from linguistic_style_transfer_pytorch.config import GeneralConfig, ModelConfig
from linguistic_style_transfer_pytorch.data_loader import TextDataset
from linguistic_style_transfer_pytorch.model import AdversarialVAE
from tqdm import tqdm, trange
import os
import numpy as np
import pickle
use_cuda = True if torch.cuda.is_available() else False
if __name__ == "__main__":
mconfig = ModelConfig()
gconfig = GeneralConfig()
weights = torch.FloatTensor(np.load(gconfig.word_embedding_path))
model = AdversarialVAE(weight=weights)
if use_cuda:
model = model.cuda()
#=============== Define dataloader ================#
train_dataset = TextDataset(mode='train')
train_dataloader = DataLoader(train_dataset, batch_size=mconfig.batch_size)
content_discriminator_params, style_discriminator_params, vae_and_classifier_params = model.get_params(
)
#============== Define optimizers ================#
# content discriminator/adversary optimizer
content_disc_opt = torch.optim.RMSprop(content_discriminator_params,
lr=mconfig.content_adversary_lr)
# style discriminaot/adversary optimizer
style_disc_opt = torch.optim.RMSprop(style_discriminator_params,
lr=mconfig.style_adversary_lr)
import torch
import os
import argparse
import numpy as np
import pickle
from linguistic_style_transfer_pytorch.config import GeneralConfig
from linguistic_style_transfer_pytorch.model import AdversarialVAE
gconfig = GeneralConfig()
# load word embeddings
weights = torch.FloatTensor(np.load(gconfig.word_embedding_path))
# load checkpoint
model_checkpoint = torch.load('checkpoints/model_epoch_20.pt')
# Load model
model = AdversarialVAE(weights=weights)
model.load_state_dict(model_checkpoint)
model.eval()
# Load average style embeddings
with open(config.avg_style_emb_path, 'rb') as f:
avg_style_embeddings = pickle.load(f)
# set avg_style_emb attribute of the model
model.avg_style_emb = avg_style_embeddings
# load word2index
with open(gconfig.w2i_file_path) as f:
word2index = json.load(f)
# load index2word
with open(gconfig.i2w_file_path) as f:
index2word = json.load(f)
label2index = {'neg': 0, 'pos': 1}
# Read input sentence
use_cuda = False
device = torch.device('cpu')
if torch.cuda.is_available():
use_cuda = True
device = torch.device('cuda:0')
print('using backend(', device, ')')
if __name__ == "__main__":
mconfig = ModelConfig()
gconfig = GeneralConfig()
weights = torch.tensor(np.load(gconfig.word_embedding_path),
device=device,
dtype=torch.float)
model = AdversarialVAE(inference=False, weight=weights, device=device)
if use_cuda:
model = model.cuda()
#=============== Define dataloader ================#
train_dataset = TextDataset(mode='train')
train_dataloader = DataLoader(train_dataset,
batch_size=mconfig.batch_size,
drop_last=True,
pin_memory=True)
content_discriminator_params, style_discriminator_params, vae_and_classifier_params = model.get_params(
)
#============== Define optimizers ================#
# content discriminator/adversary optimizer
content_disc_opt = torch.optim.RMSprop(content_discriminator_params,
lr=mconfig.content_adversary_lr)