How to evaluate XGBoost model with learning curves example 2?
How to evaluate XGBoost model with learning curves example 2?
Recipe Objective
While training a dataset sometimes we need to know how model is training with each row of data passed through it. Sometimes while training a very large dataset it takes a lots of time and for that we want to know that after passing speicific percentage of dataset what is the score of the model. So this can be done by learning curve. So here we are evaluating XGBoost with learning curves.
So this recipe is a short example of how we can visualise XGBoost model with learning curves.
Step 1 - Import the library
from numpy import loadtxt
from xgboost import XGBClassifier
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score
from matplotlib import pyplot
import matplotlib.pyplot as plt
plt.style.use("ggplot")
Here we have imported various modules like datasets, XGBClassifier and learning_curve from differnt libraries. We will understand the use of these later while using it in the in the code snippet.
For now just have a look on these imports.
Step 2 - Setup the Data
Here we have used datasets to load the inbuilt wine dataset and we have created objects X and y to store the data and the target value respectively.
dataset = loadtxt("pima.indians.diabetes.data.csv", delimiter=",")
X = dataset[:,0:8]
Y = dataset[:,8]
X_train, X_test, y_train, y_test = train_test_split(X, Y, test_size=0.33, random_state=7)
Step 3 - Model and Result
Here we are training XGBClassifier() and calculated the accuracy and the epochs.
model = XGBClassifier()
eval_set = [(X_train, y_train), (X_test, y_test)]
model.fit(X_train, y_train, eval_metric=["error", "logloss"], eval_set=eval_set, verbose=False)
y_pred = model.predict(X_test)
predictions = [round(value) for value in y_pred]
accuracy = accuracy_score(y_test, predictions)
print("Accuracy: %.2f%%" % (accuracy * 100.0))
results = model.evals_result()
epochs = len(results["validation_0"]["error"])
x_axis = range(0, epochs)
Step 4 - Ploting the Log loss and classification error
Finally, its time to plot the Log loss and classification error. We have used matplotlib to plot lines.
# plot log loss
fig, ax = pyplot.subplots(figsize=(12,12))
ax.plot(x_axis, results["validation_0"]["logloss"], label="Train")
ax.plot(x_axis, results["validation_1"]["logloss"], label="Test")
ax.legend()
pyplot.ylabel("Log Loss")
pyplot.title("XGBoost Log Loss")
pyplot.show()
# plot classification error
fig, ax = pyplot.subplots(figsize=(12,12))
ax.plot(x_axis, results["validation_0"]["error"], label="Train")
ax.plot(x_axis, results["validation_1"]["error"], label="Test")
ax.legend()
pyplot.ylabel("Classification Error")
pyplot.title("XGBoost Classification Error")
pyplot.show()
As an output we get:
Download Materials
