def set_seeds():
torch.manual_seed(SimpleRandom.get_seed())
if gpu():
torch.cuda.manual_seed_all(SimpleRandom.get_seed())
np.random.seed(SimpleRandom.get_seed())
random.seed(SimpleRandom.get_seed())
def set_seeds():
torch.manual_seed(SimpleRandom.get_seed(seed_from_config_file))
if gpu():
torch.cuda.manual_seed_all(SimpleRandom.get_seed())
np.random.seed(SimpleRandom.get_seed())
random.seed(SimpleRandom.get_seed())
def prepare(data):
#Note, we should just be passing in a sparse minibatch here! Doing todense on the entire datset is silly
if issparse(data):
data = data.todense()
v = torch.FloatTensor(np.array(data))
if gpu():
return Variable(v.cuda())
return Variable(v)
def prepare(data):
data = data.todense()
v = torch.FloatTensor(np.array(data))
if gpu():
return Variable(v.cuda())
return Variable(v)
def prepare_with_labels(data, labels):
data = data.todense()
v = torch.FloatTensor(np.array(data))
if gpu():
return Variable(v.cuda()), Variable(torch.LongTensor(labels).cuda())
return Variable(v), Variable(torch.LongTensor(labels))
train_ds.read()
dev_ds.read()
test_ds = None
if args.test is not None:
test_ds = DataSet(file=args.test, reader=jlr, formatter=formatter)
test_ds.read()
train_feats, dev_feats, test_feats = f.load(train_ds, dev_ds, test_ds)
f.save_vocab(mname)
input_shape = train_feats[0].shape[1]
model = SimpleMLP(input_shape,100,3)
if gpu():
model.cuda()
if model_exists(mname) and os.getenv("TRAIN").lower() not in ["y","1","t","yes"]:
model.load_state_dict(torch.load("models/{0}.model".format(mname)))
else:
train(model, train_feats, 500, 1e-2, 90,dev_feats,early_stopping=EarlyStopping(mname))
torch.save(model.state_dict(), "models/{0}.model".format(mname))
print_evaluation(model, dev_feats, FEVERLabelSchema())
if args.test is not None:
print_evaluation(model, test_feats, FEVERLabelSchema())