eval_batch_size = 64
train_data = batchify(corpus.train, args.batch_size)
val_data = batchify(corpus.valid, eval_batch_size)
test_data = batchify(corpus.test, eval_batch_size)
print('INFO: Data loaded.')
###############################################################################
# Build the model
###############################################################################
ntokens = len(corpus.dictionary)
model = PRPN(ntokens, args.emsize, args.nhid, args.nlayers, args.nslots,
args.nlookback, args.resolution, args.dropout, args.idropout,
args.rdropout, args.tied, args.hard, args.res)
if args.cuda:
model.cuda()
# criterion = nn.CrossEntropyLoss()
def criterion(input, targets, targets_mask):
targets_mask = targets_mask.view(-1)
input = input.view(-1, ntokens)
input = F.log_softmax(input, dim=-1)
loss = torch.gather(input, 1, targets[:, None]).view(-1)
loss = (-loss * targets_mask).sum() / targets_mask.sum()
return loss
return data_batched
eval_batch_size = 64
train_data = batchify(corpus.train, args.batch_size)
val_data = batchify(corpus.valid, eval_batch_size)
test_data = batchify(corpus.test, eval_batch_size)
###############################################################################
# Build the model
###############################################################################
ntokens = len(corpus.dictionary)
model = PRPN(ntokens, args.emsize, args.nhid, args.nlayers,
args.nslots, args.nlookback, args.resolution,
args.dropout, args.idropout, args.rdropout,
args.tied, args.hard, args.res)
if args.cuda:
model.cuda()
# criterion = nn.CrossEntropyLoss()
def criterion(input, targets, targets_mask):
targets_mask = targets_mask.view(-1)
input = input.view(-1, ntokens)
input = F.log_softmax(input, dim=-1)
loss = torch.gather(input, 1, targets[:, None]).view(-1)
loss = (-loss * targets_mask).sum() / targets_mask.sum()
return loss