Example to run MolGX with a pretrained model¶

MolGX has been pretrained with a partial subset of the QM9 database consisting of 10 samples. This simple example shows how to generate molecules with the pretrained MolGX under GT4SD. Please check here if you are interested in using the full capability of MolGX.

In [1]:

from gt4sd.algorithms.conditional_generation.molgx.core import MolGX, MolGXQM9Generator

import logging
logging.disable(logging.INFO)

Using TensorFlow backend.

INFO:toxsmi.utils.wrappers:Class weights are (1, 1).
15:41:01   Class weights are (1, 1).
INFO:toxsmi.utils.wrappers:Class weights are (1, 1).
15:41:01   Class weights are (1, 1).
INFO:tape.models.modeling_utils:Better speed can be achieved with apex installed from https://www.github.com/nvidia/apex .
15:41:05   Better speed can be achieved with apex installed from https://www.github.com/nvidia/apex .

In [2]:

configuration = MolGXQM9Generator()
algorithm = MolGX(configuration=configuration)
items = list(algorithm.sample(3))
print(items)

{'target_property': {'homo': (-10, 10), 'lumo': (-10, 10)}, 'use_linear_model': True, 'num_candidate': 2, 'max_candidate': 5, 'max_solution': 10, 'max_node': 50000, 'beam_size': 2000, 'without_estimate': True, 'use_specific_rings': True, 'use_fragment_const': False}
['c1cocn1', 'c1cnoc1', 'c1ccoc1']