fix for the seed

        weight_init_stddevs=weight_init_stddevs,

        bias_init_consts=bias_init_consts, dropouts=dropouts, penalty=penalty, 

        penalty_type=penalty_type, batch_size=batch_size, 

        learning_rate=learning_rate, verbosity=verbosity)

        learning_rate=learning_rate, verbosity=verbosity, seed=seed)

    print('-------------------------------------')

    print('Start fitting by multitask DNN')

        bypass_bias_init_consts=bypass_bias_init_consts,

        bypass_dropouts=bypass_dropouts, penalty=penalty, 

        penalty_type=penalty_type, batch_size=batch_size,

        learning_rate=learning_rate, verbosity=verbosity)

        learning_rate=learning_rate, verbosity=verbosity, seed=seed)

    print('--------------------------------------------')

    print('Start fitting by robust multitask DNN')

    model_logreg = dc.models.TensorflowLogisticRegression(len(tasks),

        n_features, penalty=penalty, penalty_type=penalty_type, 

        batch_size=batch_size, learning_rate=learning_rate, 

        verbosity=verbosity)

        verbosity=verbosity, seed=seed)

    print('-------------------------------------')

    print('Start fitting by logistic regression')

    learning_rate = hyper_parameters['learning_rate']

    n_filters = hyper_parameters['n_filters']

    n_fully_connected_nodes = hyper_parameters['n_fully_connected_nodes']

    seed = hyper_parameters['seed']

    g = tf.Graph()

    sess = tf.Session(graph=g)

  os.makedirs(base_dir_o)

  #Datasets and models used in the benchmark test, all=all the datasets

  dataset_name = 'muv'

  dataset_name = 'tox21'

  model = 'tf'

  #input hyperparameters

  hps['graphconv'] = [{'batch_size': 50, 'nb_epoch': 10, 

                       'learning_rate': 0.001, 'n_filters': 64, 

                       'n_fully_connected_nodes': 128, 'seed': None}]

                       'n_fully_connected_nodes': 128}]

  hps['rf'] = [{'n_estimators': 500}]