在test.py文件中，DataLoader(test_data, shuffle=True, batch_size=batch_size) shuffle应该设为False，这是一个很低级的BUG，卡了我一天，在最意想不到的地方有BUG

您好，可以提供一个训练数据的格式案例吗，谢谢

想获得两个句子相似的概率，_, out_classes = output_probabilities.max(dim=1)的_是否是两个句子相似的概率

你好，这个词向量矩阵与vocab是否错位？ embedding_matrix[idx + 1] = model[word]#词向量矩阵，然而vocab前两位是pad 和 unk. 而词向量第一个位置为的。

在看esim模型时，lstm 的embedding 初始化参数使用预训练的，预训练word2vec的model存在69840token，vocab.txt存在69842token,补充了2个[PAD]和[UNK]这样token,在构建预训练imbedding参数矩阵时，embedding_matrix = np.zeros((len(model.index2word) + 1, model.vector_size))，这行代码是否需要改成embedding_matrix = np.zeros((len(model.index2word) + 2, model.vector_size))，这样[PAD]和[UNK]这样token分别对应第0行和1行，与word2idx = {word: index for index, word in enumerate(vocab)}这样相对应.

我之前用bert做编码，再用孪生网络，效果比ESIM差不少

I've downloaded the .tar models of ESIM and BiMPM, but I can't unzip them using any command.

Hi,I want to know how to define the train data (validation data) label?now i am define 1 means match ,and 0 means not match,but I train model ,there is a error occurs,I find maybe the label doesn't match to network output logits and probs .Thanks!

请问训练的时候出现这个问题 Length of all samples has to be greater than 0, but found an element in 'lengths' that is <= 0，如何解决呀

labels 的内容是什么

Traceback (most recent call last):
File ".\test.py", line 51, in
main("d:\test.csv", "models/best.pth.tar")
File ".\test.py", line 43, in main
model.load_state_dict(checkpoint["model"])
File "C:\Python\Python38\lib\site-packages\torch\nn\modules\module.py", line 1044, in load_state_dict
raise RuntimeError('Error(s) in loading state_dict for {}:\n\t{}'.format(
RuntimeError: Error(s) in loading state_dict for BertModelTest:
Missing key(s) in state_dict: "bert.bert.embeddings.position_ids".

如何解决？