def test_untokenise_smiles_with_simplify_rings(self, smiles_str, split,
tokens):
token_list = ['?', 'C', '1', '2', '%12', '%']
tokeniser = smiles.SmilesTokeniser(token_list,
splitting_method=split,
simplify_rings=True)
smiles_result = tokeniser.untokenise_smiles(tokens)
assert smiles_result == smiles_str
def test_tokenise_smiles_all_tokens(self, smiles_str, tokens):
token_list = [
'?', 'C', 'N', 'O', 'Cl', 'Br', '[Pt]', '(', ')', '=', '1', '2',
'3', '%10'
]
tokeniser = smiles.SmilesTokeniser(token_list)
tokens_result = tokeniser.tokenise_smiles(smiles_str)
assert tokens_result == tokens
def test_tokenise_smiles_with_sos_eos(self, smiles_str, token_sos,
token_eos, tokens):
token_list = ['C', 'N', 'O', 'Cl', 'Br', '<SOS>', '<EOS>']
tokeniser = smiles.SmilesTokeniser(token_list,
token_sos=token_sos,
token_eos=token_eos)
tokens_result = tokeniser.tokenise_smiles(smiles_str)
assert tokens_result == tokens
def test_tokenise_smiles_with_simplify_rings(self, smiles_str, split,
tokens):
token_list = ['?', 'C', '1', '2', '3']
tokeniser = smiles.SmilesTokeniser(token_list,
splitting_method=split,
simplify_rings=True)
tokens_result = tokeniser.tokenise_smiles(smiles_str)
assert tokens_result == tokens
def test_tokenise_smiles_with_padding(self, smiles_str, token_pad, length,
truncate, tokens):
token_list = ['C', 'N', 'O', 'Cl', 'Br', '_', ' ']
tokeniser = smiles.SmilesTokeniser(token_list,
token_padding=token_pad,
sequence_length=length,
truncate_sequence=truncate)
tokens_result = tokeniser.tokenise_smiles(smiles_str)
assert tokens_result == tokens
def test_tokenise_smiles_with_unknown_placeholder(self, smiles_str, split,
placeholder, tokens):
token_list = ['C', 'Br', 'B', 'r', '?', '[Pt]', '[', 'P', 't', ']']
tokeniser = smiles.SmilesTokeniser(token_list,
splitting_method=split,
token_unknown=placeholder)
tokens_result = tokeniser.tokenise_smiles(smiles_str)
assert tokens_result == tokens
assert tokeniser.missing_tokens == {'S', 'N'}
def test_tokenise_smiles_characters(self, smiles_str, tokens):
token_list = [
'?', 'C', 'N', 'O', 'l', 'B', '[', 'P', 't', ']', '(', ')', '=',
'1', '2', '3', '%', '0', 'r'
]
tokeniser = smiles.SmilesTokeniser(token_list,
splitting_method='characters')
tokens_result = tokeniser.tokenise_smiles(smiles_str)
assert tokens_result == tokens
def test_tokenise_smiles_halogens_only(self, smiles_str, tokens):
token_list = [
'?', 'C', 'N', 'O', 'Cl', 'Br', '[', 'P', 't', ']', '(', ')', '=',
'1', '2', '3', '%', '0'
]
tokeniser = smiles.SmilesTokeniser(token_list,
splitting_method='halogens_only')
tokens_result = tokeniser.tokenise_smiles(smiles_str)
assert tokens_result == tokens
def test_tokenise_smiles_with_pad_sos_eos(self, smiles_str, token_sos,
token_eos, tokens):
token_list = ['C', 'N', 'O', 'Cl', 'Br', '<SOS>', '<EOS>', '_']
tokeniser = smiles.SmilesTokeniser(token_list,
token_sos=token_sos,
token_eos=token_eos,
token_padding='_',
sequence_length=7)
tokens_result = tokeniser.tokenise_smiles(smiles_str)
assert tokens_result == tokens
def test_exception_if_seq_length_not_padding(self):
token_list = ['?', 'C', '1', '2', '%12', '%']
with pytest.raises(ValueError):
_ = smiles.SmilesTokeniser(token_list, sequence_length=50)
def test_exception_if_padding_not_seq_length(self):
token_list = ['?', 'C', '1', '2', '%12', '%']
with pytest.raises(ValueError):
_ = smiles.SmilesTokeniser(token_list, token_padding='?')