Upload 3 files

app_KPI_706.py

@@ -0,0 +1,159 @@
+
+import streamlit as st
+import pandas as pd
+import numpy as np
+import matplotlib.pyplot as plt
+import seaborn as sns
+import pickle
+
+# Load model and encoders
+with open('model_kpi_706.pkl', 'rb') as file:
+    model, department_encoder, region_encoder, education_encoder, gender_encoder, recruitment_channel_encoder = pickle.load(file)
+
+# Load your DataFrame
+# Replace 'your_data.csv' with the actual file name or URL
+df = pd.read_csv('Uncleaned_employees_final_dataset.csv')
+df = df.drop('employee_id', axis=1)
+
+# Streamlit App
+st.title('Employee KPIs App')
+
+# Define a session state to remember tab selections
+if 'tab_selected' not in st.session_state:
+    st.session_state.tab_selected = 0
+
+# Create tabs for prediction and visualization
+tabs = ['Predict KPIs', 'Visualize Data', 'Predict from CSV']
+selected_tab = st.radio('Select Tab:', tabs, index=st.session_state.tab_selected)
+
+# Tab selection logic
+if selected_tab != st.session_state.tab_selected:
+    st.session_state.tab_selected = tabs.index(selected_tab)
+
+# Tab 1: Predict KPIs
+if st.session_state.tab_selected == 0:
+    st.header('Predict KPIs')
+
+    # User Input Form
+    department = st.selectbox('Department', department_encoder.classes_)
+    region = st.selectbox('Region', region_encoder.classes_)
+    education = st.selectbox('Education', education_encoder.classes_)
+    gender = st.radio('Gender', gender_encoder.classes_)
+    recruitment_channel = st.selectbox('Recruitment Channel', recruitment_channel_encoder.classes_)
+    no_of_trainings = st.slider('Number of Trainings', 1, 10, 1)
+    age = st.slider('Age', 18, 60, 30)
+    previous_year_rating = st.slider('Previous Year Rating', 1.0, 5.0, 3.0)
+    length_of_service = st.slider('Length of Service', 1, 20, 5)
+    awards_won = st.checkbox('Awards Won')
+    avg_training_score = st.slider('Average Training Score', 40, 100, 70)
+
+    # Create a DataFrame for the user input
+    user_input = pd.DataFrame({
+        'department': [department],
+        'region': [region],
+        'education': [education],
+        'gender': [gender],
+        'recruitment_channel': [recruitment_channel],
+        'no_of_trainings': [no_of_trainings],
+        'age': [age],
+        'previous_year_rating': [previous_year_rating],
+        'length_of_service': [length_of_service],
+        'awards_won': [1 if awards_won else 0],
+        'avg_training_score': [avg_training_score]
+    })
+
+    # Categorical Data Encoding
+    user_input['department'] = department_encoder.transform(user_input['department'])
+    user_input['region'] = region_encoder.transform(user_input['region'])
+    user_input['education'] = education_encoder.transform(user_input['education'])
+    user_input['gender'] = gender_encoder.transform(user_input['gender'])
+    user_input['recruitment_channel'] = recruitment_channel_encoder.transform(user_input['recruitment_channel'])
+
+    # Predicting
+    prediction = model.predict(user_input)
+
+    # Display Result
+    st.subheader('Prediction Result:')
+    st.write('KPIs_met_more_than_80:', prediction[0])
+
+# Tab 2: Visualize Data
+elif st.session_state.tab_selected == 1:
+    st.header('Visualize Data')
+
+    # Select condition feature
+    condition_feature = st.selectbox('Select Condition Feature:', df.columns)
+
+    # Set default condition values
+    default_condition_values = ['Select All'] + df[condition_feature].unique().tolist()
+
+    # Select condition values
+    condition_values = st.multiselect('Select Condition Values:', default_condition_values)
+
+    # Handle 'Select All' choice
+    if 'Select All' in condition_values:
+        condition_values = df[condition_feature].unique().tolist()
+
+    if len(condition_values) > 0:
+        # Filter DataFrame based on selected condition
+        filtered_df = df[df[condition_feature].isin(condition_values)]
+
+        # Plot the number of employees based on KPIs
+        fig, ax = plt.subplots(figsize=(14, 8))
+        sns.countplot(x=condition_feature, hue='KPIs_met_more_than_80', data=filtered_df, palette='viridis')
+        plt.title('Number of Employees based on KPIs')
+        plt.xlabel(condition_feature)
+        plt.ylabel('Number of Employees')
+        st.pyplot(fig)
+
+# Tab 3: Predict from CSV
+elif st.session_state.tab_selected == 2:
+    st.header('Predict from CSV')
+
+    # Upload CSV file
+    uploaded_file = st.file_uploader("Upload CSV file", type=["csv"])
+    # uploaded_file
+    
+    if uploaded_file is not None:
+        # Read CSV file
+        csv_df_org = pd.read_csv(uploaded_file)
+        csv_df_org = csv_df_org.dropna()
+        # csv_df_org.columns
+        
+        csv_df = csv_df_org.copy()
+        csv_df = csv_df.drop('employee_id',axis=1)
+        
+        
+        
+         # Categorical Data Encoding
+        csv_df['department'] = department_encoder.transform(csv_df['department'])
+        csv_df['region'] = region_encoder.transform(csv_df['region'])
+        csv_df['education'] = education_encoder.transform(csv_df['education'])
+        csv_df['gender'] = gender_encoder.transform(csv_df['gender'])
+        csv_df['recruitment_channel'] = recruitment_channel_encoder.transform(csv_df['recruitment_channel'])
+
+
+        # Predicting
+        predictions = model.predict(csv_df)
+
+        # Add predictions to the DataFrame
+        csv_df_org['KPIs_met_more_than_80'] = predictions
+
+        # Display the DataFrame with predictions
+        st.subheader('Predicted Results:')
+        st.write(csv_df_org)
+
+        # Visualize predictions based on a selected feature
+        st.subheader('Visualize Predictions')
+
+        # Select feature for visualization
+        feature_for_visualization = st.selectbox('Select Feature for Visualization:', csv_df_org.columns)
+
+        # Plot the number of employees based on KPIs for the selected feature
+        fig, ax = plt.subplots(figsize=(14, 8))
+        sns.countplot(x=feature_for_visualization, hue='KPIs_met_more_than_80', data=csv_df_org, palette='viridis')
+        plt.title(f'Number of Employees based on KPIs - {feature_for_visualization}')
+        plt.xlabel(feature_for_visualization)
+        plt.ylabel('Number of Employees')
+        st.pyplot(fig)
+
+        

model_kpi_706.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e85a1d1c74943028f6989a7358e490931be09b1b88acd16da7805719c0f339d9
+size 136186

requirements.txt

@@ -0,0 +1,7 @@
+
+matplotlib
+seaborn
+XGBoost
+scikit-learn
+pandas
+