gt4sd.algorithms.conditional_generation.guacamol.implementation.moses_vae module

Moses VAE implementation.

Summary

Classes:

VAE

Reference

class VAE(model_path, model_config_path, vocab_path, n_samples, n_batch, max_len, device='cpu')[source]

Bases: object

__init__(model_path, model_config_path, vocab_path, n_samples, n_batch, max_len, device='cpu')[source]

Initialize VAE.

Parameters

  • model_path (str) – path from where to load the model.

  • model_config_path (str) – path from where to load the model config.

  • vocab_path (str) – path from where to load the vocab.

  • n_samples (int) – number of samples to sample.

  • n_batch (int) – size of the batch.

  • max_len (int) – max length of SMILES.

  • device (str) – device used for computation. Defaults to cpu.

get_generator()[source]

Create an instance of the VaeGenerator.

Return type

VaeGenerator

Returns

an instance of VaeGenerator.

__dict__ = mappingproxy({'__module__': 'gt4sd.algorithms.conditional_generation.guacamol.implementation.moses_vae', '__init__': <function VAE.__init__>, 'get_generator': <function VAE.get_generator>, '__dict__': <attribute '__dict__' of 'VAE' objects>, '__weakref__': <attribute '__weakref__' of 'VAE' objects>, '__doc__': None, '__annotations__': {}})
__doc__ = None
__module__ = 'gt4sd.algorithms.conditional_generation.guacamol.implementation.moses_vae'
__weakref__

list of weak references to the object (if defined)