only 0.71 acc on Weibo Dataset

[Tangnameless]

On the Weibo dataset, I only got 71 accuracy socre.
I didn't change your model or training parameters.
Cause I don't have image_embed.pkl and xx_content_segmented.txt, I can only preprocess the data according to my own guess.

[gymbeijing]

Hi @Tangnameless , I was trying to reproduce the model on Twitter dataset. But I found some file missing. How did you handle the missing files on the Weibo dataset?

[Tangnameless]

Hi @Tangnameless , I was trying to reproduce the model on Twitter dataset. But I found some file missing. How did you handle the missing files on the Weibo dataset?

没有进行Twitter数据集的实验，对于Weibo数据集

1. 对于缺少的文本分词，我使用jieba进行中文分词，然后按照论文说的，用训练集自己训练32维的word2vec词向量。（直觉上觉得先把微博文本翻译成英文再嵌入多此一举）
2. 对于缺少的图像嵌入，直接使用pytorch提供的预训练vgg-19，提取倒数第二层，输出一个4096维的向量。
   由于不知道确切的预处理步骤，复现效果不理想