Develop Reference

How to convert XGBoost model SHAP values from log odds to probabilities?

I trained an XGBoost Classifier and am trying to generate SHAP contributions in probabilities. I understand the output of shap.TreeExplainer for XGBoost models is in log odds ratios. I expect the expected_value of explainer to be equal or close to the average predicted value in the dataset. However, I get exptected_value of -2.7776 (explainer.expected_value), which corresponds to probability of 0.0585 (expit(-2.7776)). This is significantlty lower compared to average predicted score of 0.21. Is there any step I am missing in converting expected value to probability?
# Import libraries
import numpy as np
from sklearn.ensemble import GradientBoostingClassifier
from sklearn.datasets import make_classification
from sklearn.model_selection import train_test_split
from sklearn.metrics import confusion_matrix, classification_report
import shap
from scipy.special import expit
# Generate data
X, Y = make_classification(n_samples=10000,
n_features=20,
n_redundant=0,
n_classes=2,
random_state=17,
weights = [0.8, 0.2])
# Split into train and test
x_train, x_test, y_train, y_test = train_test_split(X, Y, test_size=0.1, random_state=7)
# Data check
print('Target rate: {:.0%}'.format(sum(Y)/len(Y)))
print('Target rate in train dataset: {:.0%}'.format(sum(y_train)/len(y_train)))
print('Target rate in test dataset: {:.0%}'.format(sum(y_test)/len(y_test)))
print('Total observations: {:.0f}'.format(len(X)))
print('Train observations: {:.0f}'.format(len(x_train)))
print('Test observations: {:.0f}'.format(len(x_test)))
# Train XGBoost model
model = GradientBoostingClassifier(
n_estimators = 50,
max_depth = 3,
random_state = 17
)
model.fit(x_train, y_train)
# Get accuracy score and confusion matrix for train and test datasets
# There doesn't seem to be issues with model performance, it is pretty close for train and test datasets
acc_train = model.score(x_train, y_train)
acc_test = model.score(x_test, y_test)
cm_train = confusion_matrix(y_train, y_pred_class_train, normalize = 'true')
cl_report_train = classification_report(y_train, y_pred_class_train)
cm_test = confusion_matrix(y_test, y_pred_class_test, normalize = 'true')
cl_report_test = classification_report(y_test, y_pred_class_test)
# Print results
print('MODEL ACCURACY:\n \
training data: {:.2%}\n \
test data: {:.2%}'.format(acc_train, acc_test))
print('\nCONFUSION MATRIX (train data):\n {}'.format(cm_train.round(3)))
print('\nCLASSIFICATION REPORT (train data):\n {}'.format(cl_report_train))
print('\nCONFUSION MATRIX (test data):\n {}'.format(cm_test.round(3)))
print('\nCLASSIFICATION REPORT (test data):\n {}'.format(cl_report_test))
# Check average predicted score
# Train
y_pred_prob_train = model.predict_proba(x_train)
y_pred_class_train = model.predict(x_train)
print('Train: Average predicted score: {:.2%}'.format(np.mean(y_pred_prob_train[:,1])))
# Test
y_pred_prob_test = model.predict_proba(x_test)
y_pred_class_test = model.predict(x_test)
print('Test: Average predicted score: {:.2%}'.format(np.mean(y_pred_prob_test[:,1])))
# Get SHAP values in log odds for test dataset
explainer = shap.TreeExplainer(model)
shap_values = explainer.shap_values(x_test)
# Check SHAP expected value
print('SHAP expected value: {:.4f}'.format(explainer.expected_value[0]))
print('SHAP expected value transformed: {:.4f}'.format(expit(explainer.expected_value[0])))
print('Average predicted value: {:.4f}'.format(np.mean(y_pred_prob_test[:,1])))
# Average predicted value is ~ 0.21 while shap expected value only ~ 0.06.

How to evaluate the effect of different methods of handling missing values?

I am a total beginner and I am trying to compare different methods of handling missing data. In order to evaluate the effect of each method (drop raws with missing values, drop columns with missigness over 40%, impute with the mean, impute with the KNN), I compare the results of the LDA accuracy and LogReg accuracy on the training set between a dataset with 10% missing values, 20% missing values against the results of the original complete dataset. Unfortunately, I get pretty much the same results even between the complete dataset and the dataset with 20% missing-ness. I don't know what I am doing wrong.
from numpy import nan
from numpy import isnan
from pandas import read_csv
from sklearn.impute import SimpleImputer
from sklearn.discriminant_analysis import LinearDiscriminantAnalysis
from sklearn.model_selection import KFold
from sklearn.model_selection import cross_val_score
#dataset = read_csv('telecom_churn_rev10.csv')
dataset = read_csv('telecom_churn_rev20.csv')
dataset = dataset.replace(nan, 0)
values = dataset.values
X = values[:,1:11]
y = values[:,0]
dataset.fillna(dataset.mean(), inplace=True)
#dataset.fillna(dataset.mode(), inplace=True)
print(dataset.isnull().sum())
imputer = SimpleImputer(missing_values = nan, strategy = 'mean')
transformed_values = imputer.fit_transform(X)
print('Missing: %d' % isnan(transformed_values).sum())
model = LinearDiscriminantAnalysis()
cv = KFold(n_splits = 3, shuffle = True, random_state = 1)
result = cross_val_score(model, X, y, cv = cv, scoring = 'accuracy')
print('Accuracy: %.3f' % result.mean())
#print('Accuracy: %.3f' % result.mode())
print(dataset.describe())
print(dataset.head(20))
from sklearn.model_selection import train_test_split
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2, random_state = 0)
from sklearn.preprocessing import StandardScaler
sc = StandardScaler()
X_train = sc.fit_transform(X_train)
X_test = sc.transform(X_test)
from sklearn.linear_model import LogisticRegression
classifier = LogisticRegression(random_state = 0)
classifier.fit(X_train, y_train)
y_pred = classifier.predict(X_test)
from sklearn.metrics import confusion_matrix, accuracy_score
cm = confusion_matrix(y_test, y_pred)
print(cm)
accuracy_score(y_test,y_pred)
from sklearn import metrics
# make predictions on X
expected = y
predicted = classifier.predict(X)
# summarize the fit of the model
print(metrics.classification_report(expected, predicted))
print(metrics.confusion_matrix(expected, predicted))
# make predictions on X test
expected = y_test
predicted = classifier.predict(X_test)
# summarize the fit of the model
print(metrics.confusion_matrix(expected, predicted))
print(metrics.classification_report(expected, predicted))

You replace all your missing values with 0 at that line : dataset = dataset.replace(nan, 0). After this line, you have a full dataset without missing values. So, the .fillna() and the SimpleImputer() are useless after that line.

MAE using Pipeline and GridSearchCV

I am facing a challenge finding Mean Average Error (MAE) using Pipeline and GridSearchCV
Background:
I have worked on a Data Science project (MWE as below) where a MAE value would be returned of a classifier as it's performance metric.
#Library
import pandas as pd
from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import train_test_split
from sklearn.metrics import mean_absolute_error
#Data import and preparation
data = pd.read_csv("data.csv")
data_features = ['location','event_type_count','log_feature_count','total_volume','resource_type_count','severity_type']
X = data[data_features]
y = data.fault_severity
#Train Validation Split for Cross Validation
X_train, X_valid, y_train, y_valid = train_test_split(X, y, train_size=0.8, test_size=0.2, random_state=0)
#RandomForest Modeling
RF_model = RandomForestClassifier(n_estimators=100, random_state=0)
RF_model.fit(X_train, y_train)
#RandomForest Prediction
y_predict = RF_model.predict(X_valid)
#MAE
print(mean_absolute_error(y_valid, y_predict))
#Output:
# 0.38727149627623564
Challenge:
Now I am trying to implement the same using Pipeline and GridSearchCV (MWE as below). The expectation is the same MAE value would be returned as above. Unfortunately I could not get it right using the 3 approaches below.
#Library
import pandas as pd
from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import train_test_split
from sklearn.pipeline import Pipeline
from sklearn.model_selection import GridSearchCV
#Data import and preparation
data = pd.read_csv("data.csv")
data_features = ['location','event_type_count','log_feature_count','total_volume','resource_type_count','severity_type']
X = data[data_features]
y = data.fault_severity
#Train Validation Split for Cross Validation
X_train, X_valid, y_train, y_valid = train_test_split(X, y, train_size=0.8, test_size=0.2, random_state=0)
#RandomForest Modeling via Pipeline and Hyper-parameter tuning
steps = [('rf', RandomForestClassifier(random_state=0))]
pipeline = Pipeline(steps) # define the pipeline object.
parameters = {'rf__n_estimators':[100]}
grid = GridSearchCV(pipeline, param_grid=parameters, scoring='neg_mean_squared_error', cv=None, refit=True)
grid.fit(X_train, y_train)
#Approach 1:
print(grid.best_score_)
# Output:
# -0.508130081300813
#Approach 2:
y_predict=grid.predict(X_valid)
print("score = %3.2f"%(grid.score(y_predict, y_valid)))
# Output:
# ValueError: Expected 2D array, got 1D array instead:
# array=[0. 0. 0. ... 0. 1. 0.].
# Reshape your data either using array.reshape(-1, 1) if your data has a single feature or array.reshape(1, -1) if it contains a single sample.
#Approach 3:
y_predict_df = pd.DataFrame(y_predict.reshape(len(y_predict), -1),columns=['fault_severity'])
print("score = %3.2f"%(grid.score(y_predict_df, y_valid)))
# Output:
# ValueError: Number of features of the model must match the input. Model n_features is 6 and input n_features is 1
Discussion:
Approach 1:
As in GridSearchCV() the scoring variable is set to neg_mean_squared_error, tried to read the grid.best_score_. But it did not get the same MAE result.
Approach 2:
Tried to get the y_predict values using grid.predict(X_valid). Then tried to get the MAE using grid.score(y_predict, y_valid) as the scoring variable in GridSearchCV() is set to neg_mean_squared_error. It returned a ValueError complaining "Expected 2D array, got 1D array instead".
Approach 3:
Tried to reshape y_predict and it did not work either. This time it returned "ValueError: Number of features of the model must match the input."
It would be helpful if you can assist to point where I could have made the error?
If you need, the data.csv is available at https://www.dropbox.com/s/t1h53jg1hy4x33b/data.csv
Thank you very much

You are trying to compare mean_absolute_error with neg_mean_squared_error which is very different refer here for more details. You should have used neg_mean_absolute_error in your GridSearchCV object creation like shown below:
grid = GridSearchCV(pipeline, param_grid=parameters,scoring='neg_mean_absolute_error', cv=None, refit=True)
Also, the score method in sklearn takes (X,y) as inputs, where x is your input feature of shape (n_samples, n_features) and y is the target labels, you need to change your grid.score(y_predict, y_valid) into grid.score(X_valid, y_valid).

How to compute "y_train_true, y_train_prob, y_test_true, y_test_prob"?

I have computed X_train, X_test, y_train, y_test. But I can not compute y_train_true, y_train_prob, y_test_true, y_test_prob.
How can I compute y_train_true, y_train_prob, y_test_true, y_test_prob from the following code ?
X_train:
X_test:
y_train:
y_test:
N.B,
y_train_true: True binary labels of 0 or 1 in the training dataset
y_train_prob: Probability in range {0,1} predicted by the model for the training dataset
y_test_true: True binary labels of 0 or 1 in the testing dataset
y_test_prob: Probability in range {0,1} predicted by the model for the testing dataset
Code :
# Split test and train data
import numpy as np
from sklearn.model_selection import train_test_split
X = np.array(dataset.ix[:, 1:10])
y = np.array(dataset['benign_malignant'])
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=0)
#Define Classifier and ====
from sklearn.neighbors import KNeighborsClassifier
knn = KNeighborsClassifier(n_neighbors = 5, metric = 'minkowski', p = 2)
# knn = KNeighborsClassifier(n_neighbors=11)
knn.fit(X_train, y_train)
# Predicting the Test set results
y_pred = knn.predict(X_train)

Well in your case y_train and y_test is already y_train_true and y_test_true. To get y_train_prob and y_test_prob, you need to take a model. I don't know which dataset you're using but it seems to be a binary classification problem so that you could use logistic regression to do this so,
from sklearn.neighbors import KNeighborsClassifier
knn = KNeighborsClassifier(n_neighbors = 5, metric = 'minkowski', p = 2)
knn.fit(X_train, y_train)
y_train_prob = knn.predict_proba(X_train)
y_test_prob = knn.predict_proba(X_test)

Large mean squared error in sklearn regressors

I'm a beginner in machine learning and I want to build a model to predict the price of houses. I prepared a dataset by crawling a local housing website and it consists 1000 samples and only 4 features (latitude, longitude, area and number of rooms).
I tried RandomForestRegressor and LinearSVR models in sklearn, but I can't train the model properly and the MSE is super high.
MSE almost equals 90,000,000 (the true values of prices' range are between 5,000,000 - 900,000,000)
Here is my code:
import numpy as np
from sklearn.svm import LinearSVR
import pandas as pd
import csv
from sklearn.preprocessing import normalize
from sklearn.model_selection import train_test_split
df = pd.read_csv('dataset.csv', index_col=False)
X = df.drop('price', axis=1)
X_data = X.values
Y_data = df.price.values
X_train, X_test, Y_train, Y_test = train_test_split(X_data, Y_data, test_size=0.2, random_state=5)
rgr = RandomForestRegressor(n_estimators=100)
svr = LinearSVR()
rgr.fit(X_train, Y_train)
svr.fit(X_train, Y_train)
MSEs = cross_val_score(estimator=rgr,
X=X_train,
y=Y_train,
scoring='mean_squared_error',
cv=5)
MSEsSVR = cross_val_score(estimator=svr,
X=X_train,
y=Y_train,
scoring='mean_squared_error',
cv=5)
MSEs *= -1
RMSEs = np.sqrt(MSEs)
print("Root mean squared error with 95% confidence interval:")
print("{:.3f} (+/- {:.3f})".format(RMSEs.mean(), RMSEs.std()*2))
print("")
Is the problem with my dataset and count of features? How can I build a prediction model with this type of dataset?