update

app.py

@@ -1,157 +1,69 @@
-import streamlit as st
-import numpy as np
 import joblib
-from sklearn.preprocessing import StandardScaler
 import pandas as pd
-import matplotlib.pyplot as plt
-import seaborn as sns
+import streamlit as st
 from huggingface_hub import hf_hub_download
 
+REPO_ID = "chanyaphas/creditc"
 
-REPO_ID = "flokabukie/Credit_Card_Fraud_Detection"
-
-# Load the non-anomaly data
-non_anomaly_csv_filename = 'non_anomaly_data.csv'
-non_anomaly_csv_filename = hf_hub_download(repo_id=REPO_ID, filename=non_anomaly_csv_filename)
-non_anomaly_df = pd.read_csv(non_anomaly_csv_filename)
-
-# Load the Isolation Forest model
-model_filename = "IsolationForest.joblib"
-
-isolation_forest = joblib.load(
-    hf_hub_download(repo_id=REPO_ID, filename=model_filename)
+model = joblib.load(
+    hf_hub_download(repo_id=REPO_ID, filename="model.joblib")
 )
 
-# Load the StandardScaler
-scaler_filename = "StandardScaler.joblib"
-
-scaler = joblib.load(
-    hf_hub_download(repo_id=REPO_ID, filename=scaler_filename)
+unique_values = joblib.load(
+    hf_hub_download(repo_id=REPO_ID, filename="unique_values.joblib")
 )
 
-st.title("Anomaly Detection App with Isolation Forest")
-
-st.sidebar.title("Input Feature Values")
-transaction_dollar_amount = st.sidebar.slider("Transaction Dollar Amount", min_value=0.0, max_value=10000.0)
-longitude = st.sidebar.slider("Longitude (Long)", min_value=-180.0, max_value=180.0)
-latitude = st.sidebar.slider("Latitude (Lat)", min_value=-90.0, max_value=90.0)
-credit_card_limit = st.sidebar.slider("Credit Card Limit", min_value=0, max_value=50000)
-year = st.sidebar.slider("Year", min_value=2000, max_value=2030)
-month = st.sidebar.slider("Month", min_value=1, max_value=12)
-day = st.sidebar.slider("Day", min_value=1, max_value=31)
-
-submitted = st.sidebar.button("Submit")
-
-if submitted:
-    input_data = {
-        'transaction_dollar_amount': transaction_dollar_amount,
-        'Long': longitude,
-        'Lat': latitude,
-        'credit_card_limit': credit_card_limit,
-        'year': year,
-        'month': month,
-        'day': day
-    }
-
-    selected_columns = pd.DataFrame([input_data])
-
-    # Standardize the input data using the loaded StandardScaler
-    selected_columns_scaled = scaler.transform(selected_columns)
-
-    # Apply Isolation Forest for anomaly detection on the non-anomaly dataset
-    non_anomaly_scores = isolation_forest.decision_function(scaler.transform(non_anomaly_df))
-
-# Apply Isolation Forest for anomaly detection on your single input data
-    your_anomaly_score = isolation_forest.decision_function(selected_columns_scaled)[0]
-
-
-
-    # Calculate the minimum and maximum anomaly scores from non-anomaly data
-    min_non_anomaly_score = np.min(non_anomaly_scores)
-    max_non_anomaly_score = np.max(non_anomaly_scores)
-
-# Add a margin of error for the range
-    margin = 0.5
-    min_threshold = min_non_anomaly_score - margin
-    max_threshold = max_non_anomaly_score + margin
-
-    # Determine if the input data point is an anomaly based on the score
-    #is_anomaly = your_anomaly_score >= np.percentile(non_anomaly_scores, 95)
-
-    # Determine if the input data point is an anomaly based on the score
-    is_anomaly = your_anomaly_score < min_threshold or your_anomaly_score > max_threshold
-
-
-# Print the anomaly status
-    st.subheader("Anomaly Classification")
-    if is_anomaly:
-        st.write("Prediction Result: 🚨 Anomaly Detected!")
-    else:
-        st.write("Prediction Result: ✅ Not Anomaly")
-
-# Create a bar plot to visualize the anomaly score distribution and your data point's score
-    plt.figure(figsize=(8, 5))
-
-# Plot the distribution of anomaly scores from the non-anomaly dataset
-    sns.histplot(non_anomaly_scores, kde=True, color='gray', label='Non-Anomaly Score Distribution')
-
-# Plot your data point's anomaly score
-    plt.axvline(x=your_anomaly_score, color='blue', linestyle='dashed', label='Your Data Point')
-
-# Set labels and title
-    plt.xlabel('Anomaly Score')
-    plt.ylabel('Frequency')
-    plt.title('Anomaly Score Distribution and Your Data Point')
-    plt.legend()
-#plt.grid(True)
-
-# Display the histogram plot
-    st.pyplot(plt)
-
-
-# Explain the results
-    st.write("The input data point has been classified as an anomaly." if is_anomaly
-            else "The input data point is not classified as an anomaly.")
-    st.write("The anomaly score is:", your_anomaly_score)
-    st.write("The threshold for anomaly detection is:", min_threshold, "to", max_threshold)
-
-    # Create a scatter plot for longitude and latitude
-    fig, ax = plt.subplots(figsize=(10, 8))
-
-# Plot non-anomaly data
-    sns.scatterplot(data=non_anomaly_df, x='Long', y='Lat', color='lightgrey', label='Normal 🏙️', ax=ax)
-
-# Plot input data
-    if is_anomaly:
-        ax.scatter(selected_columns['Long'], selected_columns['Lat'], color='red', label='Suspicious 🚩', s=100, marker='x')
-        anomaly_marker = 'Suspicious 🚩'
-    else:
-        ax.scatter(selected_columns['Long'], selected_columns['Lat'], color='green', label='Valid ✅', s=100, marker='o')
-        anomaly_marker = 'Valid ✅'
-
-    ax.set_xlabel("Longitude")
-    ax.set_ylabel("Latitude")
-    ax.set_title("Location Plot: Anomaly Detection 🗺️")
-    ax.legend()
-    ax.grid(True)
-
-# Show the scatter plot in Streamlit
-    st.subheader("Location Plot: Anomaly Detection 🗺️")
-    st.pyplot(fig)
-
-# Explanation based on the anomaly classification
-    st.subheader("Anomaly Classification")
-    if your_anomaly_score < min_threshold or your_anomaly_score > max_threshold:
-        st.write("Prediction Result: 🚨 Anomaly Detected!")
-    else:
-        st.write("Prediction Result: ✅ Not Anomaly")
-
-# Explain the results
-    # Explain the results
-    st.write("The location plot visualizes the anomaly detection result based on longitude and latitude.")
-    if your_anomaly_score < min_threshold or your_anomaly_score > max_threshold:
-        st.write("The input data point is marked as Suspicious 🚩 due to its anomaly score.")
-        st.write("The red 'x' marker indicates a suspicious location.")
-    else:
-        st.write("The input data point is marked as Valid ✅ due to its anomaly score.")
-        st.write("The green 'o' marker indicates a valid location.")
+EDU_DICT = {'Lower secondary': 1,
+            'Secondary / secondary special': 2,
+            'Academic degree': 3,
+            'Incomplete higher': 4,
+            'Higher education' : 5
+            }
+
+def main():
+    st.title("Credit Card Approval Prediction")
+
+    with st.form("questionaire"):
+
+        Gender = st.selectbox('Gender', unique_values['CODE_GENDER'])
+        Own_car = st.selectbox('Own_car', unique_values['FLAG_OWN_CAR'])
+        Property = st.selectbox('Property', unique_values['FLAG_OWN_REALTY'])
+        Income_type = st.selectbox('Income_type', unique_values['NAME_INCOME_TYPE'])
+        Marital_status = st.selectbox('Marital_status', unique_values['NAME_FAMILY_STATUS'])
+        Housing_type = st.selectbox('Housing_type', unique_values['NAME_HOUSING_TYPE'])
+        Education = st.selectbox('Education', unique_values['NAME_EDUCATION_TYPE'])
+
+        Income = st.slider('Income', min_value=27000, max_value=1575000)
+        Children = st.slider('Children', min_value=0, max_value=19)
+        Day_Employed = st.slider('Day_Employed', min_value=0, max_value=3)
+        Flag_Mobile = st.slider('Flag_Mobile', min_value=0, max_value=1)
+        Flag_work_phone = st.slider('Flag_work_phone', min_value=0, max_value=1)
+        Flag_Phone = st.slider('Flag_Phone', min_value=0, max_value=1)
+        Flag_Email = st.slider('Flag_Email', min_value=0, max_value=1)
+        Family_mem = st.slider('Family_mem', min_value=1, max_value=20)
+
+        clicked = st.form_submit_button("Result")
+        if clicked:
+            result = model.predict(pd.DataFrame({
+                                                "CODE_GENDER": [Gender],
+                                                "FLAG_OWN_CAR": [Own_car],
+                                                "FLAG_OWN_REALTY": [Property],
+                                                "CNT_CHILDREN": [Children],
+                                                "AMT_INCOME_TOTAL": [Income],
+                                                "NAME_INCOME_TYPE": [Income_type],
+                                                "NAME_EDUCATION_TYPE": [EDU_DICT[Education]],
+                                                "NAME_FAMILY_STATUS": [Marital_status],
+                                                "NAME_HOUSING_TYPE": [Housing_type],
+                                                "DAYS_EMPLOYED": [Day_Employed],
+                                                "FLAG_MOBIL": [Flag_Mobile],
+                                                "FLAG_WORK_PHONE": [Flag_work_phone],
+                                                "FLAG_PHONE": [Flag_Phone],
+                                                "FLAG_EMAIL": [Flag_Email],
+                                                "CNT_FAM_MEMBERS": [Family_mem]}))
+
+            result = 'Pass' if result[0] == 1 else 'Did not Pass'
+
+            st.success('Credit Card approval prediction results is {}'.format(result))
+
+if __name__ == '__main__':
+    main()

app_2.py

@@ -19,6 +19,21 @@ model = joblib.load(
     hf_hub_download(repo_id=REPO_ID, filename=FILENAME)
 )
 
+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(
+    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

@@ -1,10 +1,6 @@
-streamlit==1.25.0
-numpy==1.23.5
+joblib
+pandas
 scikit-learn==1.2.2
-sklearn-pandas==2.2.0
-joblib==1.3.2
-matplotlib==3.7.1
-matplotlib-inline==0.1.6
-matplotlib-venn==0.11.9
-seaborn==0.12.2
-huggingface_hub==0.11
+xgboost==1.7.6
+altair<5
+huggingface_hub