# Import the neccessary modules for data manipulation and visual representation import pandas as pd import numpy as np import matplotlib.pyplot as plt import matplotlib as matplot import seaborn as sns %matplotlib inline

df = pd.read_csv('HR_comma_sep.csv.txt')

# Examine the dataset df.head()

# Can you check to see if there are any missing values in our data set df.isnull().any()

# Rename Columns # Renaming certain columns for better readability df = df.rename(columns={'satisfaction_level': 'satisfaction', 'last_evaluation': 'evaluation', 'number_project': 'projectCount', 'average_montly_hours': 'averageMonthlyHours', 'time_spend_company': 'yearsAtCompany', 'Work_accident': 'workAccident', 'promotion_last_5years': 'promotion', 'sales' : 'department', 'left' : 'turnover' }) df.head(3)

# Check the type of our features. Are there any data inconsistencies? df.dtypes

# How many employees are in the dataset? df.shape

# Calculate the turnover rate of our company's dataset. What's the rate of turnover? turnover_rate = df.turnover.value_counts() / 14999 turnover_rate

# Display the statistical overview of the employees df.describe()

# Display the mean summary of Employees (Turnover V.S. Non-turnover). What do you notice between the groups? turnover_Summary = df.groupby('turnover') turnover_Summary.mean()

# Create a correlation matrix. What features correlate the most with turnover? What other correlations did you find? corr = df.corr() corr = (corr) sns.heatmap(corr, xticklabels=corr.columns.values, yticklabels=corr.columns.values) plt.title('Heatmap of Correlation Matrix') corr

# Plot the distribution of Employee Satisfaction, Evaluation, and Project Count. What story can you tell? # Set up the matplotlib figure f, axes = plt.subplots(ncols=3, figsize=(15, 6)) # Graph Employee Satisfaction sns.distplot(df.satisfaction, kde=False, color="g", ax=axes[0]).set_title('Employee Satisfaction Distribution') axes[0].set_ylabel('Employee Count') # Graph Employee Evaluation sns.distplot(df.evaluation, kde=False, color="r", ax=axes[1]).set_title('Employee Evaluation Distribution') axes[1].set_ylabel('Employee Count') # Graph Employee Average Monthly Hours sns.distplot(df.averageMonthlyHours, kde=False, color="b", ax=axes[2]).set_title('Employee Average Monthly Hours Distribution') axes[2].set_ylabel('Employee Count')

cat_var = ['department','salary','turnover','promotion'] num_var = ['satisfaction','evaluation','projectCount','averageMonthlyHours','yearsAtCompany', 'workAccident'] categorical_df = pd.get_dummies(df[cat_var], drop_first=True) numerical_df = df[num_var] new_df = pd.concat([categorical_df,numerical_df], axis=1) new_df.head()

from sklearn.model_selection import train_test_split from sklearn.metrics import accuracy_score, classification_report, precision_score, recall_score, confusion_matrix, precision_recall_curve

# Create the X and y set X = new_df.iloc[:,1:] y = new_df.iloc[:,0] # Define train and test X_train, X_test, y_train, y_test = train_test_split(X,y,test_size=0.15, random_state=123, stratify=y)

%%time # Check accuracy of Logistic Model# Check  from sklearn.linear_model import LogisticRegression # Define the Logistic Regression Model lr = LogisticRegression(class_weight='balanced') # Fit the Logistic Regression Model to the train set lr.fit(X_train, y_train) print ("Logistic accuracy is %2.2f" % accuracy_score(y_test, lr.predict(X_test)))

%%time from sklearn import model_selection # Define the 10-Fold Cross Validation kfold = model_selection.KFold(n_splits=10, random_state=7) # Define the Logistic Regression Model lrCV = LogisticRegression() # Define the evaluation metric  scoring = 'roc_auc' # Train the Logistic Regression Model on the 10-Fold Cross Validation lr_results = model_selection.cross_val_score(lrCV, X_train, y_train, cv=kfold, scoring=scoring)

# Print out the 10 scores from the training. Notice how you get a wide range of scores compared to one single training lr_results

# Print out the mean and standard deviation of the training score lr_auc = lr_results.mean() print("The Logistic Regression AUC: %.3f and the STD is (%.3f)" % (lr_auc, lr_results.std()))

from sklearn.metrics import roc_auc_score print ("\n\n ---Logistic Regression Model---") lr_auc = roc_auc_score(y_test, lr.predict(X_test)) print ("Logistic Regression AUC = %2.2f" % lr_auc) print(classification_report(y_test, lr.predict(X_test)))

%%time from sklearn.ensemble import RandomForestClassifier # Random Forest Model rf = RandomForestClassifier( class_weight="balanced" ) # Fit the RF Model rf = rf.fit(X_train, y_train)

%%time rf_results = model_selection.cross_val_score(rf, X_train, y_train, cv=kfold, scoring=scoring) rf_results

# Print out the mean and standard deviation of the training score rf_auc = rf_results.mean() print("The Random Forest AUC: %.3f and the STD is (%.3f)" % (rf_auc, rf_results.std()))

from sklearn.metrics import roc_auc_score print ("\n\n ---Random Forest Model---") rf_roc_auc = roc_auc_score(y_test, rf.predict(X_test)) print ("Random Forest AUC = %2.2f" % rf_roc_auc) print(classification_report(y_test, rf.predict(X_test)))

%%time from sklearn.svm import SVC svclassifier = SVC(kernel='rbf', probability=True) svc = svclassifier.fit(X_train,y_train)

%%time svc_result = model_selection.cross_val_score(svc, X_train, y_train, cv=kfold, scoring=scoring) svc_result

# Print out the mean and standard deviation of the training score svc_auc = svc_result.mean() print("The Supper Vector Classifier AUC: %.3f and the STD is (%.3f)" % (rf_auc, rf_results.std()))

from sklearn.metrics import roc_auc_score print ("\n\n ---Support Vector Model---") rf_roc_auc = roc_auc_score(y_test, rf.predict(X_test)) print ("Support Vector Classifier AUC = %2.2f" % rf_roc_auc) print(classification_report(y_test, rf.predict(X_test)))

# Create ROC Graph from sklearn.metrics import roc_curve fpr, tpr, thresholds = roc_curve(y_test, lr.predict_proba(X_test)[:,1]) rf_fpr, rf_tpr, rf_thresholds = roc_curve(y_test, rf.predict_proba(X_test)[:,1]) svc_fpr, svc_tpr, svc_thresholds = roc_curve(y_test, svc.predict_proba(X_test)[:,1]) plt.figure() # Plot Logistic Regression ROC plt.plot(fpr, tpr, label='Logistic Regression (area = %0.2f)' % lr_auc) # Plot Random Forest ROC plt.plot(rf_fpr, rf_tpr, label='Random Forest (area = %0.2f)' % rf_auc) # Plot Decision Tree ROC plt.plot(svc_fpr, svc_tpr, label='Support Vector Classifier (area = %0.2f)' % svc_auc) # Plot Base Rate ROC plt.plot([0,1], [0,1],label='Base Rate' 'k--') plt.xlim([0.0, 1.0]) plt.ylim([0.0, 1.05]) plt.xlabel('False Positive Rate') plt.ylabel('True Positive Rate') plt.title('ROC Graph') plt.legend(loc="lower right") plt.show()

# Get Feature Importances feature_importances = pd.DataFrame(rf.feature_importances_, index = X_train.columns, columns=['importance']).sort_values('importance', ascending=False) feature_importances = feature_importances.reset_index() feature_importances

sns.set(style="whitegrid") # Initialize the matplotlib figure f, ax = plt.subplots(figsize=(13, 7)) # Plot the total schools per city sns.set_color_codes("pastel") sns.barplot(x="importance", y='index', data=feature_importances, label="Total", color="b")

rf.predict_proba(X_test)[175:200,]