unit test for molnet

from __future__ import print_function

from __future__ import division

from __future__ import unicode_literals

from deepchem.molnet.load_function.chembl_datasets import load_chembl

from deepchem.molnet.load_function.clintox_datasets import load_clintox

from deepchem.molnet.load_function.delaney_datasets import load_delaney

"""

Tests for molnet function 

"""

from __future__ import print_function

from __future__ import division

from __future__ import unicode_literals

import os

import unittest

import numpy as np

import pandas as pd

import deepchem as dc

import csv

class TestMolnet(unittest.TestCase):

  """

  Test basic function of molnet

  """

  def setUp(self):

    super(TestMolnet, self).setUp()

    self.current_dir = os.path.dirname(os.path.abspath(__file__))

  def test_delaney_graphconvreg(self):

    """Tests molnet benchmarking on delaney with graphconvreg."""

    datasets = ['delaney']

    model = 'graphconvreg'

    split = 'random'

    out_path = self.current_dir

    dc.molnet.run_benchmark(datasets, model, split=split, out_path=out_path)

    with open(os.path.join(out_path, 'results.csv'), 'r') as f:

      reader = csv.reader(f)

      for lastrow in reader:

        pass

      assert lastrow[-4] == model

      assert lastrow[-5] == 'valid'

      assert lastrow[-3] > 0.75

  def test_qm7_multitask(self):

    """Tests molnet benchmarking on qm7 with multitask network."""

    datasets = ['qm7']

    model = 'tf_regression'

    split = 'random'

    out_path = self.current_dir

    dc.molnet.run_benchmark(datasets, model, split=split, out_path=out_path)

    with open(os.path.join(out_path, 'results.csv'), 'r') as f:

      reader = csv.reader(f)

      for lastrow in reader:

        pass

      assert lastrow[-4] == model + '_ft'

      assert lastrow[-5] == 'valid'

      assert lastrow[-3] > 0.95

  def test_tox21_multitask(self):

    """Tests molnet benchmarking on tox21 with multitask network."""

    datasets = ['tox21']

    model = 'tf'

    split = 'random'

    out_path = self.current_dir

    dc.molnet.run_benchmark(datasets, model, split=split, out_path=out_path)

    with open(os.path.join(out_path, 'results.csv'), 'r') as f:

      reader = csv.reader(f)

      for lastrow in reader:

        pass

      assert lastrow[-4] == model

      assert lastrow[-5] == 'valid'

      assert lastrow[-3] > 0.75