Commit 6c4008c8 authored by Ubuntu's avatar Ubuntu
Browse files

run yapf

parent 8a672422

deepchem/models/tests/test_generalize.py

+23 −22
Original line number Diff line number Diff line
@@ -21,6 +21,7 @@ from sklearn.linear_model import LinearRegression 
from sklearn.linear_model import LogisticRegression

import xgboost





class TestGeneralize(unittest.TestCase):

  """

  Test that models can learn generalizable models on simple datasets.
@@ -71,10 +72,10 @@ class TestGeneralize(unittest.TestCase): 
    transformers = [

        dc.trans.NormalizationTransformer(

            transform_X=True, dataset=train_dataset),

        dc.trans.ClippingTransformer(

            transform_X=True, dataset=train_dataset),

        dc.trans.ClippingTransformer(transform_X=True, dataset=train_dataset),

        dc.trans.NormalizationTransformer(

            transform_y=True, dataset=train_dataset)]

            transform_y=True, dataset=train_dataset)

    ]

    for data in [train_dataset, test_dataset]:

      for transformer in transformers:

        data = transformer.transform(data)
@@ -115,9 +116,11 @@ class TestGeneralize(unittest.TestCase): 
    test_dataset = dc.data.DiskDataset.from_numpy(X_test, y_test)



    regression_metric = dc.metrics.Metric(dc.metrics.r2_score)



    def model_builder(model_dir):

      sklearn_model = LinearRegression()

      return dc.models.SklearnModel(sklearn_model, model_dir)



    model = dc.models.SingletaskToMultitask(tasks, model_builder)



    # Fit trained model
@@ -234,12 +237,10 @@ class TestGeneralize(unittest.TestCase): 


    regression_metric = dc.metrics.Metric(dc.metrics.mae_score)

    esr = {'early_stopping_rounds': 50}



    def model_builder(model_dir):

      xgb_model = xgboost.XGBRegressor(n_estimators=50, seed=123)

      return dc.models.XGBoostModel(xgb_model,

                                    model_dir,

                                    verbose=False,

                                    **esr)

      return dc.models.XGBoostModel(xgb_model, model_dir, verbose=False, **esr)



    model = dc.models.SingletaskToMultitask(tasks, model_builder)

deepchem/models/xgboost_models/__init__.py

+51 −48
Original line number Diff line number Diff line
@@ -13,12 +13,17 @@ from sklearn.cross_validation import train_test_split 
from sklearn.grid_search import GridSearchCV

import tempfile





class XGBoostModel(SklearnModel):

  """

  Abstract base class for XGBoost model.

  """

  def __init__(self, model_instance=None, model_dir=None,

               verbose=True, **kwargs):



  def __init__(self,

               model_instance=None,

               model_dir=None,

               verbose=True,

               **kwargs):

    """Abstract class for XGBoost models.

    Parameters:

    -----------
@@ -42,7 +47,6 @@ class XGBoostModel(SklearnModel): 
    else:

      self.early_stopping_rounds = 50





  def fit(self, dataset, **kwargs):

    """

    Fits XGBoost model to data.
@@ -65,12 +69,12 @@ class XGBoostModel(SklearnModel): 


    # Find optimal n_estimators based on original learning_rate

    # and early_stopping_rounds

    X_train, X_test, y_train, y_test = train_test_split(X, y,

                                                        test_size = 0.2,

                                                        random_state=seed,

                                                        stratify=stratify)

    X_train, X_test, y_train, y_test = train_test_split(

        X, y, test_size=0.2, random_state=seed, stratify=stratify)



    self.model_instance.fit(X_train, y_train,

    self.model_instance.fit(

        X_train,

        y_train,

        early_stopping_rounds=self.early_stopping_rounds,

        eval_metric=xgb_metric,

        eval_set=[(X_train, y_train), (X_test, y_test)],
@@ -79,8 +83,8 @@ class XGBoostModel(SklearnModel): 
    # n_estimator increased to 1/0.8 = 1.25 time.

    estimated_best_round = np.round(self.model_instance.best_ntree_limit * 1.25)

    self.model_instance.n_estimators = np.int64(estimated_best_round)

    self.model_instance.fit(X_train, y_train, eval_metric=xgb_metric,

			    verbose=self.verbose)

    self.model_instance.fit(

        X_train, y_train, eval_metric=xgb_metric, verbose=self.verbose)



  def _search_param(self, metric, X, y):

    '''
@@ -88,10 +92,9 @@ class XGBoostModel(SklearnModel): 
    '''

    # Make sure user specified params are in the grid.

    max_depth_grid = list(np.unique([self.model_instance.max_depth, 5, 7]))

    colsample_bytree_grid = list(np.unique(

                                [self.model_instance.colsample_bytree,0.66,0.9]))

    reg_lambda_grid = list(np.unique(

                                [self.model_instance.reg_lambda,1,5]))

    colsample_bytree_grid = list(

        np.unique([self.model_instance.colsample_bytree, 0.66, 0.9]))

    reg_lambda_grid = list(np.unique([self.model_instance.reg_lambda, 1, 5]))

    param_grid = {

        'max_depth': max_depth_grid,

        'learning_rate': [max(self.model_instance.learning_rate, 0.3)],
@@ -108,8 +111,8 @@ class XGBoostModel(SklearnModel): 
        'base_score': [self.model_instance.base_score],

        'seed': [self.model_instance.seed]

    }

    grid_search = GridSearchCV(self.model_instance, param_grid, cv=2,

                                refit=False, scoring=metric)

    grid_search = GridSearchCV(

        self.model_instance, param_grid, cv=2, refit=False, scoring=metric)

    grid_search.fit(X, y)

    best_params = grid_search.best_params_

    # Change params back original params

examples/benchmark.py

+58 −49
Original line number Diff line number Diff line
@@ -58,6 +58,7 @@ from clintox.clintox_datasets import load_clintox 
from hiv.hiv_datasets import load_hiv

import xgboost





def benchmark_loading_datasets(hyper_parameters,

                               dataset='tox21',

                               model='tf',
@@ -313,8 +314,9 @@ def benchmark_classification(train_dataset, 
  if metric == 'auc':

    classification_metric = dc.metrics.Metric(dc.metrics.roc_auc_score, np.mean)



  assert model in ['rf', 'tf', 'tf_robust', 'logreg', 'irv', 'graphconv',

                   'xgb_classifier']

  assert model in [

      'rf', 'tf', 'tf_robust', 'logreg', 'irv', 'graphconv', 'xgb_classifier'

  ]



  if model == 'tf':

    # Loading hyper parameters
@@ -584,6 +586,7 @@ def benchmark_classification(train_dataset, 
    early_stopping_rounds = hyper_parameters['early_stopping_rounds']



    esr = {'early_stopping_rounds': early_stopping_rounds}



    # Building xgboost classification model

    def model_builder(model_dir_xgb):

      xgboost_model = xgboost.XGBClassifier(
@@ -601,9 +604,9 @@ def benchmark_classification(train_dataset, 
          scale_pos_weight=scale_pos_weight,

          base_score=base_score,

          seed=seed)

        return dc.models.xgboost_models.XGBoostModel(xgboost_model,

                                                            model_dir_xgb,

      return dc.models.xgboost_models.XGBoostModel(xgboost_model, model_dir_xgb,

                                                   **esr)



    model_xgb = dc.models.multitask.SingletaskToMultitask(tasks, model_builder)



    print('-------------------------------------')
@@ -680,8 +683,9 @@ def benchmark_regression(train_dataset, 
    regression_metric = dc.metrics.Metric(dc.metrics.mean_absolute_error,

                                          np.mean)



  assert model in ['tf_regression', 'rf_regression', 'graphconvreg',

                   'xgb_regression']

  assert model in [

      'tf_regression', 'rf_regression', 'graphconvreg', 'xgb_regression'

  ]



  if model == 'tf_regression':

    # Loading hyper parameters
@@ -843,6 +847,7 @@ def benchmark_regression(train_dataset, 
    early_stopping_rounds = hyper_parameters['early_stopping_rounds']



    esr = {'early_stopping_rounds': early_stopping_rounds}



    # Building xgboost classification model

    def model_builder(model_dir_xgb):

      xgboost_model = xgboost.XGBRegressor(
@@ -860,9 +865,9 @@ def benchmark_regression(train_dataset, 
          scale_pos_weight=scale_pos_weight,

          base_score=base_score,

          seed=seed)

        return dc.models.xgboost_models.XGBoostModel(xgboost_model,

                                                            model_dir_xgb,

      return dc.models.xgboost_models.XGBoostModel(xgboost_model, model_dir_xgb,

                                                   **esr)



    model_xgb = dc.models.multitask.SingletaskToMultitask(tasks, model_builder)



    print('-------------------------------------')
@@ -902,7 +907,8 @@ if __name__ == '__main__': 
      dest='model_args',

      default=[],

      help='Choice of model: tf, tf_robust, logreg, rf, irv, graphconv, ' +

      'tf_regression, rf_regression, graphconvreg, xgb_classifier, xgb_regression')

      'tf_regression, rf_regression, graphconvreg, xgb_classifier, xgb_regression'

  )

  parser.add_argument(

      '-d',

      action='append',
@@ -1062,8 +1068,10 @@ if __name__ == '__main__': 
          'tox21', 'sider', 'muv', 'toxcast', 'pcba', 'clintox', 'hiv'

      ]:

        for model in models:

          if model in ['tf', 'tf_robust', 'logreg', 'graphconv', 'rf', 'irv',

                        'xgb_classifier']:

          if model in [

              'tf', 'tf_robust', 'logreg', 'graphconv', 'rf', 'irv',

              'xgb_classifier'

          ]:

            benchmark_loading_datasets(

                hps,

                dataset=dataset,
@@ -1073,8 +1081,9 @@ if __name__ == '__main__': 
                test=test)

      else:

        for model in models:

          if model in ['tf_regression', 'rf_regression', 'graphconvreg',

                        'xgb_regression']:

          if model in [

              'tf_regression', 'rf_regression', 'graphconvreg', 'xgb_regression'

          ]:

            benchmark_loading_datasets(

                hps,

                dataset=dataset,

CRA Git | Maintained and supported by SUSTech CRA and CCSE