Issues
- 1
Ship small datasets with the repo
#22 opened by tttthomasssss - 1
Reduce memory footprint
#34 opened by tttthomasssss - 1
Support `sacred`
#21 opened by tttthomasssss - 1
Check `feather`
#20 opened by tttthomasssss - 1
Automatic wikipedia extraction
#1 opened by tttthomasssss - 1
More flexible Serialisation
#28 opened by tttthomasssss - 0
- 2
Documentation & Examples
#3 opened by tttthomasssss - 0
Check the N/d normalisation term in PPMI
#32 opened by tttthomasssss - 0
Meaningful logging levels
#31 opened by tttthomasssss - 0
Test Coverage
#15 opened by tttthomasssss - 1
Add package to PyPI
#16 opened by tttthomasssss - 0
Be explicit about log bases
#12 opened by tttthomasssss - 0
- 1
- 1
add NMF to dim_reduction
#30 opened by tttthomasssss - 0
Offer some base settings
#24 opened by tttthomasssss - 1
Support for other PMI variants
#5 opened by tttthomasssss - 1
PPMI option to add context vectors
#23 opened by tttthomasssss - 1
- 0
Support for sampled window size
#27 opened by tttthomasssss - 0
Add extrinsic evaluation
#25 opened by tttthomasssss - 0
Implement subsampling
#8 opened by tttthomasssss - 0
ad Incremental `fit()` options
#18 opened by tttthomasssss - 2
- 3
Rethinking serialisation/deserialisation
#17 opened by tttthomasssss - 1
incremental `fit()` options
#13 opened by tttthomasssss - 0
- 0
- 0
Implement `transform`
#7 opened by tttthomasssss - 0
- 0
Optimise co-occurrence matrix construction
#14 opened by tttthomasssss - 0
Support asymmetric context windows
#6 opened by tttthomasssss - 0
Fix SPPMI
#4 opened by tttthomasssss