search

fabianp / mord

Ordinal regression algorithms
Other
263 stars 71 forks source link

Pandas has deprecated ".ix", so it has to be replaced with "iloc" in … #24

Closed thomas-haslwanter closed 3 years ago

thomas-haslwanter commented 3 years ago

…'base.py'

Also, are you sure the code for the example is correct? When I inspect the values of

clf4.predict(X) in housing.py, line 47, I get all "1"s predicted. Nevertheless, the accuracy is better than some of the other predictions.

fabianp commented 3 years ago

looks good to me

thomas-haslwanter commented 3 years ago

Fabian, one more question: How does one have to update https://github.com/fabianp/minirank/blob/master/minirank/logistic.py to make it work with "mord"?

The structure of sklearn has changed, and from sklearn import cross_validation no longer works.

fabianp commented 3 years ago

I tried something simple, https://github.com/fabianp/minirank/commit/3693a1f7ae468dc18506684b5df7e3e194bcb65f hopefully that works

thomas-haslwanter commented 3 years ago

Nope, it spit the dummy at ShuffleSplit Given your warning in https://github.com/fabianp/minirank it would probably be better to adapt that code to use 'mord' in "minirank". The reason I am working on this: I used your code in my book Introduction to Statistics with Python with the code available in https://github.com/thomas-haslwanter/statsintro_python/tree/master/ISP/Code_Quantlets/13_LogisticRegression/OrdinalLogisticRegression

And since Springer has asked me for a 2nd Ed, I am updating it. With your permission, I would like to again use your figures image for that book. Would that be OK?

fabianp commented 3 years ago

sure

On Tue, Jan 12, 2021 at 8:49 AM Thomas Haslwanter notifications@github.com wrote:

Nope, it spit the dummy at ShuffleSplit Given your warning in https://github.com/fabianp/minirank it would probably be better to adapt that code for use in "minirank". The reason I am working on this: I used your code in my book Introduction to Statistics with Python https://www.springer.com/de/book/9783319283159 with the code available in

https://github.com/thomas-haslwanter/statsintro_python/tree/master/ISP/Code_Quantlets/13_LogisticRegression/OrdinalLogisticRegression

And since Springer has asked me for a 2nd Ed, I am updating it. With your permission, I would like to again use your figures [image: image] https://user-images.githubusercontent.com/1530437/104322493-54b00d80-54e5-11eb-9470-4b2217d24383.png for that book. Would that be OK?

— You are receiving this because you modified the open/close state. Reply to this email directly, view it on GitHub https://github.com/fabianp/mord/pull/24#issuecomment-758666476, or unsubscribe https://github.com/notifications/unsubscribe-auth/AACDZB6LWW462B3IVL7B4T3SZRHPHANCNFSM4V7FAGSA .

thomas-haslwanter commented 3 years ago

Thank you!