source/nli.py [270:283]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
corr_best = 0
# loop multiple times over the dataset
for epoch in range(args.nepoch):

    loss_epoch = 0.0
    print('Ep {:4d}'.format(epoch), end='')
    # for inputs, labels in train_loader:
    for i, data in enumerate(train_loader, 0):
        # get the inputs
        inputs, labels = data
        labels = labels.long()
        if args.gpu >= 0:
            inputs = inputs.cuda()
            labels = labels.cuda()
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -



source/sent_classif.py [219:232]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
corr_best = 0
# loop multiple times over the dataset
for epoch in range(args.nepoch):

    loss_epoch = 0.0
    print('Ep {:4d}'.format(epoch), end='')
    # for inputs, labels in train_loader:
    for i, data in enumerate(train_loader, 0):
        # get the inputs
        inputs, labels = data
        labels = labels.long()
        if args.gpu >= 0:
            inputs = inputs.cuda()
            labels = labels.cuda()
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -