GLOVE_train_test_set.hdf5
收藏NIAID Data Ecosystem2026-05-02 收录
下载链接：
https://zenodo.org/record/12759355
下载链接
链接失效反馈官方服务：
资源简介：
GLOVE_100 dataset contains one million pre-trained Global Vectors for Word Representation (GloVe) embeddings that capture semantic relationships between words. These 100-dimensional embedding vectors were pre-trained on the combined Wikipedia 2014 + Gigaword 5th Edition corpora (6B tokens, 400K vocabulary) and cover a wide range of English words, enabling comprehensive assessments of approximate similarity search methods across diverse vocabulary and word relationships.
创建时间：
2024-07-17
