modelling code

######## Create sample weights

import numpy as np

def CreateBalancedSampleWeights(y_train, largest_class_weight_coef):
    classes = y_train.unique()
    classes.sort()
    class_samples = np.bincount(y_train)
    total_samples = class_samples.sum()
    n_classes = len(class_samples)
    weights = total_samples / (n_classes * class_samples * 1.0)
    class_weight_dict = {key : value for (key, value) in zip(classes, weights)}
    class_weight_dict[classes[1]] = class_weight_dict[classes[1]] * largest_class_weight_coef
    sample_weights = [class_weight_dict[y] for y in y_train]

    return sample_weights
    
train_sample_weight = CreateBalancedSampleWeights(label_train, largest_class_weight_coef=some_coef) ## some_coef is the occurence of majority class
  
  
clf.fit(X_train, label_train, sample_weight=train_sample_weight, eval_metric=eval_metric)