TopicVec
TopicVec is the source code for "Generative Topic Embedding: a Continuous Representation of Documents" (ACL 2016).
PSDVec (in folder 'psdvec') is the source code for "A Generative Word Embedding Model and its Low Rank Positive Semidefinite Solution" (EMNLP 2015).
Update v0.7:
The topic inference is now 6 times faster.
Update v0.6:
Algorithm update:
topicvecDir.py: uses exact inference instead a second-order approximation in the M-step.
Update v0.5:
Main algorithm:
topicvecDir.py: uses a Dirichlet prior for topic mixting proportions.
####Required files on Dropbox: https://www.dropbox.com/sh/lqbk3iioobegbp8/AACc8Kfr1KZIkKl9bGaIrOjfa?dl=0
- Pretrained 180000 embeddings (25000 cores) in 3 archives. For faster loading into Python, 25000-180000-500-BLK-8.0.vec.npy can be used;
- Unigram files top1grams-wiki.txt & top1grams-reuters.txt;
- RCV1 cleansed corpus ( before downloading, please apply for permission from NIST according to: http://trec.nist.gov/data/reuters/reuters.html ).
If you are in China, you can also download the above files from baidu netdisk without the hassle of "climbing over the wall": https://pan.baidu.com/s/1gVmRhK1HA2XwVWZbZHHLZQ#list/path=%2F