George Walker comments

Results 7 comments of


                                            George Walker

Direct using of categorical features

I can see from the regression example [here](https://github.com/ArdalanM/pyLightGBM/blob/master/notebooks/regression_example_kaggle_allstate.ipynb) that you integerise categorical features and then include these with continuous variables: ``` le = LabelEncoder() for col in df.columns.tolist(): if 'cat'...

raise StringLengthException if vectoriser is applied to strings that …

@ParticularMiner i have updated the test and code but obviously adding this regex replacement has a computational cost, interested what you think

raise StringLengthException if vectoriser is applied to strings that …

@ParticularMiner how is this? I can squash commits if required

how to handle 'ValueError: empty vocabulary; perhaps the documents only contain stop words' in group_similar_strings

please see https://github.com/Bergvca/string_grouper/pull/67

Tone deaf package name?

this still needs to be changed!

process.extract() with scorer partial_ratio returns wrong results

in python 3.7, a shorter example that gives the same result: ``` >>> fuzz.partial_ratio("thane", "t hosa na e thane ws") 40 ```

package is missing a licence

https://github.com/Meteorix/pylcs/pull/11