reset_learning_rate->reset_parameter

[wxchan]

refer to #129

@guolinke I test some other parameters, it is indeed also not efficient.

[msftclas]

Hi @wxchan, I'm your friendly neighborhood Microsoft Pull Request Bot (You can call me MSBOT). Thanks for your contribution!
You've already signed the contribution license agreement. Thanks!

The agreement was validated by Microsoft and real humans are currently evaluating your PR.

TTYL, MSBOT;

[wxchan]

@guolinke lambdarank task causes segfault, it should be some bug. Try following examples:

def test_lambdarank():
    from sklearn.datasets import load_svmlight_file
    X_train, y_train = load_svmlight_file('../../examples/lambdarank/rank.train')
    X_test, y_test = load_svmlight_file('../../examples/lambdarank/rank.test')
    q_train = np.loadtxt('../../examples/lambdarank/rank.train.query')
    q_test= np.loadtxt('../../examples/lambdarank/rank.test.query')
    lgb_model = lgb.LGBMRanker().fit(X_train, y_train, group=q_train, eval_at=[1])
    preds = lgb_model.predict(X_test)

[guolinke]

@wxchan fixed, you can check again, and add it to the test.

I will optimize the reset parameter logic to make it faster.