How to prepare a one-hot encoding in scikit-learn for a multiclass logistic regression?

Question

I am trying to classify 4 classes from the following DataFrame using one-hot encoding in scikit-learn:

          K   T_STAR                 REGIME
15   90.929  0.95524  BoilingInducedBreakup
9   117.483  0.89386                 Splash
16   97.764  1.17972  BoilingInducedBreakup
13   76.917  0.91399  BoilingInducedBreakup
6    44.889  0.95725  BoilingInducedBreakup
20  151.662  0.56287                 Splash
12   67.155  1.22842     ReboundWithBreakup
7   114.747  0.47618                 Splash
17  121.731  0.52956                 Splash
12   29.397  0.88702             Deposition
14   31.733  0.69154             Deposition
13  119.433  0.39422                 Splash
21   97.913  1.21309     ReboundWithBreakup
10  117.544  0.18538                 Splash
27   76.957  0.52879             Deposition
22  155.842  0.17559                 Splash
3    25.620  0.18680             Deposition
30  151.773  1.23027     ReboundWithBreakup
34   91.146  0.90138             Deposition
19   58.095  0.46110             Deposition
14   85.596  0.97520  BoilingInducedBreakup
41   97.783  0.16985             Deposition
0    16.683  0.99355             Deposition
28  122.022  1.22977     ReboundWithBreakup
0    25.570  1.24686     ReboundWithBreakup
3   113.315  0.48886                 Splash
7    31.873  1.30497     ReboundWithBreakup
0   108.488  0.73423                 Splash
2    25.725  1.29953     ReboundWithBreakup
37   97.695  0.50930             Deposition

Here is the sample as CSV:

,K,T_STAR,REGIME
15,90.929,0.95524,BoilingInducedBreakup
9,117.483,0.89386,Splash
16,97.764,1.17972,BoilingInducedBreakup
13,76.917,0.91399,BoilingInducedBreakup
6,44.889,0.95725,BoilingInducedBreakup
20,151.662,0.56287,Splash
12,67.155,1.22842,ReboundWithBreakup
7,114.747,0.47618,Splash
17,121.731,0.52956,Splash
12,29.397,0.88702,Deposition
14,31.733,0.69154,Deposition
13,119.433,0.39422,Splash
21,97.913,1.21309,ReboundWithBreakup
10,117.544,0.18538,Splash
27,76.957,0.52879,Deposition
22,155.842,0.17559,Splash
3,25.62,0.1868,Deposition
30,151.773,1.23027,ReboundWithBreakup
34,91.146,0.90138,Deposition
19,58.095,0.4611,Deposition
14,85.596,0.9752,BoilingInducedBreakup
41,97.783,0.16985,Deposition
0,16.683,0.99355,Deposition
28,122.022,1.22977,ReboundWithBreakup
0,25.57,1.24686,ReboundWithBreakup
3,113.315,0.48886,Splash
7,31.873,1.30497,ReboundWithBreakup
0,108.488,0.73423,Splash
2,25.725,1.29953,ReboundWithBreakup
37,97.695,0.5093,Deposition

Features vector is two-dimensional (K,T_STAR) and REGIMES are the categories, that are not ordered in any way.

This is what I did so far for one-hot encoding and scaling:

from sklearn.compose import ColumnTransformer
from sklearn.preprocessing import MinMaxScaler 
from sklearn.preprocessing import OneHotEncoder 
num_attribs = ["K", "T_STAR"] 
cat_attribs = ["REGIME"]
preproc_pipeline = ColumnTransformer([("num", MinMaxScaler(), num_attribs),
                                      ("cat", OneHotEncoder(),  cat_attribs)])
regimes_df_prepared = preproc_pipeline.fit_transform(regimes_df)

However, when I print a few first lines of regimes_df_prepared I get

array([[0.73836403, 0.19766192, 0.        , 0.        , 0.        ,
        1.        ],
       [0.43284301, 0.65556065, 1.        , 0.        , 0.        ,
        0.        ],
       [0.97076007, 0.93419198, 0.        , 0.        , 1.        ,
        0.        ],
       [0.96996242, 0.34623652, 0.        , 0.        , 0.        ,
        1.        ],
       [0.10915571, 1.        , 0.        , 0.        , 1.        ,
        0.        ]])

So the one-hot encoding seems to have worked, but the problem is that the feature vectors are packed together with the encoding in this array.

If I try to train the model like this:

from sklearn.linear_model import LogisticRegression

logreg_ovr = LogisticRegression(solver='lbfgs', max_iter=10000, multi_class='ovr')
logreg_ovr.fit(regimes_df_prepared, regimes_df["REGIME"])
print("Model training score : %.3f" % logreg_ovr.score(regimes_df_prepared, regimes_df["REGIME"]))

The score is 1.0, which can't be (overfitting?).

Now I want the model to predict a category at a (K, T_STAR) pair

logreg_ovr.predict([[40,0.6]])

And I get an error

ValueError: X has 2 features per sample; expecting 6

as suspected, the model sees the entire row of regimes_df_prepared as a feature vector. How can I avoid this?

Shihab Shahriar Khan Shihab Shahriar Khan · Accepted Answer · 2020-04-02T14:21:47

Target labels should not be one-hot encoded, sklearn have LabelEncoder for that. In your case, the working code for data preprocessing would be something like:

X,y = regimes_df[num_attribs].values,regimes_df['REGIME'].values
y = LabelEncoder().fit_transform(y)

I have noticed that you're calculating score on same data used to train the model, which will naturally lead to overfitting. Please use something like train_test_split or cross_val_score to properly evaluate the performance of your model.

How to prepare a one-hot encoding in scikit-learn for a multiclass logistic regression?

1 Answers