Loading deepchem/hyperparameters/__init__.py +3 −14 Original line number Diff line number Diff line Loading @@ -16,12 +16,8 @@ class HyperparamOpt(object): Provides simple hyperparameter search capabilities. """ def __init__(self, model_class, tasks, task_types, fit_transformers=None, verbosity=None): def __init__(self, model_class, verbosity=None): self.model_class = model_class self.tasks = tasks self.task_types = task_types self.fit_transformers = fit_transformers assert verbosity in [None, "low", "high"] self.verbosity = verbosity Loading Loading @@ -73,15 +69,8 @@ class HyperparamOpt(object): model_dir = tempfile.mkdtemp() else: model_dir = tempfile.mkdtemp() #TODO(JG) Fit transformers for TF models if self.fit_transformers: model = self.model_class( self.tasks, self.task_types, model_params, model_dir, fit_transformers=self.fit_transformers, verbosity=self.verbosity) else: model = self.model_class( self.tasks, self.task_types, model_params, model_dir, verbosity=self.verbosity) model = self.model_class(model_params, model_dir) model.fit(train_dataset) model.save() Loading deepchem/hyperparameters/tests/test_hyperparam_opt.py +32 −70 Original line number Diff line number Diff line Loading @@ -27,6 +27,7 @@ from sklearn.ensemble import RandomForestRegressor from deepchem.datasets import Dataset from deepchem.hyperparameters import HyperparamOpt from deepchem.models.keras_models.fcnet import MultiTaskDNN from deepchem.models.keras_models import KerasModel from deepchem.models.tensorflow_models import TensorflowModel from deepchem.models.tensorflow_models.fcnet import TensorflowMultiTaskClassifier from deepchem.splits import ScaffoldSplitter Loading Loading @@ -146,9 +147,9 @@ class TestHyperparamOptAPI(TestAPI): tasks = ["task0", "task1", "task2", "task3", "task4", "task5", "task6", "task7", "task8", "task9", "task10", "task11", "task12", "task13", "task14", "task15", "task16"] task_types = {task: task_type for task in tasks} featurizer = CircularFingerprint(size=1024) n_features = 1024 featurizer = CircularFingerprint(size=n_features) loader = DataLoader(tasks=tasks, smiles_field=self.smiles_field, featurizer=featurizer, Loading @@ -161,30 +162,13 @@ class TestHyperparamOptAPI(TestAPI): transformers = [] metric = Metric(metrics.matthews_corrcoef, np.mean, mode="classification") params_dict= {"nb_hidden": [5, 10], "activation": ["relu"], "dropout": [.5], "learning_rate": [.01], "momentum": [.9], "nesterov": [False], "decay": [1e-4], "batch_size": [5], "nb_epoch": [2], "init": ["glorot_uniform"], "nb_layers": [1], "batchnorm": [False], "data_shape": [train_dataset.get_data_shape()]} params_dict= {"n_hidden": [5, 10]} def model_builder(tasks, task_types, model_params, task_model_dir, verbosity=None): g = tf.Graph() sess = tf.Session(graph=g) K.set_session(sess) with g.as_default(): return MultiTaskDNN(tasks, task_types, model_params, task_model_dir, model_instance=LogisticRegression()) optimizer = HyperparamOpt(MultiTaskDNN, tasks, task_types, verbosity="low") def model_builder(model_params, model_dir): keras_model = MultiTaskDNN( len(tasks), n_features, task_type, dropout=0., **model_params) return KerasModel(keras_model, self.model_dir) optimizer = HyperparamOpt(model_builder, verbosity="low") best_model, best_hyperparams, all_results = optimizer.hyperparam_search( params_dict, train_dataset, valid_dataset, transformers, metric, logdir=None) Loading @@ -198,9 +182,9 @@ class TestHyperparamOptAPI(TestAPI): tasks = ["task0", "task1", "task2", "task3", "task4", "task5", "task6", "task7", "task8", "task9", "task10", "task11", "task12", "task13", "task14", "task15", "task16"] task_types = {task: task_type for task in tasks} featurizer = CircularFingerprint(size=1024) n_features = 1024 featurizer = CircularFingerprint(size=n_features) loader = DataLoader(tasks=tasks, smiles_field=self.smiles_field, Loading @@ -212,38 +196,16 @@ class TestHyperparamOptAPI(TestAPI): train_dataset, valid_dataset, test_dataset = splitter.train_valid_test_split( dataset, self.train_dir, self.valid_dir, self.test_dir) transformers = [] transformers = [] metric = Metric(metrics.matthews_corrcoef, np.mean, mode="classification") params_dict = {"activation": ["relu"], "momentum": [.9], "batch_size": [50], "init": ["glorot_uniform"], "data_shape": [train_dataset.get_data_shape()], "learning_rate": [1e-3], "decay": [1e-6], "nb_hidden": [1000], "nb_epoch": [1], "nesterov": [False], "dropouts": [(.5,)], "nb_layers": [1], "batchnorm": [False], "layer_sizes": [(1000,)], "weight_init_stddevs": [(.1,)], "bias_init_consts": [(1.,)], "num_classes": [2], "penalty": [0.], "optimizer": ["sgd"], "num_classification_tasks": [len(task_types)] } params_dict = {"layer_sizes": [(10,), (100,)]} def model_builder(tasks, task_types, params_dict, logdir, verbosity=None): return TensorflowModel( tasks, task_types, params_dict, logdir, tf_class=TensorflowMultiTaskClassifier, verbosity=verbosity) optimizer = HyperparamOpt(model_builder, tasks, task_types, verbosity="low") def model_builder(model_params, model_dir): tensorflow_model = TensorflowMultiTaskClassifier( len(tasks), n_features, model_dir, **model_params) return TensorflowModel(tensorflow_model, model_dir) optimizer = HyperparamOpt(model_builder, verbosity="low") best_model, best_hyperparams, all_results = optimizer.hyperparam_search( params_dict, train_dataset, valid_dataset, transformers, metric, logdir=None) params_dict, train_dataset, valid_dataset, transformers, metric, logdir=None) Loading
deepchem/hyperparameters/__init__.py +3 −14 Original line number Diff line number Diff line Loading @@ -16,12 +16,8 @@ class HyperparamOpt(object): Provides simple hyperparameter search capabilities. """ def __init__(self, model_class, tasks, task_types, fit_transformers=None, verbosity=None): def __init__(self, model_class, verbosity=None): self.model_class = model_class self.tasks = tasks self.task_types = task_types self.fit_transformers = fit_transformers assert verbosity in [None, "low", "high"] self.verbosity = verbosity Loading Loading @@ -73,15 +69,8 @@ class HyperparamOpt(object): model_dir = tempfile.mkdtemp() else: model_dir = tempfile.mkdtemp() #TODO(JG) Fit transformers for TF models if self.fit_transformers: model = self.model_class( self.tasks, self.task_types, model_params, model_dir, fit_transformers=self.fit_transformers, verbosity=self.verbosity) else: model = self.model_class( self.tasks, self.task_types, model_params, model_dir, verbosity=self.verbosity) model = self.model_class(model_params, model_dir) model.fit(train_dataset) model.save() Loading
deepchem/hyperparameters/tests/test_hyperparam_opt.py +32 −70 Original line number Diff line number Diff line Loading @@ -27,6 +27,7 @@ from sklearn.ensemble import RandomForestRegressor from deepchem.datasets import Dataset from deepchem.hyperparameters import HyperparamOpt from deepchem.models.keras_models.fcnet import MultiTaskDNN from deepchem.models.keras_models import KerasModel from deepchem.models.tensorflow_models import TensorflowModel from deepchem.models.tensorflow_models.fcnet import TensorflowMultiTaskClassifier from deepchem.splits import ScaffoldSplitter Loading Loading @@ -146,9 +147,9 @@ class TestHyperparamOptAPI(TestAPI): tasks = ["task0", "task1", "task2", "task3", "task4", "task5", "task6", "task7", "task8", "task9", "task10", "task11", "task12", "task13", "task14", "task15", "task16"] task_types = {task: task_type for task in tasks} featurizer = CircularFingerprint(size=1024) n_features = 1024 featurizer = CircularFingerprint(size=n_features) loader = DataLoader(tasks=tasks, smiles_field=self.smiles_field, featurizer=featurizer, Loading @@ -161,30 +162,13 @@ class TestHyperparamOptAPI(TestAPI): transformers = [] metric = Metric(metrics.matthews_corrcoef, np.mean, mode="classification") params_dict= {"nb_hidden": [5, 10], "activation": ["relu"], "dropout": [.5], "learning_rate": [.01], "momentum": [.9], "nesterov": [False], "decay": [1e-4], "batch_size": [5], "nb_epoch": [2], "init": ["glorot_uniform"], "nb_layers": [1], "batchnorm": [False], "data_shape": [train_dataset.get_data_shape()]} params_dict= {"n_hidden": [5, 10]} def model_builder(tasks, task_types, model_params, task_model_dir, verbosity=None): g = tf.Graph() sess = tf.Session(graph=g) K.set_session(sess) with g.as_default(): return MultiTaskDNN(tasks, task_types, model_params, task_model_dir, model_instance=LogisticRegression()) optimizer = HyperparamOpt(MultiTaskDNN, tasks, task_types, verbosity="low") def model_builder(model_params, model_dir): keras_model = MultiTaskDNN( len(tasks), n_features, task_type, dropout=0., **model_params) return KerasModel(keras_model, self.model_dir) optimizer = HyperparamOpt(model_builder, verbosity="low") best_model, best_hyperparams, all_results = optimizer.hyperparam_search( params_dict, train_dataset, valid_dataset, transformers, metric, logdir=None) Loading @@ -198,9 +182,9 @@ class TestHyperparamOptAPI(TestAPI): tasks = ["task0", "task1", "task2", "task3", "task4", "task5", "task6", "task7", "task8", "task9", "task10", "task11", "task12", "task13", "task14", "task15", "task16"] task_types = {task: task_type for task in tasks} featurizer = CircularFingerprint(size=1024) n_features = 1024 featurizer = CircularFingerprint(size=n_features) loader = DataLoader(tasks=tasks, smiles_field=self.smiles_field, Loading @@ -212,38 +196,16 @@ class TestHyperparamOptAPI(TestAPI): train_dataset, valid_dataset, test_dataset = splitter.train_valid_test_split( dataset, self.train_dir, self.valid_dir, self.test_dir) transformers = [] transformers = [] metric = Metric(metrics.matthews_corrcoef, np.mean, mode="classification") params_dict = {"activation": ["relu"], "momentum": [.9], "batch_size": [50], "init": ["glorot_uniform"], "data_shape": [train_dataset.get_data_shape()], "learning_rate": [1e-3], "decay": [1e-6], "nb_hidden": [1000], "nb_epoch": [1], "nesterov": [False], "dropouts": [(.5,)], "nb_layers": [1], "batchnorm": [False], "layer_sizes": [(1000,)], "weight_init_stddevs": [(.1,)], "bias_init_consts": [(1.,)], "num_classes": [2], "penalty": [0.], "optimizer": ["sgd"], "num_classification_tasks": [len(task_types)] } params_dict = {"layer_sizes": [(10,), (100,)]} def model_builder(tasks, task_types, params_dict, logdir, verbosity=None): return TensorflowModel( tasks, task_types, params_dict, logdir, tf_class=TensorflowMultiTaskClassifier, verbosity=verbosity) optimizer = HyperparamOpt(model_builder, tasks, task_types, verbosity="low") def model_builder(model_params, model_dir): tensorflow_model = TensorflowMultiTaskClassifier( len(tasks), n_features, model_dir, **model_params) return TensorflowModel(tensorflow_model, model_dir) optimizer = HyperparamOpt(model_builder, verbosity="low") best_model, best_hyperparams, all_results = optimizer.hyperparam_search( params_dict, train_dataset, valid_dataset, transformers, metric, logdir=None) params_dict, train_dataset, valid_dataset, transformers, metric, logdir=None)
