Upload 7 files

Requirements.txt

@@ -0,0 +1,7 @@
+streamlit
+tensorflow
+numpy
+pandas
+scikit-learn
+joblib
+

Scaler/scaler.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:674b9f3f1a9277fbb786dc270fc4d96044ae4b38782e59671b456e342aa79000
+size 1239

Trained Model/churn_prediction_model.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8f9ffaedf97712c45955a5933b567769ce2bfa7f93a6bfc037dd5ca80a885225
+size 37960

app.py

@@ -0,0 +1,44 @@
+import streamlit as st
+import numpy as np
+import tensorflow as tf
+import pandas as pd 
+import joblib
+from sklearn.preprocessing import StandardScaler
+
+# load the trained model
+@st.cache_resource
+def load_model():
+    return tf.keras.models.load_model("Trained Model/churn_prediction_model.keras")
+model = load_model()
+
+# load the saved scaler 
+scaler = joblib.load("Scaler/scaler.pkl")
+st.title("Customer Churn Prediciton")
+st.markdown("<h6 style='text-align: right; color: gray;'>Developed by Abhishek</h6>", unsafe_allow_html=True)
+st.write("Enter customer details to predict churn")
+
+# creating input fields
+input = []
+dataset = pd.read_csv("Dataset/Churn_Modelling.csv")
+dataset.drop(["RowNumber","Empty","CustomerId","Geography","Gender"], axis = 1, inplace = True)
+columns = [
+    ("Credit Score", "CreditScore", 650),
+    ("Age", "Age", 30),
+    ("Tenure (Years)", "Tenure", 5),
+    ("Account Balance", "Balance", 100000),
+    ("Number of Products", "NumOfProducts", 1),
+    ("Has Credit Card (1=Yes, 0=No)", "HasCrCard", 1),
+    ("Is Active Member (1=Yes, 0=No)", "IsActiveMember", 1),
+    ("Estimated Salary", "EstimatedSalary", 50000)
+]
+for label, col, default in columns:
+    value = st.number_input(label, value=float(default), format="%.2f")
+    input.append(value)
+
+if st.button("Predict Churn"):
+    input_array = np.array([input])
+    input_scaled = scaler.transform(input_array)
+    prediction = model.predict(input_scaled)
+    result = "Churn" if prediction[0][0]>0.5 else "No Churn"
+    st.write(f"### Prediction: {result}")
+

churn prediction model.ipynb

@@ -0,0 +1,607 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "   RowNumber  CustomerId  Empty  CreditScore Geography  Gender  Age  Tenure  \\\n",
+      "0          1    15634602    NaN          619    France  Female   42       2   \n",
+      "1          2    15647311    NaN          608     Spain  Female   41       1   \n",
+      "2          3    15619304    NaN          502    France  Female   42       8   \n",
+      "3          4    15701354    NaN          699    France  Female   39       1   \n",
+      "4          5    15737888    NaN          850     Spain  Female   43       2   \n",
+      "\n",
+      "     Balance  NumOfProducts  HasCrCard  IsActiveMember  EstimatedSalary  \\\n",
+      "0       0.00              1          1               1        101348.88   \n",
+      "1   83807.86              1          0               1        112542.58   \n",
+      "2  159660.80              3          1               0        113931.57   \n",
+      "3       0.00              2          0               0         93826.63   \n",
+      "4  125510.82              1          1               1         79084.10   \n",
+      "\n",
+      "   Exited  \n",
+      "0       1  \n",
+      "1       0  \n",
+      "2       1  \n",
+      "3       0  \n",
+      "4       0  \n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "np.int64(10000)"
+      ]
+     },
+     "execution_count": 1,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "import pandas as pd\n",
+    "from sklearn.preprocessing import StandardScaler\n",
+    "\n",
+    "dataset = pd.read_csv(\"Dataset/Churn_Modelling.csv\")\n",
+    "print(dataset.head())\n",
+    "dataset.isnull().sum().sum()\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "dataset.drop([\"RowNumber\",\"Empty\",\"CustomerId\",\"Geography\",\"Gender\"], axis = 1,inplace = True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "input_data = dataset.iloc[:,: -1]\n",
+    "output_data = dataset.iloc[:, -1]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "## standardiazation \n",
+    "ssl = StandardScaler()\n",
+    "input_data = pd.DataFrame(ssl.fit_transform(input_data),columns = input_data.columns)\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Shapes after splitting:\n",
+      "X_train: (8000, 8)\n",
+      "y_train: (8000,)\n",
+      "X_test: (2000, 8)\n",
+      "y_test: (2000,)\n"
+     ]
+    }
+   ],
+   "source": [
+    "# train test split\n",
+    "from sklearn.model_selection import train_test_split\n",
+    "x_train, x_test, y_train, y_test = train_test_split(input_data,output_data , test_size= 0.2 , random_state= 42)\n",
+    "\n",
+    "print(\"Shapes after splitting:\")\n",
+    "print(\"X_train:\", x_train.shape)\n",
+    "print(\"y_train:\", y_train.shape)\n",
+    "print(\"X_test:\", x_test.shape)\n",
+    "print(\"y_test:\", y_test.shape)\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# building ann \n",
+    "import tensorflow\n",
+    "from keras.layers import Dense\n",
+    "from keras.models import Sequential"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "c:\\Users\\abhis\\AppData\\Local\\Programs\\Python\\Python312\\Lib\\site-packages\\keras\\src\\layers\\core\\dense.py:87: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n",
+      "  super().__init__(activity_regularizer=activity_regularizer, **kwargs)\n"
+     ]
+    }
+   ],
+   "source": [
+    "ann = Sequential()\n",
+    "ann.add(Dense(6, input_dim = 8, activation= \"relu\"))\n",
+    "ann.add(Dense(4,  activation= \"relu\"))\n",
+    "ann.add(Dense(2,  activation= \"relu\"))\n",
+    "ann.add(Dense(1,  activation= \"sigmoid\"))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "ann.compile(optimizer= \"adam\", loss = \"binary_crossentropy\", metrics = [\"accuracy\"])\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 1/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 2ms/step - accuracy: 0.8023 - loss: 0.5640\n",
+      "Epoch 2/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7947 - loss: 0.4930\n",
+      "Epoch 3/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7971 - loss: 0.4660\n",
+      "Epoch 4/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7970 - loss: 0.4533\n",
+      "Epoch 5/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7918 - loss: 0.4495\n",
+      "Epoch 6/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7924 - loss: 0.4447\n",
+      "Epoch 7/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7970 - loss: 0.4367\n",
+      "Epoch 8/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7936 - loss: 0.4407\n",
+      "Epoch 9/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7878 - loss: 0.4429\n",
+      "Epoch 10/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7970 - loss: 0.4297\n",
+      "Epoch 11/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7901 - loss: 0.4330\n",
+      "Epoch 12/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7954 - loss: 0.4227\n",
+      "Epoch 13/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.7922 - loss: 0.4289\n",
+      "Epoch 14/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.8070 - loss: 0.4167\n",
+      "Epoch 15/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.8193 - loss: 0.4184\n",
+      "Epoch 16/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.8175 - loss: 0.4197\n",
+      "Epoch 17/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8182 - loss: 0.4139\n",
+      "Epoch 18/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8127 - loss: 0.4322\n",
+      "Epoch 19/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8181 - loss: 0.4128\n",
+      "Epoch 20/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8202 - loss: 0.4235\n",
+      "Epoch 21/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8172 - loss: 0.4202\n",
+      "Epoch 22/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8325 - loss: 0.3909\n",
+      "Epoch 23/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8237 - loss: 0.4101\n",
+      "Epoch 24/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8311 - loss: 0.4025\n",
+      "Epoch 25/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8303 - loss: 0.4055\n",
+      "Epoch 26/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8343 - loss: 0.3941\n",
+      "Epoch 27/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8289 - loss: 0.4029\n",
+      "Epoch 28/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8345 - loss: 0.3982\n",
+      "Epoch 29/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8399 - loss: 0.3928\n",
+      "Epoch 30/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8349 - loss: 0.3970\n",
+      "Epoch 31/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8455 - loss: 0.3846\n",
+      "Epoch 32/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8465 - loss: 0.3733\n",
+      "Epoch 33/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8397 - loss: 0.3904\n",
+      "Epoch 34/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8401 - loss: 0.3864\n",
+      "Epoch 35/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8491 - loss: 0.3753\n",
+      "Epoch 36/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8478 - loss: 0.3714\n",
+      "Epoch 37/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8497 - loss: 0.3709\n",
+      "Epoch 38/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8513 - loss: 0.3698\n",
+      "Epoch 39/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8527 - loss: 0.3696\n",
+      "Epoch 40/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8612 - loss: 0.3533\n",
+      "Epoch 41/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8559 - loss: 0.3661\n",
+      "Epoch 42/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8538 - loss: 0.3576\n",
+      "Epoch 43/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8521 - loss: 0.3696\n",
+      "Epoch 44/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8591 - loss: 0.3591\n",
+      "Epoch 45/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8499 - loss: 0.3689\n",
+      "Epoch 46/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8523 - loss: 0.3640\n",
+      "Epoch 47/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━��\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8541 - loss: 0.3567\n",
+      "Epoch 48/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8563 - loss: 0.3632\n",
+      "Epoch 49/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8524 - loss: 0.3694\n",
+      "Epoch 50/50\n",
+      "\u001b[1m80/80\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.8427 - loss: 0.3741\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<keras.src.callbacks.history.History at 0x57a56ae0>"
+      ]
+     },
+     "execution_count": 9,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "ann.fit(x_train, y_train, batch_size= 100 , epochs = 50)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step\n"
+     ]
+    }
+   ],
+   "source": [
+    "# to find the accuracy \n",
+    "prd = ann.predict(x_test)\n",
+    "prd_data = []\n",
+    "for i in prd:\n",
+    "    if i[0]>0.5:\n",
+    "        prd_data.append(1)\n",
+    "    else:\n",
+    "        prd_data.append(0)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\u001b[1m250/250\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 993us/step\n"
+     ]
+    }
+   ],
+   "source": [
+    "prd1 = ann.predict(x_train)\n",
+    "prd_data1 = []\n",
+    "for i in prd1:\n",
+    "    if i[0]>0.5:\n",
+    "        prd_data1.append(1)\n",
+    "    else:\n",
+    "        prd_data1.append(0)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "(accuracy on testing data 85.35000000000001\n",
+      "(accuraacy on training data 85.5\n"
+     ]
+    }
+   ],
+   "source": [
+    "from sklearn.metrics import accuracy_score\n",
+    "# testing accuracy\n",
+    "print(f\"(accuracy on testing data {accuracy_score(y_test, prd_data)*100}\")\n",
+    "# training accuracy (to check overfitting)\n",
+    "print(f\"(accuraacy on training data {accuracy_score(y_train, prd_data1)*100}\")\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>CreditScore</th>\n",
+       "      <th>Age</th>\n",
+       "      <th>Tenure</th>\n",
+       "      <th>Balance</th>\n",
+       "      <th>NumOfProducts</th>\n",
+       "      <th>HasCrCard</th>\n",
+       "      <th>IsActiveMember</th>\n",
+       "      <th>EstimatedSalary</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>9254</th>\n",
+       "      <td>0.367013</td>\n",
+       "      <td>-0.660018</td>\n",
+       "      <td>0.341352</td>\n",
+       "      <td>-1.225848</td>\n",
+       "      <td>0.807737</td>\n",
+       "      <td>0.646092</td>\n",
+       "      <td>0.970243</td>\n",
+       "      <td>1.373784</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1561</th>\n",
+       "      <td>-0.191713</td>\n",
+       "      <td>0.293517</td>\n",
+       "      <td>-0.350204</td>\n",
+       "      <td>0.691389</td>\n",
+       "      <td>0.807737</td>\n",
+       "      <td>0.646092</td>\n",
+       "      <td>0.970243</td>\n",
+       "      <td>1.667407</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1670</th>\n",
+       "      <td>-0.947028</td>\n",
+       "      <td>-1.422847</td>\n",
+       "      <td>-0.695982</td>\n",
+       "      <td>0.613102</td>\n",
+       "      <td>-0.911583</td>\n",
+       "      <td>0.646092</td>\n",
+       "      <td>-1.030670</td>\n",
+       "      <td>-0.246910</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>6087</th>\n",
+       "      <td>-0.926334</td>\n",
+       "      <td>-1.136786</td>\n",
+       "      <td>1.378686</td>\n",
+       "      <td>0.948021</td>\n",
+       "      <td>-0.911583</td>\n",
+       "      <td>0.646092</td>\n",
+       "      <td>-1.030670</td>\n",
+       "      <td>0.921446</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>6669</th>\n",
+       "      <td>-1.381593</td>\n",
+       "      <td>1.628468</td>\n",
+       "      <td>1.378686</td>\n",
+       "      <td>1.052363</td>\n",
+       "      <td>-0.911583</td>\n",
+       "      <td>-1.547768</td>\n",
+       "      <td>-1.030670</td>\n",
+       "      <td>-1.053812</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "      CreditScore       Age    Tenure   Balance  NumOfProducts  HasCrCard  \\\n",
+       "9254     0.367013 -0.660018  0.341352 -1.225848       0.807737   0.646092   \n",
+       "1561    -0.191713  0.293517 -0.350204  0.691389       0.807737   0.646092   \n",
+       "1670    -0.947028 -1.422847 -0.695982  0.613102      -0.911583   0.646092   \n",
+       "6087    -0.926334 -1.136786  1.378686  0.948021      -0.911583   0.646092   \n",
+       "6669    -1.381593  1.628468  1.378686  1.052363      -0.911583  -1.547768   \n",
+       "\n",
+       "      IsActiveMember  EstimatedSalary  \n",
+       "9254        0.970243         1.373784  \n",
+       "1561        0.970243         1.667407  \n",
+       "1670       -1.030670        -0.246910  \n",
+       "6087       -1.030670         0.921446  \n",
+       "6669       -1.030670        -1.053812  "
+      ]
+     },
+     "execution_count": 13,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "x_train.head()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 113ms/step\n"
+     ]
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "c:\\Users\\abhis\\AppData\\Local\\Programs\\Python\\Python312\\Lib\\site-packages\\keras\\src\\models\\functional.py:238: UserWarning: The structure of `inputs` doesn't match the expected structure.\n",
+      "Expected: keras_tensor\n",
+      "Received: inputs=('Tensor(shape=(1, 8))',)\n",
+      "  warnings.warn(msg)\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "[0]"
+      ]
+     },
+     "execution_count": 14,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# checking on diffrnt data \n",
+    "import numpy as np\n",
+    "input = np.array([[-0.191713,0.293517,-0.350204,0.691389,0.807737,0.646092,0.970243,1.66740]])\n",
+    "prd1 = ann.predict([input])\n",
+    "prd_data1 = []\n",
+    "for i in prd1:\n",
+    "    if i[0]>0.5:\n",
+    "        prd_data1.append(1)\n",
+    "    else:\n",
+    "        prd_data1.append(0)\n",
+    "prd_data1"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 15,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "6252    0\n",
+       "4684    0\n",
+       "1731    0\n",
+       "4742    0\n",
+       "4521    0\n",
+       "       ..\n",
+       "6412    1\n",
+       "8285    0\n",
+       "7853    1\n",
+       "1095    1\n",
+       "6929    1\n",
+       "Name: Exited, Length: 2000, dtype: int64"
+      ]
+     },
+     "execution_count": 15,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "y_test"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 17,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# savin thhe model\n",
+    "ann.save(\"churn_prediction_model.keras\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 21,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "['scaler.pkl']"
+      ]
+     },
+     "execution_count": 21,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# saving the scaler \n",
+    "\n",
+    "import joblib \n",
+    "scaler = StandardScaler()\n",
+    "scaler.fit(dataset.iloc[:, :-1])\n",
+    "joblib.dump(scaler, \"scaler.pkl\")"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.12.0"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}