Commit b6580cf7 authored by ZHENQIN WU's avatar ZHENQIN WU
Browse files

qm7 add in

parent 7f6cf0e1

examples/benchmark.py

+56 −22
Original line number Diff line number Diff line
@@ -18,8 +18,8 @@ Giving regression performances of: 
    MultitaskDNN(tf_regression),

    Random forest(rf_regression),

    Graph convolution regression(graphconvreg)

on datasets: delaney(ESOL), nci, kaggle, pdbbind, qm7, 

             chembl, sampl(FreeSolv)

on datasets: delaney(ESOL), nci, kaggle, pdbbind, 

             qm7, qm7b, chembl, sampl(FreeSolv)



time estimation listed in README file
@@ -51,7 +51,7 @@ from kaggle.kaggle_datasets import load_kaggle 
from delaney.delaney_datasets import load_delaney

from pdbbind.pdbbind_datasets import load_pdbbind_grid

from chembl.chembl_datasets import load_chembl

from qm7.qm7_datasets import load_qm7

from qm7.qm7_datasets import load_qm7_from_mat, load_qm7b_from_mat

from sampl.sampl_datasets import load_sampl

from clintox.clintox_datasets import load_clintox
@@ -73,7 +73,7 @@ def benchmark_loading_datasets(hyper_parameters, 
  dataset: string, optional (default='tox21')

      choice of which dataset to use, should be: tox21, muv, sider, 

      toxcast, pcba, delaney, kaggle, nci, clintox, pdbbind, chembl,

      qm7, sampl

      qm7, qm7b, sampl

  model: string,  optional (default='tf')

      choice of which model to use, should be: rf, tf, tf_robust, logreg,

      irv, graphconv, tf_regression, rf_regression, graphconvreg
@@ -86,7 +86,7 @@ def benchmark_loading_datasets(hyper_parameters, 
  if dataset in ['muv', 'pcba', 'tox21', 'sider', 'toxcast', 'clintox']:

    mode = 'classification'

  elif dataset in [

      'kaggle', 'delaney', 'nci', 'pdbbind', 'chembl', 'qm7', 'sampl'

      'kaggle', 'delaney', 'nci', 'pdbbind', 'chembl', 'qm7', 'qm7b', 'sampl'

  ]:

    mode = 'regression'

  else:
@@ -121,9 +121,19 @@ def benchmark_loading_datasets(hyper_parameters, 
    if not model in ['tf_regression', 'rf_regression']:

      return



  if dataset in ['qm7']:

    featurizer = None  # qm7 is already featurized

    if split in ['scaffold', 'butina']:

      return  # qm7 accept index and random splitter

    if not model in ['tf_regression']:

      return



  if split in ['year']:

    if not dataset in ['chembl']:

      return

  if split in ['indice', 'stratified']:

    if not dataset in ['qm7', 'qm7b']:

      return

  elif not split in [None, 'index', 'random', 'scaffold', 'butina']:

    raise ValueError('Splitter function not supported')
@@ -138,7 +148,8 @@ def benchmark_loading_datasets(hyper_parameters, 
      'delaney': load_delaney,

      'pdbbind': load_pdbbind_grid,

      'chembl': load_chembl,

      'qm7': load_qm7,

      'qm7': load_qm7_from_mat,

      'qm7b': load_qm7b_from_mat,

      'sampl': load_sampl,

      'clintox': load_clintox

  }
@@ -157,10 +168,16 @@ def benchmark_loading_datasets(hyper_parameters, 
        featurizer=featurizer)



  train_dataset, valid_dataset, test_dataset = all_dataset

  fit_transformers = []

  if dataset in ['qm7', 'qm7b']:

    fit_transformers = [dc.trans.CoulombFitTransformer(train_dataset)]



  time_finish_loading = time.time()

  # time_finish_loading-time_start is the time(s) used for dataset loading

  if dataset in ['kaggle', 'pdbbind', 'qm7']:

  if dataset in ['kaggle', 'pdbbind']:

    n_features = train_dataset.get_data_shape()[0]

  elif dataset in ['qm7', 'qm7b']:

    n_features = list(train_dataset.get_data_shape())

    # dataset has customized features



  # running model
@@ -191,6 +208,7 @@ def benchmark_loading_datasets(hyper_parameters, 
          n_features,

          metric=metric,

          model=model,

          fit_transformers=fit_transformers,

          test=test)

    time_finish_fitting = time.time()
@@ -551,6 +569,7 @@ def benchmark_regression(train_dataset, 
                         n_features,

                         metric='r2',

                         model='tf_regression',

                         fit_transformers=[],

                         seed=123,

                         test=False):

  """
@@ -610,8 +629,23 @@ def benchmark_regression(train_dataset, 
    batch_size = hyper_parameters['batch_size']

    nb_epoch = hyper_parameters['nb_epoch']

    learning_rate = hyper_parameters['learning_rate']



    if len(fit_transformers) > 0:

      model_tf_regression = dc.models.TensorflowMultiTaskFitTransformRegressor(

          n_tasks=len(tasks),

          n_features=n_features,

          layer_sizes=layer_sizes,

          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,

          fit_transformers=fit_transformers,

          n_eval=10,

          seed=seed)

    # Building tensorflow MultiTaskDNN model

    else:

      model_tf_regression = dc.models.TensorflowMultiTaskRegressor(

          len(tasks),

          n_features,
@@ -752,7 +786,7 @@ if __name__ == '__main__': 
      dest='dataset_args',

      default=[],

      help='Choice of dataset: tox21, sider, muv, toxcast, pcba, ' +

      'kaggle, delaney, nci, pdbbind, chembl, sampl, qm7, clintox')

      'kaggle, delaney, nci, pdbbind, chembl, sampl, qm7, qm7b, clintox')

  parser.add_argument(

      '-t',

      action='store_true',
@@ -778,7 +812,7 @@ if __name__ == '__main__': 
  if len(datasets) == 0:

    datasets = [

        'tox21', 'sider', 'muv', 'toxcast', 'pcba', 'clintox', 'sampl',

        'delaney', 'nci', 'kaggle', 'pdbbind', 'chembl', 'qm7'

        'delaney', 'nci', 'kaggle', 'pdbbind', 'chembl', 'qm7b'

    ]



  #input hyperparameters

examples/qm7/qm7_datasets.py

+13 −7
Original line number Diff line number Diff line
@@ -13,9 +13,12 @@ import scipy.io 
import csv



def load_qm7_from_mat(featurizer=None, split='stratified'):

  current_dir = os.path.dirname(os.path.realpath(__file__))

  dataset_file = os.path.join(

      current_dir, "./qm7.mat")

  

  if not os.path.exists('qm7.mat'): os.system('wget http://www.quantum-machine.org/data/qm7.mat')

  dataset = scipy.io.loadmat('qm7.mat')

  if not os.path.exists(dataset_file): os.system('wget -P '+current_dir+' http://www.quantum-machine.org/data/qm7.mat')

  dataset = scipy.io.loadmat(dataset_file)

  

  X = dataset['X']

  y = dataset['T']
@@ -51,13 +54,16 @@ def load_qm7_from_mat(featurizer=None, split='stratified'): 
    valid_dataset = transformer.transform(valid_dataset)

    test_dataset = transformer.transform(test_dataset)



  qm7_tasks = np.arange(y.shape[1])

  qm7_tasks = np.arange(y.shape[0])

  return qm7_tasks, (train_dataset, valid_dataset, test_dataset), transformers



def load_qm7b_from_mat(featurizer=None, split='stratified'):

  current_dir = os.path.dirname(os.path.realpath(__file__))

  dataset_file = os.path.join(

      current_dir, "./qm7b.mat")

  

  if not os.path.exists('qm7b.mat'): os.system('wget http://www.quantum-machine.org/data/qm7b.mat')

  dataset = scipy.io.loadmat('qm7b.mat')

  if not os.path.exists(dataset_file): os.system('wget -P '+current_dir+' http://www.quantum-machine.org/data/qm7b.mat')

  dataset = scipy.io.loadmat(dataset_file)

 

  X = dataset['X']

  y = dataset['T']

CRA Git | Maintained and supported by SUSTech CRA and CCSE