freeze_embeddings=args.freeze_emb, gpu=args.gpu
)
if args.gpu:
model.load_state_dict(torch.load('models/{}.bin'
.format(args.model
+ utils.getModelName(args))))
else:
model.load_state_dict(torch.load('models/{}.bin'
.format(args.model
+ utils.getModelName(args)),
map_location=lambda storage, loc: storage)
)
# Samples latent and conditional codes randomly from prior
z = model.sample_z_prior(1)
c = model.sample_c_prior(1)
# Generate positive sample given z
c[0, 0], c[0, 1] = 1, 0
_, c_idx = torch.max(c, dim=1)
sample_idxs = model.sample_sentence(z, c, temp=0.1)
print('\nSentiment: {}'.format(dataset.idx2label(int(c_idx))))
print('Generated: {}'.format(dataset.idxs2sentence(sample_idxs)))
# Generate negative sample from the same z
c[0, 0], c[0, 1] = 0, 1
_, c_idx = torch.max(c, dim=1)
if args.gpu:
model.load_state_dict(
torch.load('models/{}.bin'.format(args.model +
utils.getModelName(args))))
else:
model.load_state_dict(
torch.load('models/{}.bin'.format(args.model +
utils.getModelName(args)),
map_location=lambda storage, loc: storage))
outputFile = open("models/gen" + utils.getModelName(args) + ".out.txt", "w+")
for i in range(n_iter):
# Samples latent and conditional codes randomly from prior
z = model.sample_z_prior(1)
c = model.sample_c_prior(1)
# Generate positive sample given z
for j in range(args.num_classes):
for k in range(args.num_classes):
c[0, k] = 0
c[0, j] = 1
_, c_idx = torch.max(c, dim=1)
sample_idxs = model.sample_sentence(z, c, temp=0.1)
outputFile.write('{}\t{}\n'.format(dataset.idxs2sentence(sample_idxs),
str(dataset.idx2label(int(c_idx)))))