candidate = news_words[all_train_pn[i]] ：IndexError: index 29998 is out of bounds for axis 0 with size 51
Opened this issue · 9 comments
File "E:/project/IJCAI2019-NAML-master/naml.py", line 278, in generate_batch_data_train
candidate = news_words[all_train_pn[i]] # new[]
IndexError: index 29998 is out of bounds for axis 0 with size 51
HI：
i think new_words are record the words of title，but the all_train_pn are record the id of news，so it get a Error。
File "E:/project/IJCAI2019-NAML-master/naml.py", line 278, in generate_batch_data_train
candidate = news_words[all_train_pn[i]] # new[]
IndexError: index 29998 is out of bounds for axis 0 with size 51HI：
i think new_words are record the words of title，but the all_train_pn are record the id of news，so it get a Error。
Clearly, the sample data is limited...
File "E:/project/IJCAI2019-NAML-master/naml.py", line 278, in generate_batch_data_train
candidate = news_words[all_train_pn[i]] # new[]
IndexError: index 29998 is out of bounds for axis 0 with size 51
HI：
i think new_words are record the words of title，but the all_train_pn are record the id of news，so it get a Error。Clearly, the sample data is limited...
我看了源码的处理方式，all_train_pn是收集编码后的title id。实际上，两个csv的记录不一样的，一份是记录title编码后的id，另一份是还没有编码之前的title。有好的数据集通知我一下，就差数据集了。作者提供这份样本数据很难跑得起来。
File "E:/project/IJCAI2019-NAML-master/naml.py", line 278, in generate_batch_data_train
candidate = news_words[all_train_pn[i]] # new[]
IndexError: index 29998 is out of bounds for axis 0 with size 51
HI：
i think new_words are record the words of title，but the all_train_pn are record the id of news，so it get a Error。Clearly, the sample data is limited...
老哥后来解决了这个问题吗
老哥后来解决了这个问题吗
好像解决了，用了自己的数据集重写了
老哥后来解决了这个问题吗
好像解决了，用了自己的数据集重写了
老哥的数据集和代码还留着吗，如果方便的话想试试。我这边跑通了但是有新闻没点击数据，不太好验证。
老哥后来解决了这个问题吗
好像解决了，用了自己的数据集重写了
老哥的数据集和代码还留着吗，如果方便的话想试试。我这边跑通了但是有新闻没点击数据，不太好验证。
抱歉啊，老哥，那个是公司的代码
老哥后来解决了这个问题吗
你好，请问你解决这个问题了吗，我不知道怎么去找合适的数据集
很早之前的事情，有点忘记了