Commit 3dd366c2 authored by Bharath Ramsundar's avatar Bharath Ramsundar
Browse files

Some bugfixes

parent ef843889

deepchem/datasets/__init__.py

+6 −0
Original line number Diff line number Diff line
@@ -637,6 +637,12 @@ def convert_df_to_numpy(df, feature_type, tasks, mol_id_field): 
      if y[ind, task] == "":

        missing[ind, task] = 1

  x_list = list(df[feature_type].values)

  ############################################################## DEBUG

  print("x_list")

  print(x_list)

  print("[type(elt) for elt in x_list]")

  print([type(elt) for elt in x_list])

  ############################################################## DEBUG

  valid_inds = np.array([1 if elt.size > 0 else 0 for elt in x_list], dtype=bool)

  x_list = [elt for (is_valid, elt) in zip(valid_inds, x_list) if is_valid]

  x = np.squeeze(np.array(x_list))

deepchem/featurizers/featurize.py

+8 −1
Original line number Diff line number Diff line
@@ -430,8 +430,15 @@ class DataFeaturizer(object): 
    ############################################################## DEBUG

    df[featurizer.feature_fields] = df[featurizer.feature_fields].apply(pd.to_numeric)

    X_shard = df.as_matrix(columns=featurizer.feature_fields)

    df[featurizer.__class__.__name__] = X_shard.tolist()

    ############################################################## DEBUG

    #df[featurizer.__class__.__name__] = X_shard.tolist()

    df[featurizer.__class__.__name__] = [np.array(elt) for elt in X_shard.tolist()]

    ############################################################## DEBUG

    print("X_shard")

    print(X_shard)

    print("type(X_shard)")

    print(type(X_shard))

    print("[type(elt) for elt in X_shard.tolist()]")

    print([type(elt) for elt in X_shard.tolist()])

    #for ind, row in df.iterrows():

    #  # pandas rows are tuples (row_num, row_data)

    #  feature_list = []

deepchem/models/tests/test_api.py

+58 −54
Original line number Diff line number Diff line
@@ -95,7 +95,7 @@ class TestModelAPI(TestAPI): 
                                smiles_field=self.smiles_field,

                                featurizer=featurizer,

                                verbosity="low")

    dataset = featurizer.featurize(input_file, self.data_dir)

    dataset = featurizer.featurize(input_file, self.data_dir, debug=True)



    splitter = SpecifiedSplitter(input_file, "split")

    train_dataset, test_dataset = splitter.train_test_split(
@@ -230,60 +230,64 @@ class TestModelAPI(TestAPI): 
    evaluator = Evaluator(model, test_dataset, transformers, verbosity=True)

    _ = evaluator.compute_model_performance(regression_metrics)



  def test_singletask_keras_mlp_USF_regression_API(self):

    """Test of singletask MLP User Specified Features regression API."""

    from deepchem.models.keras_models.fcnet import SingleTaskDNN

    featurizer = UserDefinedFeaturizer(["evals"])

    tasks = ["u0"]

    task_type = "regression"

    task_types = {task: task_type for task in tasks}

    model_params = {"nb_hidden": 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}



    input_file = os.path.join(self.current_dir, "gbd3k.pkl.gz")

    featurizer = DataFeaturizer(tasks=tasks,

                                smiles_field=self.smiles_field,

                                featurizer=featurizer,

                                verbosity="low")

    dataset = featurizer.featurize(input_file, self.data_dir)



    splitter = ScaffoldSplitter()

    train_dataset, test_dataset = splitter.train_test_split(

        dataset, self.train_dir, self.test_dir)



    input_transformers = [

      NormalizationTransformer(transform_X=True, dataset=train_dataset),

      ClippingTransformer(transform_X=True, dataset=train_dataset)]

    output_transformers = [

      NormalizationTransformer(transform_y=True, dataset=train_dataset)]

    transformers = input_transformers + output_transformers



    for dataset in [train_dataset, test_dataset]:

      for transformer in transformers:

        transformer.transform(dataset)



    model_params["data_shape"] = train_dataset.get_data_shape()

    regression_metrics = [Metric(metrics.r2_score),

                          Metric(metrics.mean_squared_error),

                          Metric(metrics.mean_absolute_error)]



    model = SingleTaskDNN(tasks, task_types, model_params, self.model_dir)



    # Fit trained model

    model.fit(train_dataset)

    model.save()



    # Eval model on train

    evaluator = Evaluator(model, train_dataset, transformers, verbosity=True)

    _ = evaluator.compute_model_performance(regression_metrics)



    # Eval model on test

    evaluator = Evaluator(model, test_dataset, transformers, verbosity=True)

    _ = evaluator.compute_model_performance(regression_metrics)

  #### TODO(rbharath): This test is being disabled since deepchem no longer

  #### accepts this format of input. Decide whether this test should be deleted

  #### altogether or replaced.

  #def test_singletask_keras_mlp_USF_regression_API(self):

  #  """Test of singletask MLP User Specified Features regression API."""

  #  from deepchem.models.keras_models.fcnet import SingleTaskDNN

  #  featurizer = UserDefinedFeaturizer(["evals"])

  #  tasks = ["u0"]

  #  task_type = "regression"

  #  task_types = {task: task_type for task in tasks}

  #  model_params = {"nb_hidden": 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}



  #  input_file = os.path.join(self.current_dir, "gbd3k.pkl.gz")

  #  featurizer = DataFeaturizer(tasks=tasks,

  #                              smiles_field=self.smiles_field,

  #                              featurizer=featurizer,

  #                              verbosity="low")

  #  dataset = featurizer.featurize(input_file, self.data_dir)



  #  splitter = ScaffoldSplitter()

  #  train_dataset, test_dataset = splitter.train_test_split(

  #      dataset, self.train_dir, self.test_dir)



  #  input_transformers = [

  #    NormalizationTransformer(transform_X=True, dataset=train_dataset),

  #    ClippingTransformer(transform_X=True, dataset=train_dataset)]

  #  output_transformers = [

  #    NormalizationTransformer(transform_y=True, dataset=train_dataset)]

  #  transformers = input_transformers + output_transformers



  #  for dataset in [train_dataset, test_dataset]:

  #    for transformer in transformers:

  #      transformer.transform(dataset)



  #  model_params["data_shape"] = train_dataset.get_data_shape()

  #  regression_metrics = [Metric(metrics.r2_score),

  #                        Metric(metrics.mean_squared_error),

  #                        Metric(metrics.mean_absolute_error)]



  #  model = SingleTaskDNN(tasks, task_types, model_params, self.model_dir)



  #  # Fit trained model

  #  model.fit(train_dataset)

  #  model.save()



  #  # Eval model on train

  #  evaluator = Evaluator(model, train_dataset, transformers, verbosity=True)

  #  _ = evaluator.compute_model_performance(regression_metrics)



  #  # Eval model on test

  #  evaluator = Evaluator(model, test_dataset, transformers, verbosity=True)

  #  _ = evaluator.compute_model_performance(regression_metrics)





  def test_multitask_keras_mlp_ECFP_classification_API(self):

CRA Git | Maintained and supported by SUSTech CRA and CCSE