session-based-news-recommendation icon indicating copy to clipboard operation
session-based-news-recommendation copied to clipboard

Missing files on MIND dataset

Open LUUUUUUZ opened this issue 2 years ago • 6 comments
trafficstars

Could you share some missing files on MIND dataset? python mind_preprocess.py --content_info It seems ../data/mind/articles_embeddings_1.pkl is missing, which I thought maybe is generated from get_content_vec.py. But how could I load the pre-trained model?

Besides, python data_process/generate_neighbor.py It seems publish_time_1.txt is missing?

I will really appreciate it if you could provide above two files. Thank you!

LUUUUUUZ avatar Dec 01 '22 10:12 LUUUUUUZ

articles_embeddings_1.pkl: articles_embeddings_1.zip

publish_time_1.txt can be generated in mind_preprocess.py.

summmeer avatar Dec 01 '22 11:12 summmeer

Thanks! Have a good day ;)

LUUUUUUZ avatar Dec 01 '22 12:12 LUUUUUUZ

Sorry to bother again. It seems MIND/articles_category.pkl is also missing. Although it's simple to generate but I don't know the form of this data structure. I will appreciate it if you could provide this file. Thanks!

LUUUUUUZ avatar Dec 01 '22 14:12 LUUUUUUZ

And could you share your environment? like python version and tf version?

LUUUUUUZ avatar Dec 01 '22 14:12 LUUUUUUZ

articles_category is a dictionary, pseudo code:

articles_category = {}
for each news_id, news_category:
    articles_category[news_id] = news_category
pickle.dump(articles_category, open('../articles_category.pkl', 'wb'))

Environment info: Python: 3.6.9, TensorFlow: 1.14.0

summmeer avatar Dec 02 '22 02:12 summmeer

Thank you so much! It's so nice of you to help me with so many problems. Enjoy your weekend! :)

LUUUUUUZ avatar Dec 02 '22 07:12 LUUUUUUZ