model for deep learning and tailwind css modified

.gitignore

@@ -20,6 +20,8 @@ mlruns
 mlartifacts/
 mlruns/
 
+node_modules/
+
 
 # Distribution / packaging
 .Python

Dockerfile

@@ -1,43 +1,24 @@
-# Use official Python image as the base image
+# Use the official Python image as a base image
 FROM python:3.10
 
-# Set environment variables
-ENV PYTHONDONTWRITEBYTECODE 1
-ENV PYTHONUNBUFFERED 1
-ENV FLASK_APP=app
-
-# Copy and install system dependencies
-RUN apt-get update && apt-get install -y \
-    build-essential \
-    libpq-dev \
-    curl \
-    && rm -rf /var/lib/apt/lists/*
-
-# Install Node.js and npm for TailwindCSS
-RUN curl -fsSL https://deb.nodesource.com/setup_18.x | bash - && \
-    apt-get install -y nodejs && \
-    apt-get clean && rm -rf /var/lib/apt/lists/*
-
-# Set the working directory
+# Set the working directory in the container
 WORKDIR /app
 
-# Copy project dependencies first to improve caching
+# Copy the requirements file into the container
 COPY requirements.txt .
 
-RUN pip install --no-cache-dir --upgrade pip && \
-    pip install --no-cache-dir -r requirements.txt
-# Copy Node dependencies and install them
-COPY app/package.json .
-RUN npm install
+# Install the dependencies
+RUN pip install --no-cache-dir -r requirements.txt
 
-# Copy the rest of the project files
-COPY app/. .
+# Copy the entire application into the container
+COPY . .
 
-# Build CSS with Tailwind
-RUN npm run create-css
+# Set environment variables
+ENV FLASK_APP=app
+ENV FLASK_ENV=development
 
-# Expose port 5000 for the Flask app
+# Expose the port the app runs on
 EXPOSE 5000
 
-# Run the Flask app
+# Command to run the application
 CMD ["flask", "run", "--host=0.0.0.0"]

app/__init__.py

@@ -2,7 +2,7 @@
 from dotenv import load_dotenv
 from flask import Flask
 
-from app import pages, transactions, data, predict
+from app import pages, transactions, data, xgb_model
 from app.models import db
 from app.database import init_db_command
 from app.dashboard import (
@@ -35,7 +35,7 @@ def create_app():
     app.register_blueprint(pages.bp)
     app.register_blueprint(transactions.bp)
     app.register_blueprint(data.bp)
-    app.register_blueprint(predict.bp)
+    app.register_blueprint(xgb_model.bp)
 
     # Initialize Dash apps and pass the `db` object for querying the database
     create_summary_dash_app(app, db)

app/dashboard.py

@@ -1,7 +1,7 @@
 import pandas as pd
 import plotly.express as px
 from dash import Dash, dcc, html, Input, Output
-from sqlalchemy import create_engine, func
+from sqlalchemy import func
 from app.models import Features 
 
 def create_summary_dash_app(flask_app, db):
@@ -101,7 +101,7 @@ def update_device_browser_analysis(_):
             )
             device_browser_df = pd.DataFrame(device_browser_data, columns=['device_id', 'browser', 'fraud_cases'])
 
-        fig = px.bar(device_browser_df, x='device_id', y='fraud_cases', color='browser', title='Fraud by Device and Browser')
+        fig = px.bar(device_browser_df, x='fraud_cases', y='device_id', color='browser', title='Fraud by Device and Browser')
         return fig
 
     return dash_app

app/data.py

@@ -3,31 +3,13 @@
 import pandas as pd
 import matplotlib.pyplot as plt
 from dotenv import load_dotenv
-from flask import Blueprint, render_template, jsonify
+from flask import Blueprint, render_template
 from sqlalchemy import create_engine
-from sklearn.model_selection import train_test_split
-from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score, confusion_matrix
-from sklearn.preprocessing import LabelEncoder
-from imblearn.over_sampling import SMOTE
-from xgboost import XGBClassifier
-from sklearn import preprocessing
-import mlflow
-from mlflow import pyfunc
-import xgboost as xgb
-from dash import Dash, dcc, html, Input, Output
-from flask import Blueprint
-from dash import Dash
-import dash_core_components as dcc
-import dash_html_components as html
 
 load_dotenv()
 
 bp = Blueprint("data", __name__)
 
-# Load the model
-model = xgb.Booster()
-model = model.load_model("model.xgb")
-
 @bp.route("/data")
 def data():
     # Retrieve database URL from environment variables
@@ -59,56 +41,3 @@ def data():
     features = df.to_dict(orient='records')
 
     return render_template('data/data.html', features=features)
-
-@bp.route('/result')
-def result():
-    database_url = os.getenv('DATABASE_URL')
-    if not database_url:
-        raise RuntimeError("DATABASE_URL environment variable not set.")
-
-    engine = create_engine(database_url)
-
-    # Fetch data from database
-    data = pd.read_sql("SELECT * FROM transactions ORDER BY ctid DESC LIMIT 1", engine)
-
-    ID = data['user_id']
-
-    df = data
-
-    # Initialize a LabelEncoder object
-    label_encoder = LabelEncoder()
-
-    # Iterate through each column in the dataframe
-    for column in df.columns:
-        # Check if the column datatype is not numeric
-        if df[column].dtype not in ['int64', 'float64']:
-            # Fit label encoder and transform values
-            df[column] = label_encoder.fit_transform(df[column])
-
-    # Get the numeric columns
-    numeric_columns = df.select_dtypes(include=['number']).columns
-
-    # Create the StandardScaler
-    transform = preprocessing.StandardScaler()
-
-    # Fit the scaler on the data (calculate mean and standard deviation)
-    transform.fit(df[numeric_columns])
-
-    # Transform the data using the fitted transform and reassign it to X
-    df[numeric_columns] = transform.transform(df[numeric_columns])
-
-    df = xgb.DMatrix(df)
-
-    prediction = None
-
-    if model is None:
-        print("Model is not loaded. Please check the loading process.")
-    else:
-        prediction = model.predict(df)
-
-
-    # Make prediction
-    #prediction = model.predict(df)
-    print("Prediction: ", prediction)
-
-    return render_template('data/result.html', prediction=prediction, ID=ID)

app/models.py

@@ -33,7 +33,7 @@ class Features(db.Model):
     sex = db.Column(db.String)
     age = db.Column(db.Integer)
     ip_address = db.Column(db.String)
-    class_ = db.Column('class', db.Integer)  # Fraud class: 0 = non-fraud, 1 = fraud
+    class_ = db.Column('class', db.Integer)
     lower_bound_ip_address = db.Column(db.BigInteger)
     upper_bound_ip_address = db.Column(db.BigInteger)
     country = db.Column(db.String)

app/schema.sql

@@ -1,17 +1,35 @@
 DROP TABLE IF EXISTS transactions;
 
 CREATE TABLE transactions (
-    user_id INT PRIMARY KEY, -- user_id as the unique identifier
-    signup_time TIMESTAMP NOT NULL, -- Signup time as timestamp
-    purchase_time TIMESTAMP NOT NULL, -- Purchase time as timestamp
-    purchase_value FLOAT NOT NULL, -- Purchase value as a float
-    device_id VARCHAR NOT NULL, -- Device ID as a string
-    source VARCHAR NOT NULL, -- Source as a string (e.g., app, web)
-    browser VARCHAR NOT NULL, -- Browser as a string
-    sex VARCHAR NOT NULL, -- Sex as a string
-    age INT NOT NULL, -- Age as integer
-    ip_address FLOAT NOT NULL, -- IP address as a float
-    lower_bound_ip_address FLOAT NOT NULL, -- Lower bound of IP address
-    upper_bound_ip_address FLOAT NOT NULL, -- Upper bound of IP address
-    country VARCHAR NOT NULL -- Country as a string
+    user_id INT PRIMARY KEY, 
+    signup_time TIMESTAMP NOT NULL,
+    purchase_time TIMESTAMP NOT NULL, 
+    purchase_value FLOAT NOT NULL,
+    device_id VARCHAR NOT NULL,
+    source VARCHAR NOT NULL, 
+    browser VARCHAR NOT NULL, 
+    sex VARCHAR NOT NULL, 
+    age INT NOT NULL, 
+    ip_address FLOAT NOT NULL, 
+    lower_bound_ip_address FLOAT NOT NULL, 
+    upper_bound_ip_address FLOAT NOT NULL, 
+    country VARCHAR NOT NULL 
+
+DROP TABLE IF EXISTS features;
+
+CREATE TABLE features (
+    user_id SERIAL PRIMARY KEY,
+    signup_time TIMESTAMP,
+    purchase_time TIMESTAMP,
+    purchase_value FLOAT,
+    device_id VARCHAR,
+    source VARCHAR,
+    browser VARCHAR,
+    sex VARCHAR,
+    age INTEGER,
+    ip_address VARCHAR,
+    class INTEGER,
+    lower_bound_ip_address BIGINT,
+    upper_bound_ip_address BIGINT,
+    country VARCHAR
 );