machine-intelligence-laboratory/OptimalNumberOfTopics
A set of methods for finding an appropriate number of topics in a text collection
PythonMIT
Issues
- 0
TopicNet datasets not loading
#101 opened by Alvant - 0
Dataset.load: only once for each dataset
#98 opened by Alvant - 0
SearchInterval: standardize topic grid
#97 opened by Alvant - 0
Search interval v2: simple list as input
#88 opened by Alvant - 0
Exclude notebooks from code stats
#93 opened by Alvant - 0
Stability method: speed up subdatasets creation
#87 opened by Alvant - 0
Notebook with additional scores computation
#90 opened by Alvant - 2
- 0
Wrong paper in docstring for lift score
#86 opened by Alvant - 2
- 0
brunet_metric
#89 opened by Alvant - 0
- 0
data_utils -> into the topka
#85 opened by Alvant - 0
Perplexity == HoldoutPerplexity (But Shouldn't Be)
#77 opened by Alvant - 0
Merge Lift Score
#80 opened by Alvant - 0
- 0
Run experiments with KnownModel.TLESS
#70 opened by Alvant - 2
- 0
Arun score may break sometimes
#73 opened by Alvant - 0
- 1
- 0
Test for scores save/load
#45 opened by Alvant - 0
Big data mode
#57 opened by Alvant - 1
Actualize notebooks: load datasets using TopicNet
#50 opened by Alvant - 0
Test for test script
#48 opened by Alvant - 0
Stability approach. Holdout + Restarts
#17 opened by Alvant - 4
list index out of range
#51 opened by SergeyShk - 1
Manage repo topics when kinda ready
#7 opened by Alvant - 1
- 0
DRY code for Dataset save/load in scores
#46 opened by Alvant - 2
no need to calculate scores at each iteration
#19 opened by bt2901 - 0
Scores. Settle down: minimize all
#24 opened by Alvant - 0
Demo notebook
#5 opened by Alvant - 0
- 0
Smoke Tests
#20 opened by Alvant - 0
Search methods: save/load + KeyboardInterrupt
#36 opened by Alvant - 0
Add Holdout perplexity score
#25 opened by Alvant - 0
Topic bank approach. Simple
#6 opened by Alvant - 0
- 2
Add renormalization & entropy score in run_search
#10 opened by Alvant - 1
Add (top-tokens + intratext) coherence scores
#13 opened by Alvant - 0
Method to compute several scores together
#9 opened by Alvant - 0
Refactor mean-std-computation stuff in scores
#12 opened by Alvant - 1
Entropy-based approach
#4 opened by Alvant