AnFreTh/STREAM
A versatile Python package engineered for seamless topic modeling, topic evaluation, and topic visualization. Ideal for text analysis, natural language processing (NLP), and research in the social sciences, STREAM simplifies the extraction, interpretation, and visualization of topics from large, complex datasets.
PythonMIT
Issues
- 2
- 3
Pruning when performing HPO for TNTM (and CTM)
#59 opened by ArikReuter - 1
- 1
add github as data source option
#75 opened by mkumar73 - 1
delete models/ctmneg_utils folder
#73 opened by AnFreTh - 0
add learning rate to hparam optimization in NTMs
#66 opened by AnFreTh - 1
Import time and huggingface hub
#64 opened by AnFreTh - 0
Make metrics usable for other models
#17 opened by AnFreTh - 0
Include real word embeddings for TNTM, ETM
#62 opened by AnFreTh - 1
Refactor TMDataset
#55 opened by mkumar73 - 0
Include Word2Vec for metrics
#19 opened by AnFreTh - 1
Adapt metrics to make HPO possible
#14 opened by AnFreTh - 0
Test and Check LDA implementation
#27 opened by AnFreTh - 1
- 1
- 1
refactor default preprocessing config
#47 opened by mkumar73 - 1
stream to stream_topic
#45 opened by mkumar73 - 1
- 1
stream commons to avoid circular import
#48 opened by mkumar73 - 1
- 1
Reuters dataset
#20 opened by AnFreTh - 1
BBC News Dataset
#21 opened by AnFreTh - 0
NAM usable with new structure
#16 opened by AnFreTh - 0
Adapt WordCluTM to new structure
#29 opened by AnFreTh - 0
20 Newsgroup dataset
#24 opened by AnFreTh - 2
Reddit Dataset
#22 opened by AnFreTh - 1
Stocktwits dataset
#23 opened by AnFreTh - 0
Spotify dataset
#25 opened by AnFreTh - 0
Poliblogs dataset
#26 opened by AnFreTh - 0
- 0
- NeuralLDA
#13 opened by AnFreTh - 0
- 0
- 0
- 1
- 0
Make current metrics more efficient
#18 opened by AnFreTh - 1
Manage branch protections rules
#3 opened by AnFreTh - 0
- 0
- 0
Additional Preprocessing
#4 opened by AnFreTh - 0
new languages
#5 opened by AnFreTh