def test_statistics(self, qm9_dataset, split_path, args):
# test for statistics not in split file
if os.path.exists(split_path):
os.remove(split_path)
os.makedirs(os.path.dirname(split_path), exist_ok=True)
train, val, test = spk.data.train_test_split(qm9_dataset, 10, 5,
split_path)
train_loader = spk.data.AtomsLoader(train, batch_size=5)
mean, stddev = get_statistics(
split_path=split_path,
train_loader=train_loader,
args=args,
atomref=None,
per_atom=False,
)
energies = []
for batch in train_loader:
energies.append(batch["energy_U0"])
assert_almost_equal(torch.cat(energies).mean(), mean["energy_U0"], 2)
# test for statistics in split file
split_file = np.load(split_path)
saved_mean = split_file["mean"]
mean, stddev = get_statistics(
split_path=split_path,
train_loader=train_loader,
args=args,
atomref=None,
per_atom=False,
)
assert_almost_equal(saved_mean, mean["energy_U0"])
# test assertion on wrong split file
with pytest.raises(Exception):
get_statistics(
split_path="I/do/not/exist.npz",
train_loader=train_loader,
args=args,
atomref=None,
per_atom=False,
)
def main(args):
# setup
train_args = setup_run(args)
device = torch.device("cuda" if args.cuda else "cpu")
# get dataset
environment_provider = get_environment_provider(train_args, device=device)
dataset = get_dataset(train_args,
environment_provider=environment_provider)
# get dataloaders
split_path = os.path.join(args.modelpath, "split.npz")
train_loader, val_loader, test_loader = get_loaders(args,
dataset=dataset,
split_path=split_path,
logging=logging)
# define metrics
metrics = get_metrics(train_args)
# train or evaluate
if args.mode == "train":
# get statistics
atomref = dataset.get_atomref(args.property)
mean, stddev = get_statistics(
args=args,
split_path=split_path,
train_loader=train_loader,
atomref=atomref,
divide_by_atoms=get_divide_by_atoms(args),
logging=logging,
)
# build model
model = get_model(args,
train_loader,
mean,
stddev,
atomref,
logging=logging)
# build trainer
logging.info("training...")
trainer = get_trainer(args, model, train_loader, val_loader, metrics)
# run training
trainer.train(device, n_epochs=args.n_epochs)
logging.info("...training done!")
else:
raise ("Use the original SchnetPack script instead.")
def main(args):
#building model and dataset
device = torch.device("cuda" if args.cuda else "cpu")
environment_provider = spk.environment.AseEnvironmentProvider(cutoff=5.0)
omdb = './omdb'
if args.mode == "train":
if not os.path.exists(os.path.join(args.model_path)):
os.makedirs(args.model_path)
spk.utils.spk_utils.set_random_seed(None)
if not os.path.exists('omdb'):
os.makedirs(omdb)
omdData = OrganicMaterialsDatabase(
args.datapath,
download=False,
load_only=[args.property],
environment_provider=environment_provider)
# split_path = os.path.join(args.model_path, "split.npz")
split_path = os.path.join(
'/home/s3754715/gnn_molecule/schnetpack/model_2020-06-23-18-44-59',
"split.npz")
train, val, test = spk.train_test_split(data=omdData,
num_train=9000,
num_val=1000,
split_file=split_path)
print('-----------')
print(len(train))
print(len(val))
print(len(test))
print('-------------')
train_loader = spk.AtomsLoader(train,
batch_size=16,
sampler=RandomSampler(train),
num_workers=4
#pin_memory=True
)
val_loader = spk.AtomsLoader(val, batch_size=16, num_workers=2)
test_loader = spk.AtomsLoader(test, batch_size=16, num_workers=2)
atomref = omdData.get_atomref(args.property)
mean, stddev = get_statistics(
args=args,
split_path=split_path,
train_loader=train_loader,
atomref=atomref,
divide_by_atoms=get_divide_by_atoms(args),
logging=logging)
# means, stddevs = train_loader.get_statistics(
# args.property, get_divide_by_atoms(args),atomref
# )
model_train = model(args, omdData, atomref, mean, stddev)
trainer = train_model(args, model_train, train_loader, val_loader)
print('started training')
trainer.train(device=device, n_epochs=args.n_epochs)
print('training finished')
sch_model = torch.load(os.path.join(args.model_path, 'best_model'))
err = 0
sch_model.eval()
for count, batch in enumerate(test_loader):
# move batch to GPU, if necessary
batch = {k: v.to(device) for k, v in batch.items()}
# apply model
pred = sch_model(batch)
# calculate absolute error
tmp = torch.sum(
torch.abs(pred[args.property] - batch[args.property]))
tmp = tmp.detach().cpu().numpy(
) # detach from graph & convert to numpy
err += tmp
print(tmp)
# log progress
percent = '{:3.2f}'.format(count / len(test_loader) * 100)
print('Progress:',
percent + '%' + ' ' * (5 - len(percent)),
end="\r")
err /= len(test)
print('Test MAE', np.round(err, 3), 'eV =',
np.round(err / (kcal / mol), 3), 'kcal/mol')
#plot results
plot_results(args)
elif args.mode == "pred":
print('predictionsss')
sch_model = torch.load(os.path.join(args.model_path, 'best_model'),
map_location=torch.device(device))
#reading test data
# test_dataset = AtomsData('./cod_predict.db')
# test_loader = spk.AtomsLoader(test_dataset, batch_size=32)
#reading stored cod list
#cod_list = np.load('./cod_id_list_old.npy')
omdData = OrganicMaterialsDatabase(
args.datapath,
download=True,
load_only=[args.property],
environment_provider=environment_provider)
split_path = os.path.join(args.model_path, "split.npz")
train, val, test = spk.train_test_split(data=omdData,
num_train=9000,
num_val=1000,
split_file=split_path)
print(len(test))
test_loader = spk.AtomsLoader(
test,
batch_size=32, #num_workers=2
)
mean_abs_err = 0
prediction_list = []
actual_value_list = []
print('Started generating predictions')
for count, batch in enumerate(test_loader):
# move batch to GPU, if necessary
print('before batch')
batch = {k: v.to(device) for k, v in batch.items()}
print('after batch')
# apply model
pred = sch_model(batch)
prediction_list.extend(
pred['band_gap'].detach().cpu().numpy().flatten().tolist())
actual_value_list.extend(
batch['band_gap'].detach().cpu().numpy().flatten().tolist())
# log progress
percent = '{:3.2f}'.format(count / len(test_loader) * 100)
print('Progress:',
percent + '%' + ' ' * (5 - len(percent)),
end="\r")
cod_arr = np.genfromtxt(
os.path.join(
'/home/s3754715/gnn_molecule/schnetpack/dataset/OMDB-GAP1_v1.1',
'CODids.csv'))
cod_list = cod_arr[10000:].tolist()
results_df = pd.DataFrame({
'cod': cod_list,
'prediction': prediction_list,
'actual': actual_value_list
})
results_df.to_csv('./predictions.csv')
def main(args):
# setup
train_args = setup_run(args)
device = torch.device("cuda" if args.cuda else "cpu")
# get dataset
environment_provider = get_environment_provider(train_args, device=device)
dataset = get_dataset(train_args,
environment_provider=environment_provider)
# get dataloaders
split_path = os.path.join(args.modelpath, "split.npz")
train_loader, val_loader, test_loader = get_loaders(args,
dataset=dataset,
split_path=split_path,
logging=logging)
# define metrics
metrics = get_metrics(train_args)
# train or evaluate
if args.mode == "train":
# get statistics
atomref = dataset.get_atomref(args.property)
mean, stddev = get_statistics(
args=args,
split_path=split_path,
train_loader=train_loader,
atomref=atomref,
divide_by_atoms=get_divide_by_atoms(args),
logging=logging,
)
# build model
model = get_model(args,
train_loader,
mean,
stddev,
atomref,
logging=logging)
# build trainer
logging.info("training...")
trainer = get_trainer(args, model, train_loader, val_loader, metrics)
# run training
trainer.train(device, n_epochs=args.n_epochs)
logging.info("...training done!")
elif args.mode == "eval":
# remove old evaluation files
evaluation_fp = os.path.join(args.modelpath, "evaluation.txt")
if os.path.exists(evaluation_fp):
if args.overwrite:
os.remove(evaluation_fp)
else:
raise ScriptError(
"The evaluation file does already exist at {}! Add overwrite flag"
" to remove.".format(evaluation_fp))
# load model
logging.info("loading trained model...")
model = torch.load(os.path.join(args.modelpath, "best_model"))
# run evaluation
logging.info("evaluating...")
if spk.utils.get_derivative(train_args) is None:
with torch.no_grad():
evaluate(
args,
model,
train_loader,
val_loader,
test_loader,
device,
metrics=metrics,
)
else:
evaluate(
args,
model,
train_loader,
val_loader,
test_loader,
device,
metrics=metrics,
)
logging.info("... evaluation done!")
else:
raise ScriptError("Unknown mode: {}".format(args.mode))
def main(args):
# setup
# train_args = setup_run(args)
logging.info("CUDA is used: " + str(args.cuda))
if args.cuda:
logging.info("CUDA is available: " + str(torch.cuda.is_available()))
device = torch.device("cuda" if args.cuda else "cpu")
# get dataset
dataset = get_dataset(args)
# get dataloaders
split_path = os.path.join(args.modelpath, "split.npz")
train_loader, val_loader, test_loader = get_loaders(args,
dataset=dataset,
split_path=split_path,
logging=logging)
# define metrics
metrics = get_metrics(args)
# train or evaluate
if args.mode == "train":
# get statistics
atomref = dataset.get_atomref(args.property)
divide_by_atoms = settings.divide_by_atoms[args.property]
mean, stddev = get_statistics(
args=args,
split_path=split_path,
train_loader=train_loader,
atomref=atomref,
divide_by_atoms=divide_by_atoms,
logging=logging,
)
aggregation_mode = settings.pooling_mode[args.property]
# build model
model = get_model(args,
train_loader,
mean,
stddev,
atomref,
aggregation_mode,
logging=logging)
# build trainer
logging.info("training...")
trainer = get_trainer(args, model, train_loader, val_loader, metrics)
# run training
trainer.train(device, n_epochs=args.n_epochs)
logging.info("...training done!")
elif args.mode == "eval":
# remove old evaluation files
evaluation_fp = os.path.join(args.modelpath, "evaluation.txt")
if os.path.exists(evaluation_fp):
if args.overwrite:
os.remove(evaluation_fp)
else:
raise ScriptError(
"The evaluation file does already exist at {}! Add overwrite flag"
" to remove.".format(evaluation_fp))
# load model
logging.info("loading trained model...")
model = torch.load(os.path.join(args.modelpath, "best_model"))
# run evaluation
logging.info("evaluating...")
if args.dataset != "md17":
with torch.no_grad():
evaluate(
args,
model,
train_loader,
val_loader,
test_loader,
device,
metrics=metrics,
)
else:
evaluate(
args,
model,
train_loader,
val_loader,
test_loader,
device,
metrics=metrics,
)
logging.info("... evaluation done!")
else:
raise ScriptError("Unknown mode: {}".format(args.mode))