Changelogs » Lime

Lime

0.1.1.36

- Added mask_string parameter to lime_text, allow user to control how tokens are masked.
- Fixed bug in truncnorm sampling where min == max

0.1.1.35

- Added sparse support for LimeTabularExplainer (thanks imatiach-msft)
- Changed undiscretize function for LimeTabularExplainer, now using truncated normal
- Minor fixes: re.split in python3.7, submodular pick on non-binary tasks, others

0.1.1.33

- LimeTabularExplainer accepts statistics rather than a dataset now
- various small fixes

0.1.1.32


      

0.1.1.31


      

0.1.1.26


      

0.1.1.25

Minor fixes.

0.1.1.24

EntropyDiscretizer did not work due to imports.

0.1.1.23

- bug in LimeImage
- bug where predict proba doesn't show

0.1.1.22

Added regression, other minor changes.

0.1.1.21

minor only

0.1.1.20

kgullikson88 added Recurrent Tabular Explainer and example.

0.1.1.19

Fixed some things in tutorials, added images.

0.1.1.18


      

0.1.1.17

- Added more discretization options
- Tutorials work with newest version of sklearn
- Default value of kernel_width scaled by number of columns in training data
- Minor changes

0.1.1.16

Fixed a bug in save_to_file related to encoding.

0.1.1.15

I'll start tracking releases more carefully from now on, and keeping this consistent with pypi releases.
This is the first release : )