str, Any] Specify parameters to featurizer, e.g. {"size": 1024} splitter_kwargs : Dict[str, Any] Specify parameters to splitter, e.g. {"seed": 42} transformer_kwargs : dict Maps transformer names to constructor arguments, e.g. {"BalancingTransformer": {"transform_x":True, "transform_y":False}} **kwargs : additional optional arguments. featurizer: Featurizer or str the featurizer to use for processing the data. Alternatively you can pass one of the names from dc.molnet.featurizers as a shortcut. splitter: Splitter or str the splitter to use for splitting the data into training, validation, and test sets. Alternatively you can pass one of the names from dc.molnet.splitters as a shortcut. If this is None, all the data will be included in a single dataset. transformers: list of TransformerGenerators or strings the Transformers to apply to the data. Each one is specified by a TransformerGenerator or, as a shortcut, one of the names from dc.molnet.transformers. reload: bool if True, the first call for a particular featurizer and splitter will cache the datasets to disk, and subsequent calls will reload the cached datasets. data_dir: str a directory to save the raw data in save_dir: str a directory to save the dataset in Returns ------- Loading @@ -112,91 +93,13 @@ def load_bandgap( Examples -------- >> import deepchem as dc >> tasks, datasets, transformers = dc.molnet.load_bandgap(reload=False) >> tasks, datasets, transformers = dc.molnet.load_bandgap() >> train_dataset, val_dataset, test_dataset = datasets >> n_tasks = len(tasks) >> n_features = train_dataset.get_data_shape()[0] >> model = dc.models.MultitaskRegressor(n_tasks, n_features) """ # Featurize logger.info("About to featurize band gap dataset.") my_tasks = ['experimental_bandgap'] # machine learning targets # Get DeepChem data directory if needed if data_dir is None: data_dir = DEFAULT_DIR if save_dir is None: save_dir = DEFAULT_DIR if issubclass(featurizer, MaterialCompositionFeaturizer): featurizer = featurizer(**featurizer_kwargs) else: raise TypeError( "featurizer must be a subclass of MaterialCompositionFeaturizer.") if issubclass(splitter, Splitter): splitter = splitter() else: raise TypeError("splitter must be a subclass of Splitter.") # Reload from disk if reload: featurizer_name = str(featurizer.__class__.__name__) splitter_name = str(splitter.__class__.__name__) save_folder = os.path.join(save_dir, "bandgap-featurized", featurizer_name, splitter_name) loaded, all_dataset, transformers = deepchem.utils.data_utils.load_dataset_from_disk( save_folder) if loaded: return my_tasks, all_dataset, transformers # First type of supported featurizers supported_featurizers: List[str] = ['ElementPropertyFingerprint'] # Load .tar.gz file if featurizer.__class__.__name__ in supported_featurizers: dataset_file = os.path.join(data_dir, 'expt_gap.json') if not os.path.exists(dataset_file): targz_file = os.path.join(data_dir, 'expt_gap.tar.gz') if not os.path.exists(targz_file): deepchem.utils.data_utils.download_url( url=BANDGAP_URL, dest_dir=data_dir) deepchem.utils.data_utils.untargz_file( os.path.join(data_dir, 'expt_gap.tar.gz'), data_dir) # Changer loader to match featurizer and data file type loader = deepchem.data.JsonLoader( tasks=my_tasks, feature_field="composition", label_field="experimental_bandgap", featurizer=featurizer) # Featurize dataset dataset = loader.create_dataset(dataset_file) train_dataset, valid_dataset, test_dataset = splitter.train_valid_test_split( dataset, **splitter_kwargs) # Initialize transformers transformers = [ DEFAULT_TRANSFORMERS[t](dataset=dataset, **transformer_kwargs[t]) if isinstance(t, str) else t( dataset=dataset, **transformer_kwargs[str(t.__name__)]) for t in transformers ] for transformer in transformers: train_dataset = transformer.transform(train_dataset) valid_dataset = transformer.transform(valid_dataset) test_dataset = transformer.transform(test_dataset) if reload: # save to disk deepchem.utils.data_utils.save_dataset_to_disk( save_folder, train_dataset, valid_dataset, test_dataset, transformers) return my_tasks, (train_dataset, valid_dataset, test_dataset), transformers loader = _BandgapLoader(featurizer, splitter, transformers, BANDGAP_TASKS, data_dir, save_dir, **kwargs) return loader.load_dataset('bandgap', reload)
deepchem/molnet/load_function/material_datasets/load_mp_formation_energy.py +48 −148 Original line number Diff line number Diff line Loading @@ -2,60 +2,42 @@ Calculated formation energies for inorganic crystals from Materials Project. """ import os import logging import deepchem from deepchem.feat import MaterialStructureFeaturizer from deepchem.splits.splitters import Splitter from deepchem.molnet.defaults import get_defaults import deepchem as dc from deepchem.molnet.load_function.molnet_loader import TransformerGenerator, _MolnetLoader from deepchem.data import Dataset from typing import List, Optional, Tuple, Union from typing import List, Tuple, Dict, Optional, Any logger = logging.getLogger(__name__) DEFAULT_DIR = deepchem.utils.data_utils.get_data_dir() MPFORME_URL = 'https://deepchemdata.s3-us-west-1.amazonaws.com/datasets/mp_formation_energy.tar.gz' MPFORME_TASKS = ['formation_energy'] # dict of accepted featurizers for this dataset # modify the returned dicts for your dataset DEFAULT_FEATURIZERS = get_defaults("feat") # Names of supported featurizers featurizers = [ 'CGCNNFeaturizer', 'SineCoulombMatrix', ] DEFAULT_FEATURIZERS = {k: DEFAULT_FEATURIZERS[k] for k in featurizers} # dict of accepted transformers DEFAULT_TRANSFORMERS = get_defaults("trans") # dict of accepted splitters DEFAULT_SPLITTERS = get_defaults("splits") class _MPFormationLoader(_MolnetLoader): # names of supported splitters splitters = ['RandomSplitter'] DEFAULT_SPLITTERS = {k: DEFAULT_SPLITTERS[k] for k in splitters} def create_dataset(self) -> Dataset: dataset_file = os.path.join(self.data_dir, 'mp_formation_energy.json') targz_file = os.path.join(self.data_dir, 'mp_formation_energy.tar.gz') if not os.path.exists(dataset_file): if not os.path.exists(targz_file): dc.utils.data_utils.download_url( url=MPFORME_URL, dest_dir=self.data_dir) dc.utils.data_utils.untargz_file(targz_file, self.data_dir) loader = dc.data.JsonLoader( tasks=self.tasks, feature_field="structure", label_field="formation_energy", featurizer=self.featurizer) return loader.create_dataset(dataset_file) def load_mp_formation_energy( featurizer=DEFAULT_FEATURIZERS['SineCoulombMatrix'], transformers: List = [DEFAULT_TRANSFORMERS['NormalizationTransformer']], splitter=DEFAULT_SPLITTERS['RandomSplitter'], featurizer: Union[dc.feat.Featurizer, str] = dc.feat.SineCoulombMatrix(), splitter: Union[dc.splits.Splitter, str, None] = 'random', transformers: List[Union[TransformerGenerator, str]] = ['normalization'], reload: bool = True, data_dir: Optional[str] = None, save_dir: Optional[str] = None, featurizer_kwargs: Dict[str, Any] = {}, splitter_kwargs: Dict[str, Any] = { 'frac_train': 0.8, 'frac_valid': 0.1, 'frac_test': 0.1 }, transformer_kwargs: Dict[str, Dict[str, Any]] = { 'NormalizationTransformer': { 'transform_X': True } }, **kwargs) -> Tuple[List, Optional[Tuple], List]: **kwargs ) -> Tuple[List[str], Tuple[Dataset, ...], List[dc.trans.Transformer]]: """Load mp formation energy dataset. Contains 132752 calculated formation energies and inorganic Loading @@ -69,27 +51,25 @@ def load_mp_formation_energy( Parameters ---------- featurizer : MaterialStructureFeaturizer (default SineCoulombMatrix) A featurizer that inherits from deepchem.feat.Featurizer. transformers : List[Transformer] A transformer that inherits from deepchem.trans.Transformer. splitter : Splitter (default RandomSplitter) A splitter that inherits from deepchem.splits.splitters.Splitter. reload : bool (default True) Try to reload dataset from disk if already downloaded. Save to disk after featurizing. data_dir : str, optional (default None) Path to datasets. save_dir : str, optional (default None) Path to featurized datasets. featurizer_kwargs : Dict[str, Any] Specify parameters to featurizer, e.g. {"size": 1024} splitter_kwargs : Dict[str, Any] Specify parameters to splitter, e.g. {"seed": 42} transformer_kwargs : dict Maps transformer names to constructor arguments, e.g. {"BalancingTransformer": {"transform_X":True, "transform_y":False}} **kwargs : additional optional arguments. featurizer: Featurizer or str the featurizer to use for processing the data. Alternatively you can pass one of the names from dc.molnet.featurizers as a shortcut. splitter: Splitter or str the splitter to use for splitting the data into training, validation, and test sets. Alternatively you can pass one of the names from dc.molnet.splitters as a shortcut. If this is None, all the data will be included in a single dataset. transformers: list of TransformerGenerators or strings the Transformers to apply to the data. Each one is specified by a TransformerGenerator or, as a shortcut, one of the names from dc.molnet.transformers. reload: bool if True, the first call for a particular featurizer and splitter will cache the datasets to disk, and subsequent calls will reload the cached datasets. data_dir: str a directory to save the raw data in save_dir: str a directory to save the dataset in Returns ------- Loading @@ -114,93 +94,13 @@ def load_mp_formation_energy( Examples -------- >> import deepchem as dc >> tasks, datasets, transformers = dc.molnet.load_mp_formation_energy(reload=False) >> tasks, datasets, transformers = dc.molnet.load_mp_formation_energy() >> train_dataset, val_dataset, test_dataset = datasets >> n_tasks = len(tasks) >> n_features = train_dataset.get_data_shape()[0] >> model = dc.models.MultitaskRegressor(n_tasks, n_features) """ # Featurize logger.info("About to featurize mp formation energy dataset.") my_tasks = ['formation_energy'] # machine learning targets # Get DeepChem data directory if needed if data_dir is None: data_dir = DEFAULT_DIR if save_dir is None: save_dir = DEFAULT_DIR if issubclass(featurizer, MaterialStructureFeaturizer): featurizer = featurizer(**featurizer_kwargs) else: raise TypeError( "featurizer must be a subclass of MaterialStructureFeaturizer.") if issubclass(splitter, Splitter): splitter = splitter() else: raise TypeError("splitter must be a subclass of Splitter.") # Reload from disk if reload: featurizer_name = str(featurizer.__class__.__name__) splitter_name = str(splitter.__class__.__name__) save_folder = os.path.join(save_dir, "mp-forme-featurized", featurizer_name, splitter_name) loaded, all_dataset, transformers = deepchem.utils.data_utils.load_dataset_from_disk( save_folder) if loaded: return my_tasks, all_dataset, transformers # First type of supported featurizers supported_featurizers: List[str] = [ 'CGCNNFeaturizer', 'SineCoulombMatrix', ] # Load .tar.gz file if featurizer.__class__.__name__ in supported_featurizers: dataset_file = os.path.join(data_dir, 'mp_formation_energy.json') if not os.path.exists(dataset_file): targz_file = os.path.join(data_dir, 'mp_formation_energy.tar.gz') if not os.path.exists(targz_file): deepchem.utils.data_utils.download_url( url=MPFORME_URL, dest_dir=data_dir) deepchem.utils.data_utils.untargz_file( os.path.join(data_dir, 'mp_formation_energy.tar.gz'), data_dir) # Changer loader to match featurizer and data file type loader = deepchem.data.JsonLoader( tasks=my_tasks, feature_field="structure", label_field="formation_energy", featurizer=featurizer) # Featurize dataset dataset = loader.create_dataset(dataset_file) train_dataset, valid_dataset, test_dataset = splitter.train_valid_test_split( dataset, **splitter_kwargs) # Initialize transformers transformers = [ DEFAULT_TRANSFORMERS[t](dataset=dataset, **transformer_kwargs[t]) if isinstance(t, str) else t( dataset=dataset, **transformer_kwargs[str(t.__name__)]) for t in transformers ] for transformer in transformers: train_dataset = transformer.transform(train_dataset) valid_dataset = transformer.transform(valid_dataset) test_dataset = transformer.transform(test_dataset) if reload: # save to disk deepchem.utils.data_utils.save_dataset_to_disk( save_folder, train_dataset, valid_dataset, test_dataset, transformers) return my_tasks, (train_dataset, valid_dataset, test_dataset), transformers loader = _MPFormationLoader(featurizer, splitter, transformers, MPFORME_TASKS, data_dir, save_dir, **kwargs) return loader.load_dataset('mp-forme', reload)
deepchem/molnet/load_function/material_datasets/load_mp_metallicity.py +48 −148 File changed.Preview size limit exceeded, changes collapsed. Show changes
