update file

app.py

@@ -1,172 +1,166 @@
-import pandas as pd
 import streamlit as st
-import numpy as np
-from matplotlib import pyplot as plt
-import pickle
-import sklearn
+import pandas as pd
 import joblib
-from PIL import Image
-import base64
+import matplotlib.pyplot as plt
 from huggingface_hub import hf_hub_download
+import time
+import base64 
+
+# Load the pre-trained numerical imputer, scaler, and model using joblib
 
-REPO_ID = "Abubakari/Sales_Prediction"
+REPO_ID = "Abubakari/Sepsis-prediction-streamlit-app"
 
 num_imputer = joblib.load(
     hf_hub_download(repo_id=REPO_ID, filename="numerical_imputer.joblib")
 )
 
-cat_imputer = joblib.load(
-    hf_hub_download(repo_id=REPO_ID, filename="categorical_imputer.joblib")
-)
-
-encoder = joblib.load(
-    hf_hub_download(repo_id=REPO_ID, filename="encoder.joblib")
-)
-
 scaler = joblib.load(
     hf_hub_download(repo_id=REPO_ID, filename="scaler.joblib")
 )
 
-dt_model = joblib.load(
+model = joblib.load(
     hf_hub_download(repo_id=REPO_ID, filename="Final_model.joblib")
 )
 
-# Add a title and subtitle
-st.write("<center><h1>Sales Prediction App</h1></center>", unsafe_allow_html=True)
-
-# Set up the layout
-col1, col2, col3 = st.columns([1, 3, 3])
-
-
-# Add a subtitle or description
-st.write("This app uses machine learning to predict sales based on certain input parameters. Simply enter the required information and click 'Predict' to get a sales prediction!")
-
-st.subheader("Enter the details to predict sales")
-
-# Add some text
-#st.write("Enter some data for Prediction.")
-
- # Create the input fields
-input_data = {}
-col1,col2 = st.columns(2)
-with col1:
-    input_data['store_nbr'] = st.slider("store_nbr",0,54)
-    input_data['products'] = st.selectbox("products", ['AUTOMOTIVE', 'CLEANING', 'BEAUTY', 'FOODS', 'STATIONERY',
-       'CELEBRATION', 'GROCERY', 'HARDWARE', 'HOME', 'LADIESWEAR',
-       'LAWN AND GARDEN', 'CLOTHING', 'LIQUOR,WINE,BEER', 'PET SUPPLIES'])
-    input_data['onpromotion'] =st.number_input("onpromotion",step=1)
-    input_data['state'] = st.selectbox("state", ['Pichincha', 'Cotopaxi', 'Chimborazo', 'Imbabura',
-       'Santo Domingo de los Tsachilas', 'Bolivar', 'Pastaza',
-       'Tungurahua', 'Guayas', 'Santa Elena', 'Los Rios', 'Azuay', 'Loja',
-       'El Oro', 'Esmeraldas', 'Manabi'])
-    input_data['store_type'] = st.selectbox("store_type",['D', 'C', 'B', 'E', 'A'])
-    input_data['cluster'] = st.number_input("cluster",step=1)
-
-with col2:
-    input_data['dcoilwtico'] = st.number_input("dcoilwtico",step=1)
-    input_data['year'] = st.number_input("year",step=1)
-    input_data['month'] = st.slider("month",1,12)
-    input_data['day'] = st.slider("day",1,31)
-    input_data['dayofweek'] = st.number_input("dayofweek,0=Sun and 6=Sat",step=1)
-    input_data['end_month'] = st.selectbox("end_month",['True','False'])
-
-
-# Define CSS style for the download button
-# Define the custom CSS
-predict_button_css = """
-    <style>
-    .predict-button {
-        background-color: #C4C4C4;
-        color: gray;
-        padding: 0.75rem 2rem;
-        border-radius: 0.5rem;
-        border: none;
-        font-size: 1.1rem;
-        font-weight: bold;
-        text-align: center;
-        margin-top: 2rem;
-    }
-    </style>
-"""
-
-download_button_css = """
-    <style>
-    .download-button {
-        background-color: #C4C4C4;
-        color: white;
-        padding: 0.75rem 2rem;
-        border-radius: 0.5rem;
-        border: none;
-        font-size: 1.1rem;
-        font-weight: bold;
-        text-align: center;
-        margin-top: 1rem;
-    }
-    </style>
-"""
-
-# Display the custom CSS
-st.markdown(predict_button_css + download_button_css, unsafe_allow_html=True)
-
-
-  # Create a button to make a prediction
-
-if st.button("Predict", key="predict_button", help="Click to make a prediction."):
-    # Convert the input data to a pandas DataFrame
-        input_df = pd.DataFrame([input_data])
-
-
-# Selecting categorical and numerical columns separately
-        cat_columns = [col for col in input_df.columns if input_df[col].dtype == 'object']
-        num_columns = [col for col in input_df.columns if input_df[col].dtype != 'object']
-
-
-# Apply the imputers
-        input_df_imputed_cat = cat_imputer.transform(input_df[cat_columns])
-        input_df_imputed_num = num_imputer.transform(input_df[num_columns])
-
-
- # Encode the categorical columns
-        input_encoded_df = pd.DataFrame(encoder.transform(input_df_imputed_cat).toarray(),
-                                   columns=encoder.get_feature_names(cat_columns))
-
-# Scale the numerical columns
-        input_df_scaled = scaler.transform(input_df_imputed_num)
-        input_scaled_df = pd.DataFrame(input_df_scaled , columns = num_columns)
-
-#joining the cat encoded and num scaled
-        final_df = pd.concat([input_encoded_df, input_scaled_df], axis=1)
-
-# Make a prediction
-        prediction = dt_model.predict(final_df)[0]
-
-
-# Display the prediction
-        st.write(f"The predicted sales are: {prediction}.")
-        input_df.to_csv("data.csv", index=False)
-        st.table(input_df)
-
-# Define custom CSS
-css = """
-table {
-    background-color: #f2f2f2;
-    color: #333333;
-}
-"""
-
-# Set custom CSS
-st.write(f'<style>{css}</style>', unsafe_allow_html=True)
-
-
-# Add the download button
-def download_csv():
-    with open("data.csv", "r") as f:
-        csv = f.read()
-    b64 = base64.b64encode(csv.encode()).decode()
-    button = f'<button class="download-button"><a href="data:file/csv;base64,{b64}" download="data.csv">Download Data CSV</a></button>'
-    return button
-
-st.markdown(
-    f'<div style="text-align: center">{download_csv()}</div>',
-    unsafe_allow_html=True
-)
+# Define a function to preprocess the input data
+def preprocess_input_data(input_data):
+    input_data_df = pd.DataFrame(input_data, columns=['PRG', 'PL', 'PR', 'SK', 'TS', 'M11', 'BD2', 'Age', 'Insurance'])
+    num_columns = input_data_df.select_dtypes(include='number').columns
+
+    input_data_imputed_num = num_imputer.transform(input_data_df[num_columns])
+    input_scaled_df = pd.DataFrame(scaler.transform(input_data_imputed_num), columns=num_columns)
+
+    return input_scaled_df
+
+
+# Define a function to make the sepsis prediction
+def predict_sepsis(input_data):
+    input_scaled_df = preprocess_input_data(input_data)
+    prediction = model.predict(input_scaled_df)[0]
+    probabilities = model.predict_proba(input_scaled_df)[0]
+    sepsis_status = "Positive" if prediction == 1 else "Negative"
+    
+    status_icon = "✔" if prediction == 1 else "✘"  # Red 'X' icon for positive sepsis prediction, green checkmark icon for negative sepsis prediction
+    sepsis_explanation = "Sepsis is a life-threatening condition caused by an infection. A positive prediction suggests that the patient might be exhibiting sepsis symptoms and requires immediate medical attention." if prediction == 1 else "Sepsis is a life-threatening condition caused by an infection. A negative prediction suggests that the patient is not currently exhibiting sepsis symptoms."
+
+    output_df = pd.DataFrame(input_data, columns=['PRG', 'PL', 'PR', 'SK', 'TS', 'M11', 'BD2', 'Age', 'Insurance'])
+    output_df['Prediction'] = sepsis_status
+    output_df['Negative Probability'] = probabilities[0]
+    output_df['Positive Probability'] = probabilities[1]
+
+    return output_df, probabilities, status_icon, sepsis_explanation
+
+# Create a Streamlit app
+def main():
+    st.title('Sepsis Prediction App')
+
+    st.image("Strealit_.jpg")
+
+    # How to use
+    st.sidebar.title('How to Use')
+    st.sidebar.markdown('1. Adjust the input parameters on the left sidebar.')
+    st.sidebar.markdown('2. Click the "Predict" button to initiate the prediction.')
+    st.sidebar.markdown('3. The app will simulate a prediction process with a progress bar.')
+    st.sidebar.markdown('4. Once the prediction is complete, the results will be displayed below.')
+
+
+    st.sidebar.title('Input Parameters')
+
+    # Input parameter explanations
+    st.sidebar.markdown('**PRG:** Plasma Glucose')
+    PRG = st.sidebar.number_input('PRG', value=0.0)
+
+    st.sidebar.markdown('**PL:** Blood Work Result 1')
+    PL = st.sidebar.number_input('PL', value=0.0)
+
+    st.sidebar.markdown('**PR:** Blood Pressure Measured')
+    PR = st.sidebar.number_input('PR', value=0.0)
+
+    st.sidebar.markdown('**SK:** Blood Work Result 2')
+    SK = st.sidebar.number_input('SK', value=0.0)
+
+    st.sidebar.markdown('**TS:** Blood Work Result 3')
+    TS = st.sidebar.number_input('TS', value=0.0)
+
+    st.sidebar.markdown('**M11:** BMI')
+    M11 = st.sidebar.number_input('M11', value=0.0)
+
+    st.sidebar.markdown('**BD2:** Blood Work Result 4')
+    BD2 = st.sidebar.number_input('BD2', value=0.0)
+
+    st.sidebar.markdown('**Age:** What is the Age of the Patient: ')
+    Age = st.sidebar.number_input('Age', value=0.0)
+
+    st.sidebar.markdown('**Insurance:** Does the patient have Insurance?')
+    insurance_options = {0: 'NO', 1: 'YES'}
+    Insurance = st.sidebar.radio('Insurance', list(insurance_options.keys()), format_func=lambda x: insurance_options[x])
+
+
+    input_data = [[PRG, PL, PR, SK, TS, M11, BD2, Age, Insurance]]
+
+    if st.sidebar.button('Predict'):
+        with st.spinner("Predicting..."):
+            # Simulate a long-running process
+            progress_bar = st.progress(0)
+            step = 20 # A big step will reduce the execution time
+            for i in range(0, 100, step):
+                time.sleep(0.1)
+                progress_bar.progress(i + step)
+
+            output_df, probabilities, status_icon, sepsis_explanation = predict_sepsis(input_data)
+
+            st.subheader('Prediction Result')
+            prediction_text = "Positive" if status_icon == "✔" else "Negative"
+            st.markdown(f"Prediction: **{prediction_text}**")
+            st.markdown(f"{status_icon} {sepsis_explanation}")
+            st.write(output_df)
+
+            # Add a download button for output_df
+            csv = output_df.to_csv(index=False)
+            b64 = base64.b64encode(csv.encode()).decode()
+            href = f'<a href="data:file/csv;base64,{b64}" download="output.csv">Download Output CSV</a>'
+            st.markdown(href, unsafe_allow_html=True)
+            
+
+            # Plot the probabilities
+            fig, ax = plt.subplots()
+            ax.bar(['Negative', 'Positive'], probabilities)
+            ax.set_xlabel('Sepsis Status')
+            ax.set_ylabel('Probability')
+            ax.set_title('Sepsis Prediction Probabilities')
+            st.pyplot(fig)
+
+            # Print feature importance
+            if hasattr(model, 'coef_'):
+                feature_importances = model.coef_[0]
+                feature_names = ['PRG', 'PL', 'PR', 'SK', 'TS', 'M11', 'BD2', 'Age', 'Insurance']
+
+                importance_df = pd.DataFrame({'Feature': feature_names, 'Importance': feature_importances})
+                importance_df = importance_df.sort_values('Importance', ascending=False)
+
+                st.subheader('Feature Importance')
+                fig, ax = plt.subplots()
+                bars = ax.bar(importance_df['Feature'], importance_df['Importance'])
+                ax.set_xlabel('Feature')
+                ax.set_ylabel('Importance')
+                ax.set_title('Feature Importance')
+                ax.tick_params(axis='x', rotation=45)
+
+                # Add data labels to the bars
+                for bar in bars:
+                    height = bar.get_height()
+                    ax.annotate(f'{height:.2f}', xy=(bar.get_x() + bar.get_width() / 2, height),
+                    xytext=(0, 3),  # 3 points vertical offset
+                    textcoords="offset points",
+                    ha='center', va='bottom')
+                st.pyplot(fig)
+
+            else:
+                st.write('Feature importance is not available for this model.')
+
+            #st.subheader('Sepsis Explanation')
+            #st.markdown(f"{status_icon} {sepsis_explanation}")
+
+
+if __name__ == '__main__':
+    main()

app_2.py

@@ -11,27 +11,14 @@ from transformers import pipeline
 import datetime
 from huggingface_hub import hf_hub_download
 
-REPO_ID = "AlbieCofie/predict-customer-churn"
+REPO_ID = "rajistics/churn-model"
+FILENAME = "churn.pkl"
 
-num_imputer = joblib.load(
-    hf_hub_download(repo_id=REPO_ID, filename="numerical_imputer.joblib")
-)
-
-cat_imputer = joblib.load(
-    hf_hub_download(repo_id=REPO_ID, filename="categorical_imputer.joblib")
-)
 
-encoder = joblib.load(
-    hf_hub_download(repo_id=REPO_ID, filename="encoder.joblib")
+model = joblib.load(
+    hf_hub_download(repo_id=REPO_ID, filename=FILENAME)
 )
 
-scaler = joblib.load(
-    hf_hub_download(repo_id=REPO_ID, filename="scaler.joblib")
-)
-
-dt_model = joblib.load(
-    hf_hub_download(repo_id=REPO_ID, filename="Final_model.joblib")
-)
 
 # Add a title and subtitle
 st.write("<center><h1>Sales Prediction App</h1></center>", unsafe_allow_html=True)

requirements.txt

@@ -1,7 +1,12 @@
-transformers==4.22.2
-streamlit
-huggingface_hub==0.9.1
+joblib==1.2.0
+matplotlib==3.7.1
+matplotlib-inline==0.1.6
+numpy==1.24.2
+pandas==1.5.3
 scikit-learn==1.2.2
-joblib
-torch
-pandas
+scipy==1.10.0
+seaborn==0.12.2
+streamlit==1.20.0
+fastapi==0.95.1
+uvicorn==0.22.0
+pydantic==1.10.7