add verbose to EnsembleClassifier

[albahnsen]

Relater to #6
Add verbose in 4 levels:

`verbose=0` (default): prints nothing like right now
`verbose=1`: Print the number of the clf that is being fitted
`verbose=2`: Print info about the parameters of the clf that is being fitted
`verbose>2`:change the verbose of the  underlying clf to self.verbose - 2 

Example:

>>> import numpy as np
>>> from sklearn.linear_model import LogisticRegression
>>> from sklearn.naive_bayes import GaussianNB
>>> from sklearn.ensemble import RandomForestClassifier
>>> from mlxtend.sklearn import EnsembleClassifier
>>> clf1 = LogisticRegression(random_state=1)
>>> clf2 = RandomForestClassifier(random_state=1)
>>> clf3 = GaussianNB()
>>> X = np.array([[-1, -1], [-2, -1], [-3, -2], [1, 1], [2, 1], [3, 2]])
>>> y = np.array([1, 1, 1, 2, 2, 2])
>>> eclf1 = EnsembleClassifier(clfs=[clf1, clf2, clf3], voting='hard', verbose=3)
>>> eclf1 = eclf1.fit(X, y)
>>> print(eclf1.predict(X))
Fitting 3 classifiers...
Fitting clf1: logisticregression (1/3)
LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True,
          intercept_scaling=1, max_iter=100, multi_class='ovr',
          penalty='l2', random_state=1, solver='liblinear', tol=0.0001,
          verbose=0)
[LibLinear]iter  1 act 2.675e+00 pre 2.380e+00 delta 7.534e-01 f 4.159e+00 |g| 7.211e+00 CG   1
iter  2 act 2.681e-01 pre 2.350e-01 delta 7.534e-01 f 1.484e+00 |g| 1.577e+00 CG   1
iter  3 act 1.622e-02 pre 1.555e-02 delta 7.534e-01 f 1.216e+00 |g| 3.205e-01 CG   1
iter  4 act 4.876e-04 pre 4.874e-04 delta 7.534e-01 f 1.200e+00 |g| 3.435e-02 CG   2
Fitting clf2: randomforestclassifier (2/3)
RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini',
            max_depth=None, max_features='auto', max_leaf_nodes=None,
            min_samples_leaf=1, min_samples_split=2,
            min_weight_fraction_leaf=0.0, n_estimators=10, n_jobs=1,
            oob_score=False, random_state=1, verbose=0, warm_start=False)
[Parallel(n_jobs=1)]: Done   1 jobs       | elapsed:    0.0s
[Parallel(n_jobs=1)]: Done  10 out of  10 | elapsed:    0.0s finished
[Parallel(n_jobs=1)]: Done   1 jobs       | elapsed:    0.0s
[Parallel(n_jobs=1)]: Done  10 out of  10 | elapsed:    0.0s finished
Fitting clf3: gaussiannb (3/3)
GaussianNB()
[1 1 1 2 2 2]

[albahnsen]

@rasbt let me know what you think. I realize that is was important to have a different level of verbose between the ensemble function and the underlying clf. other than that, is what we discussed in #6

[coveralls]

Coverage decreased (-0.29%) to 63.71% when pulling 51e751d on albahnsen:master into d895a04 on rasbt:master.

[rasbt]

@albahnsen That looks pretty good, thanks! I think there it only needs a tiny fix for the print statement before I can merge it.

[albahnsen]

@rasbt happy to help 😄. to which fix are you referring?

[rasbt]

@albahnsen You can find the error when you click on the "Details" button for the travis-ci check. I think you just forgot the parentheses to make it compatible to Python 3:

 File "/home/travis/build/rasbt/mlxtend/mlxtend/sklearn/ensemble.py", line 128
    print _name_estimators((clf,))[0][1]
                      ^
SyntaxError: invalid syntax

[coveralls]

Coverage decreased (-0.38%) to 63.62% when pulling 6bd0dce on albahnsen:master into d895a04 on rasbt:master.

[rasbt]

Looks great, thanks a lot! I just made some tiny changes so that the verbose parameter of the individual classifiers is not set for the original ones but their clones.
Also, I swapped the order of

if self.verbose > 1:

and

if self.verbose > 2: 

so that the verbose parameter is shown correctly when printing out the params, e.g,.

>>> eclf1 = EnsembleClassifier(clfs=[clf1, clf2, clf3], voting='hard', verbose=3)
>>> eclf1 = eclf1.fit(X, y)
>>> print(eclf1.predict(X))


Fitting 3 classifiers...
Fitting clf1: logisticregression (1/3)
LogisticRegression(...
      verbose=1)
...

[albahnsen]

That's great. Thanks.