We can evaluate the performance of an algorithm of a classification problem using the logistic loss function. The logistic loss of a classification algorithm is given by the following formula:
![Log\ Loss=\frac{-1}{N}\sum_{i=1}^{N}[y_i\log p_i+(1-y_i)\log (1-p_i)]\\where\ N=the\ sample\ size\\y_i=the\ actual\ class\\p_i=the\ probability\ of\ the\ actual\ class\ y_i](https://s0.wp.com/latex.php?latex=Log%5C+Loss%3D%5Cfrac%7B-1%7D%7BN%7D%5Csum_%7Bi%3D1%7D%5E%7BN%7D%5By_i%5Clog+p_i%2B%281-y_i%29%5Clog+%281-p_i%29%5D%5C%5Cwhere%5C+N%3Dthe%5C+sample%5C+size%5C%5Cy_i%3Dthe%5C+actual%5C+class%5C%5Cp_i%3Dthe%5C+probability%5C+of%5C+the%5C+actual%5C+class%5C+y_i+&bg=ffffff&fg=000&s=0&c=20201002)
We can use the following Python code to calculate log loss for a classification problem.
from sklearn.model_selection import KFold
from sklearn.linear_model import LogisticRegression
from sklearn.model_selection import cross_val_score
import pandas
data = pandas.read_csv("diabetes.csv")
D = data.values
X = D[:, :-1]
y = D[:, -1]
k_fold = KFold(n_splits=10, shuffle=True, random_state=1)
classifier = LogisticRegression(solver="liblinear")
results = cross_val_score(classifier, X, y, cv=k_fold, scoring="neg_log_loss")
mean_score = results.mean()
print("Negative Log Loss: ", mean_score)
Here, we are first using the pandas Python library to read the Pima Indians Diabetes dataset. The dataset contains various …
0 Comments