Pause before design review on avoiding feature-array dup in singletask.

    train_dict["all"], test_dict["all"] = train_data, test_data

  else:

    raise ValueError("Unsupported mode for process_datasets.")

  print "Shape of Xtrain"

  target = train_dict.itervalues().next()

  print "Shape of Xtrain"

  print np.shape(target[1])

  return train_dict, test_dict 

              entry["feature_types"].append(feature_type)

  return molecules 

#def get_target_names(paths, target_dir_name="targets"):

#  """Get names of targets in provided collections.

#

#  Parameters

#  ----------

#  paths: list 

#    List of paths to base directory.

#  """

#  target_names = []

#  for dataset_path in paths:

#    target_dir = os.path.join(dataset_path, target_dir_name)

#    target_names += [target_pickle.split(".")[0]

#        for target_pickle in os.listdir(target_dir)

#        if "pkl.gz" in target_pickle]

#  return target_names

def load_assays(paths, target_dir_name, target_names):

  """Load regression dataset labels from assays.

    fingerprint, labels = (datapoint["fingerprint"],

      datapoint["labels"])

    tensors.append(np.squeeze(fingerprint))

    #fingerprint.reshape(np.shape(X[id_ind]))

    #X[id_ind] = fingerprint

    sorted_targets = sorted(labels.keys())

    # Set labels from measurements

    for target_ind, target in enumerate(sorted_targets):

        W[id_ind][target_ind] = 0

      else:

        y[id_ind][target_ind] = labels[target]

  X = np.stack(tensors)

  if weight_positives:

    W = balance_positives(y, W)

  print("Done filling X")

  return (sorted_ids, X, y, W)

def multitask_to_singletask(dataset):