本文整理汇总了Python中datasets.prepare_data方法的典型用法代码示例。如果您正苦于以下问题：Python datasets.prepare_data方法的具体用法？Python datasets.prepare_data怎么用？Python datasets.prepare_data使用的例子？那么恭喜您, 这里精选的方法代码示例或许可以为您提供帮助。您也可以进一步了解该方法所在类datasets的用法示例。
在下文中一共展示了datasets.prepare_data方法的1个代码示例，这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞，您的评价将有助于系统推荐出更棒的Python代码示例。
示例1: evaluate
# 需要导入模块: import datasets [as 别名]
# 或者: from datasets import prepare_data [as 别名]
def evaluate(dataloader, cnn_model, rnn_model, batch_size):
cnn_model.eval()
rnn_model.eval()
s_total_loss = 0
w_total_loss = 0
for step, data in enumerate(dataloader, 0):
real_imgs, captions, cap_lens, \
class_ids, keys = prepare_data(data)
words_features, sent_code = cnn_model(real_imgs[-1])
# nef = words_features.size(1)
# words_features = words_features.view(batch_size, nef, -1)
hidden = rnn_model.init_hidden(batch_size)
words_emb, sent_emb = rnn_model(captions, cap_lens, hidden)
w_loss0, w_loss1, attn = words_loss(words_features, words_emb, labels,
cap_lens, class_ids, batch_size)
w_total_loss += (w_loss0 + w_loss1).data
s_loss0, s_loss1 = \
sent_loss(sent_code, sent_emb, labels, class_ids, batch_size)
s_total_loss += (s_loss0 + s_loss1).data
if step == 50:
break
s_cur_loss = s_total_loss[0] / step
w_cur_loss = w_total_loss[0] / step
return s_cur_loss, w_cur_loss