Loading deepchem/models/__init__.py +27 −90 Original line number Diff line number Diff line Loading @@ -9,48 +9,12 @@ import numpy as np import pandas as pd import joblib import os from keras.models import model_from_json from deepchem.utils.dataset import NumpyDataset from deepchem.utils.dataset import load_sharded_dataset from deepchem.utils.dataset import save_sharded_dataset ''' def get_parameter_filename(model_dir): """ Given model directory, obtain filename for stored parameters. """ filename = os.path.join(model_dir, "model_params.joblib") return filename ''' # TODO(rbharath): Make these instance methods... def save_sklearn_model(model, filename): """Saves sklearn model to disk using joblib.""" joblib.dump(model, filename) def save_keras_model(model, filename): """Saves keras models to disk.""" filename, _ = os.path.splitext(filename) # Note that keras requires the model architecture and weights to be stored # separately. A json file is generated that specifies the model architecture. # The weights will be stored in an h5 file. The pkl.gz file with store the # target name. json_filename = "%s.%s" % (filename, "json") h5_filename = "%s.%s" % (filename, "h5") # Save architecture json_string = model.to_json() with open(json_filename, "wb") as file_obj: file_obj.write(json_string) model.save_weights(h5_filename, overwrite=True) def get_model_filename(model_dir): """ Given model directory, obtain filename for the model itself. """ filename = os.path.join(model_dir, "model_params.joblib") return filename # TODO(rbharath): Make a static method def get_model_type(model_name): Loading @@ -66,27 +30,6 @@ def get_model_type(model_name): model_type = "sklearn" return model_type # TODO(rbharath): Make this an instance method of Model objects. def load_sklearn_model(filename): """Loads sklearn model from file on disk.""" return joblib.load(filename) def load_keras_model(filename): """Loads keras model from disk. Assumes that filename.json and filename.h5 respectively contain the model architecture and weights. """ filename, _ = os.path.splitext(filename) json_filename = "%s.%s" % (filename, "json") h5_filename = "%s.%s" % (filename, "h5") with open(json_filename) as file_obj: model = model_from_json(file_obj.read()) model.load_weights(h5_filename) return model #TODO(enf/rbharath): incorporate save, load, eval, fit features into class Model. class Model(object): """ Loading @@ -94,7 +37,9 @@ class Model(object): """ # List of registered models registered_model_types = {} def __init__(self, task_types, model_params, initialize_raw_model=True): def __init__(self, model_type, task_types, model_params, initialize_raw_model=True): self.model_type = model_type self.task_types = task_types self.model_params = model_params Loading Loading @@ -124,6 +69,18 @@ class Model(object): """ return(self.raw_model) def get_param_filename(self, out_dir): """ Given model directory, obtain filename for the model itself. """ return os.path.join(out_dir, "model_params.joblib") def get_model_filename(self, out_dir): """ Given model directory, obtain filename for the model itself. """ return os.path.join(out_dir, "model.joblib") @staticmethod def model_builder(model_type, task_types, model_params, initialize_raw_model=True): Loading @@ -132,7 +89,7 @@ class Model(object): """ if model_type in Model.registered_model_types: model = Model.registered_model_types[model_type]( task_types, model_params, initialize_raw_model) model_type, task_types, model_params, initialize_raw_model) else: raise ValueError("model_type %s is not supported" % model_type) return model Loading @@ -144,40 +101,20 @@ class Model(object): """ Model.registered_model_types[model_type] = model_class @staticmethod def load_model(model_name, model_dir): def load(self, model_dir): """Dispatcher function for loading.""" model_type = get_model_type(model_name) params = load_sharded_dataset(get_model_filename(model_dir)) model = Model.model_builder(model_name, params["task_types"], params["model_params"], initialize_raw_model=False) if model_type == "sklearn": raw_model = load_sklearn_model(get_model_filename(model_dir)) elif "keras" in model_type: raw_model = load_keras_model(get_model_filename(model_dir)) else: raise ValueError("Unsupported model_type.") model.set_raw_model(raw_model) return model #model_type = get_model_type(model_name) params = load_sharded_dataset(self.get_model_filename(model_dir)) self.model_params = params["model_params"] self.task_types = params["task_types"] self.model_type = params["model_type"] # TODO(rbharath): This really shouldn't be a static method. Make an instance # method instance. @staticmethod def save_model(model, model_name, model_dir): def save(self, out_dir): """Dispatcher function for saving.""" model_type = get_model_type(model_name) params = {"model_params" : model.model_params, "task_types" : model.task_types} save_sharded_dataset(params, get_model_filename(model_dir)) raw_model = model.get_raw_model() if model_type == "sklearn": save_sklearn_model(raw_model, get_model_filename(model_dir)) elif "keras" in model_type: save_keras_model(raw_model, get_model_filename(model_dir)) else: raise ValueError("Unsupported model_type.") params = {"model_params" : self.model_params, "task_types" : self.task_types, "model_type": self.model_type} save_sharded_dataset(params, self.get_params_filename(out_dir)) def fit(self, numpy_dataset): """ Loading deepchem/models/deep.py +39 −1 Original line number Diff line number Diff line Loading @@ -3,6 +3,7 @@ Code for processing the Google vs-datasets using keras. """ import numpy as np from keras.models import Graph from keras.models import model_from_json from keras.layers.core import Dense, Dropout from keras.optimizers import SGD from deepchem.models import Model Loading @@ -12,7 +13,7 @@ class MultiTaskDNN(Model): Abstract base class for different ML models. """ def __init__(self, task_types, model_params, initialize_raw_model=True): super(MultiTaskDNN, self).__init__(task_types, model_params, super(MultiTaskDNN, self).__init__(model_type, task_types, model_params, initialize_raw_model) if initialize_raw_model: sorted_tasks = sorted(task_types.keys()) Loading Loading @@ -116,6 +117,43 @@ class MultiTaskDNN(Model): y_pred = np.squeeze(y_pred) return y_pred def save(self, out_dir): """ Saves underlying keras model to disk. """ super(MultiTaskDNN, self).save(out_dir) model = self.get_raw_model() filename, _ = os.path.splitext(self.get_model_filename(out_dir)) # Note that keras requires the model architecture and weights to be stored # separately. A json file is generated that specifies the model architecture. # The weights will be stored in an h5 file. The pkl.gz file with store the # target name. json_filename = "%s.%s" % (filename, "json") h5_filename = "%s.%s" % (filename, "h5") # Save architecture json_string = model.to_json() with open(json_filename, "wb") as file_obj: file_obj.write(json_string) model.save_weights(h5_filename, overwrite=True) def load(self, model_dir): """ Load keras multitask DNN from disk. """ super(MultiTaskDNN, self).load(model_dir) filename = self.get_Model_filename(model_dir) filename, _ = os.path.splitext(filename) json_filename = "%s.%s" % (filename, "json") h5_filename = "%s.%s" % (filename, "h5") with open(json_filename) as file_obj: model = model_from_json(file_obj.read()) model.load_weights(h5_filename) self.raw_model = model Model.register_model_type("multitask_deep_regressor", MultiTaskDNN) Model.register_model_type("multitask_deep_classifier", MultiTaskDNN) Loading deepchem/models/standard.py +54 −7 Original line number Diff line number Diff line Loading @@ -16,24 +16,70 @@ class SklearnModel(Model): Abstract base class for different ML models. """ def __init__(self, task_types, model_params, initialize_raw_model=True): super(SklearnModel, self).__init__(task_types, model_params, initialize_raw_model) self.task_types = task_types self.model_params = model_params self.raw_model = None if initialize_raw_model: if self.modeltype == "rf_regressor": raw_model = RandomForestRegressor( n_estimators=500, n_jobs=-1, warm_start=True, max_features="sqrt") elif self.modeltype == "rf_classifier": raw_model = RandomForestClassifier( n_estimators=500, n_jobs=-1, warm_start=True, max_features="sqrt") elif modeltype == "logistic": raw_model = LogisticRegression(class_weight="auto") elif modeltype == "linear": raw_model = LinearRegression(normalize=True) elif modeltype == "ridge": raw_model = RidgeCV(alphas=[0.01, 0.1, 1.0, 10.0], normalize=True) elif modeltype == "lasso": raw_model = LassoCV(max_iter=2000, n_jobs=-1) elif modeltype == "lasso_lars": raw_model = LassoLarsCV(max_iter=2000, n_jobs=-1) elif modeltype == "elastic_net": raw_model = ElasticNetCV(max_iter=2000, n_jobs=-1) else: raise ValueError("Invalid model type provided.") def fit_on_batch(self, X, y, w): # TODO(rbharath): This is a partial implementation! Does not work for a # datasets with more than one shard. def fit(self, numpy_dataset): """ Updates existing model with new information. Fits SKLearn model to data. """ raise NotImplementedError( "Each model is responsible for its own fit_on_batch method.") for (X, y, _, _) in numpy_dataset.itershards(): self.raw_model.fit(X, y) return def predict_on_batch(self, X): """ Makes predictions on given batch of new data. """ raise NotImplementedError( "Each model is responsible for its own predict_on_batch method.") return self.raw_model.predict(X) def save(self, out_dir): """Saves sklearn model to disk using joblib.""" super(SklearnModel, self).save(out_dir) joblib.dump(self.raw_model, self.get_model_filename(out_dir)) def load(self, model_dir): """Loads sklearn model from joblib file on disk.""" super(SklearnModel, self).load(model_dir) self.raw_model = joblib.load(self.get_model_filename(model_dir) Model.register_model_type("logistic", SklearnModel) Model.register_model_type("rf_classifier", SklearnModel) Model.register_model_type("rf_regressor", SklearnModel) Model.register_model_type("linear", SklearnModel) Model.register_model_type("ridge", SklearnModel) Model.register_model_type("lasso", SklearnModel) Model.register_model_type("lasso_lars", SklearnModel) Model.register_model_type("elastic_net", SklearnModel) # TODO(rbharath): Need to fix singletask dataset support. ''' def fit_singletask_models(train_data, modeltype): """Fits singletask linear regression models to potency. Loading Loading @@ -84,3 +130,4 @@ def fit_singletask_models(train_data, modeltype): model.fit(task_X_train, task_y_train.ravel()) models[task] = model return models ''' Loading
deepchem/models/__init__.py +27 −90 Original line number Diff line number Diff line Loading @@ -9,48 +9,12 @@ import numpy as np import pandas as pd import joblib import os from keras.models import model_from_json from deepchem.utils.dataset import NumpyDataset from deepchem.utils.dataset import load_sharded_dataset from deepchem.utils.dataset import save_sharded_dataset ''' def get_parameter_filename(model_dir): """ Given model directory, obtain filename for stored parameters. """ filename = os.path.join(model_dir, "model_params.joblib") return filename ''' # TODO(rbharath): Make these instance methods... def save_sklearn_model(model, filename): """Saves sklearn model to disk using joblib.""" joblib.dump(model, filename) def save_keras_model(model, filename): """Saves keras models to disk.""" filename, _ = os.path.splitext(filename) # Note that keras requires the model architecture and weights to be stored # separately. A json file is generated that specifies the model architecture. # The weights will be stored in an h5 file. The pkl.gz file with store the # target name. json_filename = "%s.%s" % (filename, "json") h5_filename = "%s.%s" % (filename, "h5") # Save architecture json_string = model.to_json() with open(json_filename, "wb") as file_obj: file_obj.write(json_string) model.save_weights(h5_filename, overwrite=True) def get_model_filename(model_dir): """ Given model directory, obtain filename for the model itself. """ filename = os.path.join(model_dir, "model_params.joblib") return filename # TODO(rbharath): Make a static method def get_model_type(model_name): Loading @@ -66,27 +30,6 @@ def get_model_type(model_name): model_type = "sklearn" return model_type # TODO(rbharath): Make this an instance method of Model objects. def load_sklearn_model(filename): """Loads sklearn model from file on disk.""" return joblib.load(filename) def load_keras_model(filename): """Loads keras model from disk. Assumes that filename.json and filename.h5 respectively contain the model architecture and weights. """ filename, _ = os.path.splitext(filename) json_filename = "%s.%s" % (filename, "json") h5_filename = "%s.%s" % (filename, "h5") with open(json_filename) as file_obj: model = model_from_json(file_obj.read()) model.load_weights(h5_filename) return model #TODO(enf/rbharath): incorporate save, load, eval, fit features into class Model. class Model(object): """ Loading @@ -94,7 +37,9 @@ class Model(object): """ # List of registered models registered_model_types = {} def __init__(self, task_types, model_params, initialize_raw_model=True): def __init__(self, model_type, task_types, model_params, initialize_raw_model=True): self.model_type = model_type self.task_types = task_types self.model_params = model_params Loading Loading @@ -124,6 +69,18 @@ class Model(object): """ return(self.raw_model) def get_param_filename(self, out_dir): """ Given model directory, obtain filename for the model itself. """ return os.path.join(out_dir, "model_params.joblib") def get_model_filename(self, out_dir): """ Given model directory, obtain filename for the model itself. """ return os.path.join(out_dir, "model.joblib") @staticmethod def model_builder(model_type, task_types, model_params, initialize_raw_model=True): Loading @@ -132,7 +89,7 @@ class Model(object): """ if model_type in Model.registered_model_types: model = Model.registered_model_types[model_type]( task_types, model_params, initialize_raw_model) model_type, task_types, model_params, initialize_raw_model) else: raise ValueError("model_type %s is not supported" % model_type) return model Loading @@ -144,40 +101,20 @@ class Model(object): """ Model.registered_model_types[model_type] = model_class @staticmethod def load_model(model_name, model_dir): def load(self, model_dir): """Dispatcher function for loading.""" model_type = get_model_type(model_name) params = load_sharded_dataset(get_model_filename(model_dir)) model = Model.model_builder(model_name, params["task_types"], params["model_params"], initialize_raw_model=False) if model_type == "sklearn": raw_model = load_sklearn_model(get_model_filename(model_dir)) elif "keras" in model_type: raw_model = load_keras_model(get_model_filename(model_dir)) else: raise ValueError("Unsupported model_type.") model.set_raw_model(raw_model) return model #model_type = get_model_type(model_name) params = load_sharded_dataset(self.get_model_filename(model_dir)) self.model_params = params["model_params"] self.task_types = params["task_types"] self.model_type = params["model_type"] # TODO(rbharath): This really shouldn't be a static method. Make an instance # method instance. @staticmethod def save_model(model, model_name, model_dir): def save(self, out_dir): """Dispatcher function for saving.""" model_type = get_model_type(model_name) params = {"model_params" : model.model_params, "task_types" : model.task_types} save_sharded_dataset(params, get_model_filename(model_dir)) raw_model = model.get_raw_model() if model_type == "sklearn": save_sklearn_model(raw_model, get_model_filename(model_dir)) elif "keras" in model_type: save_keras_model(raw_model, get_model_filename(model_dir)) else: raise ValueError("Unsupported model_type.") params = {"model_params" : self.model_params, "task_types" : self.task_types, "model_type": self.model_type} save_sharded_dataset(params, self.get_params_filename(out_dir)) def fit(self, numpy_dataset): """ Loading
deepchem/models/deep.py +39 −1 Original line number Diff line number Diff line Loading @@ -3,6 +3,7 @@ Code for processing the Google vs-datasets using keras. """ import numpy as np from keras.models import Graph from keras.models import model_from_json from keras.layers.core import Dense, Dropout from keras.optimizers import SGD from deepchem.models import Model Loading @@ -12,7 +13,7 @@ class MultiTaskDNN(Model): Abstract base class for different ML models. """ def __init__(self, task_types, model_params, initialize_raw_model=True): super(MultiTaskDNN, self).__init__(task_types, model_params, super(MultiTaskDNN, self).__init__(model_type, task_types, model_params, initialize_raw_model) if initialize_raw_model: sorted_tasks = sorted(task_types.keys()) Loading Loading @@ -116,6 +117,43 @@ class MultiTaskDNN(Model): y_pred = np.squeeze(y_pred) return y_pred def save(self, out_dir): """ Saves underlying keras model to disk. """ super(MultiTaskDNN, self).save(out_dir) model = self.get_raw_model() filename, _ = os.path.splitext(self.get_model_filename(out_dir)) # Note that keras requires the model architecture and weights to be stored # separately. A json file is generated that specifies the model architecture. # The weights will be stored in an h5 file. The pkl.gz file with store the # target name. json_filename = "%s.%s" % (filename, "json") h5_filename = "%s.%s" % (filename, "h5") # Save architecture json_string = model.to_json() with open(json_filename, "wb") as file_obj: file_obj.write(json_string) model.save_weights(h5_filename, overwrite=True) def load(self, model_dir): """ Load keras multitask DNN from disk. """ super(MultiTaskDNN, self).load(model_dir) filename = self.get_Model_filename(model_dir) filename, _ = os.path.splitext(filename) json_filename = "%s.%s" % (filename, "json") h5_filename = "%s.%s" % (filename, "h5") with open(json_filename) as file_obj: model = model_from_json(file_obj.read()) model.load_weights(h5_filename) self.raw_model = model Model.register_model_type("multitask_deep_regressor", MultiTaskDNN) Model.register_model_type("multitask_deep_classifier", MultiTaskDNN) Loading
deepchem/models/standard.py +54 −7 Original line number Diff line number Diff line Loading @@ -16,24 +16,70 @@ class SklearnModel(Model): Abstract base class for different ML models. """ def __init__(self, task_types, model_params, initialize_raw_model=True): super(SklearnModel, self).__init__(task_types, model_params, initialize_raw_model) self.task_types = task_types self.model_params = model_params self.raw_model = None if initialize_raw_model: if self.modeltype == "rf_regressor": raw_model = RandomForestRegressor( n_estimators=500, n_jobs=-1, warm_start=True, max_features="sqrt") elif self.modeltype == "rf_classifier": raw_model = RandomForestClassifier( n_estimators=500, n_jobs=-1, warm_start=True, max_features="sqrt") elif modeltype == "logistic": raw_model = LogisticRegression(class_weight="auto") elif modeltype == "linear": raw_model = LinearRegression(normalize=True) elif modeltype == "ridge": raw_model = RidgeCV(alphas=[0.01, 0.1, 1.0, 10.0], normalize=True) elif modeltype == "lasso": raw_model = LassoCV(max_iter=2000, n_jobs=-1) elif modeltype == "lasso_lars": raw_model = LassoLarsCV(max_iter=2000, n_jobs=-1) elif modeltype == "elastic_net": raw_model = ElasticNetCV(max_iter=2000, n_jobs=-1) else: raise ValueError("Invalid model type provided.") def fit_on_batch(self, X, y, w): # TODO(rbharath): This is a partial implementation! Does not work for a # datasets with more than one shard. def fit(self, numpy_dataset): """ Updates existing model with new information. Fits SKLearn model to data. """ raise NotImplementedError( "Each model is responsible for its own fit_on_batch method.") for (X, y, _, _) in numpy_dataset.itershards(): self.raw_model.fit(X, y) return def predict_on_batch(self, X): """ Makes predictions on given batch of new data. """ raise NotImplementedError( "Each model is responsible for its own predict_on_batch method.") return self.raw_model.predict(X) def save(self, out_dir): """Saves sklearn model to disk using joblib.""" super(SklearnModel, self).save(out_dir) joblib.dump(self.raw_model, self.get_model_filename(out_dir)) def load(self, model_dir): """Loads sklearn model from joblib file on disk.""" super(SklearnModel, self).load(model_dir) self.raw_model = joblib.load(self.get_model_filename(model_dir) Model.register_model_type("logistic", SklearnModel) Model.register_model_type("rf_classifier", SklearnModel) Model.register_model_type("rf_regressor", SklearnModel) Model.register_model_type("linear", SklearnModel) Model.register_model_type("ridge", SklearnModel) Model.register_model_type("lasso", SklearnModel) Model.register_model_type("lasso_lars", SklearnModel) Model.register_model_type("elastic_net", SklearnModel) # TODO(rbharath): Need to fix singletask dataset support. ''' def fit_singletask_models(train_data, modeltype): """Fits singletask linear regression models to potency. Loading Loading @@ -84,3 +130,4 @@ def fit_singletask_models(train_data, modeltype): model.fit(task_X_train, task_y_train.ravel()) models[task] = model return models '''
