Upload 3 files

KNN.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:79d2a6ccf9a736551776e3c7539aa31a68bf01d1cdf3dcfd90f0087f43f7eed9
+size 1171

app.py

@@ -13,6 +13,8 @@ linear_model = joblib.load("linear_model.pkl")
 poly_model = joblib.load("poly_model.pkl")
 poly_features = joblib.load("poly_features.pkl")
 scaler = joblib.load("scaler.pkl")  # Load the saved StandardScaler
+knn_model = joblib.load("KNN.pkl")
+randforests_model = joblib.load("random_forests.pkl")
 
 # Function to load and preview CSV data
 def load_data(file):
@@ -33,7 +35,7 @@ def plot_population_trend(file, model_choice):
     plt.grid()
 
     plt.figure(figsize=(8,5))
-    plt.scatter(df["Year"], df["Total_Population"], label="Actual Population", color="blue", alpha=0.6)
+    plt.scatter(df["Year"], df["Population"], label="Actual Population", color="blue", alpha=0.6)
 
     X = df["Year"].values.reshape(-1, 1)  # Extract Year column
 
@@ -41,11 +43,25 @@ def plot_population_trend(file, model_choice):
         X_scaled = scaler.transform(X) 
         predictions = linear_model.predict(X_scaled)
         plt.plot(df["Year"], predictions, label="Linear Regression", color="red", linestyle="dashed")
-    else:  # Polynomial Regression
+    elif model_choice == "Polynomial Regression": # Polynomial Regression
         X_scaled = scaler.transform(X)  # Apply scaling
         X_poly = poly_features.transform(X_scaled)  # Transform for Polynomial Regression
         predictions = poly_model.predict(X_poly)
         plt.plot(df["Year"], predictions, label="Polynomial Regression", color="green")
+    elif model_choice == "KNN":  # K-Nearest Neighbors (KNN)
+            predictions = knn_model.predict(X)
+            label = "KNN"
+            color = "blue"
+            linestyle = "dotted"
+            plt.plot(df["Year"], predictions, label="KNN", color="blue")
+    else:   #Random Forests 
+            predictions = randforests_model.predict(X)
+            label = "Random Forests"
+            color = "yellow"
+            linestyle = "dotted"
+            plt.plot(df["Year"], predictions, label="Random Forests", color="yellow")
+
+    
 
     plt.xlabel("Year")
     plt.ylabel("Population")
@@ -76,11 +92,14 @@ def predict_population(file, model_choice):
         # Do NOT scale X for Linear Regression
         X_scaled = scaler.transform(X) 
         predictions = linear_model.predict(X_scaled)  
-    else:  # Polynomial Regression
+    elif model_choice== "Polynomial Regression":  # Polynomial Regression
         X_scaled = scaler.transform(X)  # Apply the same scaling as training
         X_poly = poly_features.transform(X_scaled)  # Transform for Polynomial Regression
         predictions = poly_model.predict(X_poly)
-
+    elif model_choice == "KNN":
+         predictions = knn_model.predict(X)
+    else:#random forests 
+         predictions = randforests_model.predict(X)
 
 
     df["Predicted Population"] = predictions  # Append predictions to DataFrame
@@ -116,7 +135,7 @@ interface = gr.Interface(
     fn=gradio_interface,  # Use the single wrapper function
     inputs=[
         gr.File(label="Upload CSV File"),
-        gr.Radio(["Linear Regression", "Polynomial Regression"], label="Choose Model")
+        gr.Radio(["Linear Regression", "Polynomial Regression","KNN", "Random Forests"], label="Choose Model")
     ],
     outputs=[
         gr.Dataframe(label="Preview Data"),
@@ -125,7 +144,7 @@ interface = gr.Interface(
         gr.Textbox(label="Model Performance")
     ],
     title="Population Prediction Tool",
-    description="Upload a CSV file with Year and Population data. Choose a model (Linear or Polynomial Regression) to predict future population trends."
+    description="Upload a CSV file with Year and Population data. Choose a model (Linear or Polynomial Regression, KNN or Random Forests) to predict future population trends."
 )
 
 # Launch the Gradio App

random_forests.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:02ede939310eefd2d36ea586f799203e8a6a3ba840b11b0d584389bff51c0b85
+size 9985