def average_duplicates(args: Args):
"""Averages duplicate data points in a dataset."""
print('Loading data')
header = get_header(args.data_path)
data = get_data(path=args.data_path,
smiles_columns=args.smiles_columns,
target_columns=args.target_columns)
print(f'Data size = {len(data):,}')
# Map SMILES string to lists of targets
smiles_in_order = []
smiles_to_targets = defaultdict(list)
for smiles, targets in zip(data.smiles(flatten=True), data.targets()):
smiles_to_targets[smiles].append(targets)
if len(smiles_to_targets[smiles]) == 1:
smiles_in_order.append(smiles)
# Find duplicates
duplicate_count = 0
stds = []
new_data = []
for smiles in smiles_in_order:
all_targets = smiles_to_targets[smiles]
duplicate_count += len(all_targets) - 1
num_tasks = len(all_targets[0])
targets_by_task = [[] for _ in range(num_tasks)]
for task in range(num_tasks):
for targets in all_targets:
if targets[task] is not None:
targets_by_task[task].append(targets[task])
stds.append([
np.std(task_targets) if len(task_targets) > 0 else 0.0
for task_targets in targets_by_task
])
means = [
np.mean(task_targets) if len(task_targets) > 0 else None
for task_targets in targets_by_task
]
new_data.append((smiles, means))
print(f'Number of duplicates = {duplicate_count:,}')
print(
f'Duplicate standard deviation per task = {", ".join(f":{std:.4e}" for std in np.mean(stds, axis=0))}'
)
print(f'New data size = {len(new_data):,}')
# Save new data
with open(args.save_path, 'w') as f:
f.write(','.join(header) + '\n')
for smiles, avg_targets in new_data:
f.write(smiles + ',' + ','.join(
str(value) if value is not None else ''
for value in avg_targets) + '\n')
def interpret(args: InterpretArgs) -> None:
"""
Runs interpretation of a Chemprop model using the Monte Carlo Tree Search algorithm.
:param args: A :class:`~chemprop.args.InterpretArgs` object containing arguments for interpretation.
"""
if args.number_of_molecules != 1:
raise ValueError(
"Interpreting is currently only available for single-molecule models."
)
global C_PUCT, MIN_ATOMS
chemprop_model = ChempropModel(args)
def scoring_function(smiles: List[str]) -> List[float]:
return chemprop_model(smiles)[:, args.property_id - 1]
C_PUCT = args.c_puct
MIN_ATOMS = args.min_atoms
all_smiles = get_smiles(path=args.data_path,
smiles_columns=args.smiles_columns)
header = get_header(path=args.data_path)
property_name = header[
args.property_id] if len(header) > args.property_id else 'score'
print(f'smiles,{property_name},rationale,rationale_score')
for smiles in all_smiles:
score = scoring_function([smiles])[0]
if score > args.prop_delta:
rationales = mcts(smiles=smiles[0],
scoring_function=scoring_function,
n_rollout=args.rollout,
max_atoms=args.max_atoms,
prop_delta=args.prop_delta)
else:
rationales = []
if len(rationales) == 0:
print(f'{smiles},{score:.3f},,')
else:
min_size = min(len(x.atoms) for x in rationales)
min_rationales = [
x for x in rationales if len(x.atoms) == min_size
]
rats = sorted(min_rationales, key=lambda x: x.P, reverse=True)
print(f'{smiles},{score:.3f},{rats[0].smiles},{rats[0].P:.3f}')
def make_predictions(
args: PredictArgs,
smiles: List[List[str]] = None) -> List[List[Optional[float]]]:
"""
Loads data and a trained model and uses the model to make predictions on the data.
If SMILES are provided, then makes predictions on smiles.
Otherwise makes predictions on :code:`args.test_data`.
:param args: A :class:`~chemprop.args.PredictArgs` object containing arguments for
loading data and a model and making predictions.
:param smiles: List of list of SMILES to make predictions on.
:return: A list of lists of target predictions.
"""
print('Loading training args')
train_args = load_args(args.checkpoint_paths[0])
num_tasks, task_names = train_args.num_tasks, train_args.task_names
# If features were used during training, they must be used when predicting
if ((train_args.features_path is not None
or train_args.features_generator is not None)
and args.features_path is None
and args.features_generator is None):
raise ValueError(
'Features were used during training so they must be specified again during prediction '
'using the same type of features as before (with either --features_generator or '
'--features_path and using --no_features_scaling if applicable).')
# If atom-descriptors were used during training, they must be used when predicting and vice-versa
if train_args.atom_descriptors != args.atom_descriptors:
raise ValueError(
'The use of atom descriptors is inconsistent between training and prediction. If atom descriptors '
' were used during training, they must be specified again during prediction using the same type of '
' descriptors as before. If they were not used during training, they cannot be specified during prediction.'
)
# Update predict args with training arguments to create a merged args object
for key, value in vars(train_args).items():
if not hasattr(args, key):
setattr(args, key, value)
args: Union[PredictArgs, TrainArgs]
if args.atom_descriptors == 'feature':
set_extra_atom_fdim(train_args.atom_features_size)
print('Loading data')
if smiles is not None:
full_data = get_data_from_smiles(
smiles=smiles,
skip_invalid_smiles=False,
features_generator=args.features_generator)
else:
full_data = get_data(path=args.test_path,
target_columns=[],
ignore_columns=[],
skip_invalid_smiles=False,
args=args,
store_row=not args.drop_extra_columns)
print('Validating SMILES')
full_to_valid_indices = {}
valid_index = 0
for full_index in range(len(full_data)):
if all(mol is not None for mol in full_data[full_index].mol):
full_to_valid_indices[full_index] = valid_index
valid_index += 1
test_data = MoleculeDataset(
[full_data[i] for i in sorted(full_to_valid_indices.keys())])
# Edge case if empty list of smiles is provided
if len(test_data) == 0:
return [None] * len(full_data)
print(f'Test size = {len(test_data):,}')
# Predict with each model individually and sum predictions
if args.dataset_type == 'multiclass':
sum_preds = np.zeros(
(len(test_data), num_tasks, args.multiclass_num_classes))
else:
sum_preds = np.zeros((len(test_data), num_tasks))
# Create data loader
test_data_loader = MoleculeDataLoader(dataset=test_data,
batch_size=args.batch_size,
num_workers=args.num_workers)
print(
f'Predicting with an ensemble of {len(args.checkpoint_paths)} models')
for checkpoint_path in tqdm(args.checkpoint_paths,
total=len(args.checkpoint_paths)):
# Load model and scalers
model = load_checkpoint(checkpoint_path, device=args.device)
scaler, features_scaler = load_scalers(checkpoint_path)
# Normalize features
if args.features_scaling:
test_data.reset_features_and_targets()
test_data.normalize_features(features_scaler)
# Make predictions
model_preds = predict(model=model,
data_loader=test_data_loader,
scaler=scaler)
sum_preds += np.array(model_preds)
# Ensemble predictions
avg_preds = sum_preds / len(args.checkpoint_paths)
avg_preds = avg_preds.tolist()
# Save predictions
#print(f'Saving predictions to {args.preds_path}')
assert len(test_data) == len(avg_preds)
makedirs(args.preds_path, isfile=True)
# Get prediction column names
if args.dataset_type == 'multiclass':
task_names = [
f'{name}_class_{i}' for name in task_names
for i in range(args.multiclass_num_classes)
]
else:
task_names = task_names
# Copy predictions over to full_data
for full_index, datapoint in enumerate(full_data):
valid_index = full_to_valid_indices.get(full_index, None)
preds = avg_preds[valid_index] if valid_index is not None else [
'Invalid SMILES'
] * len(task_names)
# If extra columns have been dropped, add back in SMILES columns
if args.drop_extra_columns:
datapoint.row = OrderedDict()
smiles_columns = args.smiles_columns
if None in smiles_columns:
smiles_columns = get_header(
args.test_path)[:len(smiles_columns)]
for column, smiles in zip(smiles_columns, datapoint.smiles):
datapoint.row[column] = smiles
# Add predictions columns
for pred_name, pred in zip(task_names, preds):
datapoint.row[pred_name] = pred
# Save
"""
with open(args.preds_path, 'w') as f:
writer = csv.DictWriter(f, fieldnames=full_data[0].row.keys())
writer.writeheader()
for datapoint in full_data:
writer.writerow(datapoint.row)
"""
return avg_preds, dict(full_data[0].row)
def predict():
"""Renders the predict page and makes predictions if the method is POST."""
if request.method == 'GET':
return render_predict()
# Get arguments
ckpt_id = request.form['checkpointName']
if request.form['textSmiles'] != '':
smiles = request.form['textSmiles'].split()
elif request.form['drawSmiles'] != '':
smiles = [request.form['drawSmiles']]
else:
# Upload data file with SMILES
data = request.files['data']
data_name = secure_filename(data.filename)
data_path = os.path.join(app.config['TEMP_FOLDER'], data_name)
data.save(data_path)
# Check if header is smiles
possible_smiles = get_header(data_path)[0]
smiles = [possible_smiles
] if Chem.MolFromSmiles(possible_smiles) is not None else []
# Get remaining smiles
smiles.extend(get_smiles(data_path))
models = db.get_models(ckpt_id)
model_paths = [
os.path.join(app.config['CHECKPOINT_FOLDER'], f'{model["id"]}.pt')
for model in models
]
task_names = load_task_names(model_paths[0])
num_tasks = len(task_names)
gpu = request.form.get('gpu')
train_args = load_args(model_paths[0])
# Build arguments
arguments = [
'--test_path', 'None', '--preds_path',
os.path.join(app.config['TEMP_FOLDER'],
app.config['PREDICTIONS_FILENAME']), '--checkpoint_paths',
*model_paths
]
if gpu is not None:
if gpu == 'None':
arguments.append('--no_cuda')
else:
arguments += ['--gpu', gpu]
# Handle additional features
if train_args.features_path is not None:
# TODO: make it possible to specify the features generator if trained using features_path
arguments += [
'--features_generator', 'rdkit_2d_normalized',
'--no_features_scaling'
]
elif train_args.features_generator is not None:
arguments += ['--features_generator', *train_args.features_generator]
if not train_args.features_scaling:
arguments.append('--no_features_scaling')
# Parse arguments
args = PredictArgs().parse_args(arguments)
# Run predictions
preds = make_predictions(args=args, smiles=smiles)
if all(p is None for p in preds):
return render_predict(errors=['All SMILES are invalid'])
# Replace invalid smiles with message
invalid_smiles_warning = 'Invalid SMILES String'
preds = [
pred if pred is not None else [invalid_smiles_warning] * num_tasks
for pred in preds
]
return render_predict(
predicted=True,
smiles=smiles,
num_smiles=min(10, len(smiles)),
show_more=max(0,
len(smiles) - 10),
task_names=task_names,
num_tasks=len(task_names),
preds=preds,
warnings=["List contains invalid SMILES strings"]
if None in preds else None,
errors=["No SMILES strings given"] if len(preds) == 0 else None)
def test_bad_path(self):
"""Test bad provided path """
bad_path = os.path.join(self.temp_dir.name, 'bad_path.csv')
with self.assertRaises(FileNotFoundError):
get_header(bad_path)
def test_correct_file(self):
""" Test correct input """
header = get_header(os.path.join(self.temp_dir.name, 'dummy_data.csv'))
self.assertEqual(header, ['column0', 'column1'])