print("Mean Tmelt in training data: ", train_mean)
print("Standard deviation of Tmelt in training data: ", train_std)
print("Min value of Tmelt in training data: ", np.min(y_train))
print("Max value of Tmelt in training data: ", np.max(y_train))
y_train = (y_train - train_mean) / train_std
y_test = (y_test - train_mean) / train_std
def rmse_tmelt(target, predicted, std=train_std):
mse = mean_squared_error(target, predicted)
rmse = np.sqrt(mse) * std
return rmse
from openchem.data.utils import save_smiles_property_file
save_smiles_property_file('./benchmark_datasets/melt_temp/train.smi', X_train,
y_train.reshape(-1, 1))
save_smiles_property_file('./benchmark_datasets/melt_temp/test.smi', X_test,
y_test.reshape(-1, 1))
from openchem.data.smiles_data_layer import SmilesDataset
train_dataset = SmilesDataset('./benchmark_datasets/melt_temp/train.smi',
delimiter=',',
cols_to_read=[0, 1],
tokens=tokens,
augment=True)
test_dataset = SmilesDataset('./benchmark_datasets/melt_temp/test.smi',
delimiter=',',
cols_to_read=[0, 1],
tokens=tokens)
from openchem.data.utils import get_tokens
tokens, _, _ = get_tokens(reactants)
from sklearn.model_selection import train_test_split
X_train, X_test, y_train, y_test = train_test_split(reactants,
labels,
test_size=0.2,
random_state=42)
y_mean = np.mean(y_train)
y_std = np.std(y_train)
y_train = (y_train - y_mean) / y_std
y_test = (y_test - y_mean) / y_std
from openchem.data.utils import save_smiles_property_file
save_smiles_property_file('./benchmark_datasets/reactions/train.smi',
X_train,
y_train,
delimiter=" ")
save_smiles_property_file('./benchmark_datasets/reactions/test.smi',
X_test,
y_test,
delimiter=" ")
from openchem.data.smiles_data_layer import SmilesDataset
head1_arguments = {"tokens": tokens, "delimiter": " ", "sanitize": False}
head2_arguments = {"tokens": tokens, "delimiter": " ", "sanitize": False}
train_dataset = SiameseDataset('./benchmark_datasets/reactions/train.smi',
head1_type='smiles',
head2_type='smiles',
cols_to_read=[0, 1, 2],
head1_arguments=head1_arguments,
head2_arguments=head2_arguments)
smiles = data[0]
labels = np.array(data[1:])
labels[np.where(labels=='')] = '999'
labels = labels.T
from openchem.data.utils import get_tokens
tokens, _, _ = get_tokens(smiles)
tokens = tokens + ' '
from sklearn.model_selection import train_test_split
X_train, X_test, y_train, y_test = train_test_split(smiles, labels, test_size=0.2,
random_state=42)
from openchem.data.utils import save_smiles_property_file
save_smiles_property_file('./benchmark_datasets/tox21/train.smi', X_train, y_train)
save_smiles_property_file('./benchmark_datasets/tox21/test.smi', X_test, y_test)
from openchem.data.smiles_data_layer import SmilesDataset
train_dataset = SmilesDataset('./benchmark_datasets/tox21/train.smi',
delimiter=',', cols_to_read=list(range(13)),
tokens=tokens, augment=True)
test_dataset = SmilesDataset('./benchmark_datasets/tox21/test.smi',
delimiter=',', cols_to_read=list(range(13)),
tokens=tokens)
def multitask_auc(ground_truth, predicted):
from sklearn.metrics import roc_auc_score
import numpy as np
import torch
ground_truth = np.array(ground_truth)