TEAM-82-H2O-NOVA

H2O-NOVA/README.md

@@ -0,0 +1,77 @@
+
+
+
+---
+
+#### **Project Name**:  
+**QATRA matters** 💧 - AI-Powered Water Consumption Management
+
+---
+![looooogo (1)](https://github.com/user-attachments/assets/163e9365-a86d-4d4d-8af5-921024c72ef8)
+
+#### **Problem**:
+Water scarcity 🌍 and inefficient water management  are growing concerns, especially in regions like Morocco 🇲🇦. Despite existing efforts, many households  and businesses lack tools to effectively track, manage, and reduce water consumption. Overuse, leaks, and wastage  are common problems that lead to high bills  and environmental strain . There is a need for a comprehensive, user-friendly solution to monitor, forecast, and optimize water usage.
+
+---
+
+#### **Solution**:
+**SmartWater** leverages **AI-driven technology**  to provide real-time monitoring , predictive forecasting 📊, and personalized alerts to help users optimize water consumption. Our solution empowers individuals , businesses , and government bodies  to adopt sustainable water usage practices 🌱, reduce wastage 💦, and ultimately save resources 💰 and costs. The integration of machine learning models, a chatbot 🗣, and OCR technology  makes our solution easy to use, efficient, and accessible for everyone.
+
+---
+#### **Workflow**:
+
+![Flowchart Building](https://github.com/user-attachments/assets/b2cc6812-3a6b-451c-99dc-75189dae6799)
+![Flowchart Building (1)](https://github.com/user-attachments/assets/e2c66b4b-f5da-4e14-bd14-0c5c3ddffc86)
+
+
+#### **Key Features**:
+1. **Real-Time Water Consumption Tracking** 📉:
+   - Monitors water usage from traditional water meters 💧.
+   - Provides visualizations  of current and historical water consumption data.
+
+2. **AI-Powered Forecasting** 🔮:
+   - Uses **LSTM (Long Short-Term Memory)** models  for predicting future water usage based on historical data 📈.
+
+3. **Threshold Alerts** 🚨:
+   - Notifies users when their water consumption exceeds a predefined limit .
+   - Encourages users to take corrective actions and reduce wastage.
+
+4. **Interactive Chatbot** :
+   - Provides personalized recommendations for reducing water consumption.
+
+5. **Water Bill Integration** 🧾:
+   - Scans and extracts data from water bills using **OCR** technology  to automatically update and compare consumption data.
+
+
+---
+
+#### **Impact**:
+1. **Sustainability** 🌍:
+   - Promotes responsible water consumption and contributes to long-term environmental sustainability 🌱.
+
+2. **Cost Savings** 💸:
+   - Helps users reduce water bills by providing insights on where and how to save water .
+
+4. **Local Adaptation** :
+   - The chatbot will be enhanced with **Moroccan Arabic** 🗣️ to ensure accessibility and engagement for local users, making it easier for Moroccans to adopt sustainable water practices.
+
+5. **Scalability** 📈:
+   - Our solution can be scaled to address the needs of both urban and rural areas, impacting a wide range of users across Morocco.
+
+---
+
+#### **Technologies Used**:
+- **Machine Learning**: **LSTM (Long Short-Term Memory)** for predictive modeling 📉.
+- **OCR**: **Tesseract** or **Google Vision API** for water bill scanning 🧾.
+- **Web App**: Built using **Streamlit** for real-time dashboards and analytics 📊.
+- **Chatbot**: Integrated with **Dialogflow** for natural language interactions 🗣️.
+
+---
+
+#### **Future Improvements**:
+- **IoT Integration** : Compatibility with smart water meters for real-time, automated data collection 🧭.
+- **Advanced AI Features** : More sophisticated anomaly detection algorithms for leaks or inefficiencies 🏚️.
+
+#### **Links**:
+* DEMO:https://www.youtube.com/watch?v=ZPSJRCKDiYc&ab_channel=MeriameBoulaghrous
+* PITCH:https://www.youtube.com/watch?v=1HCLl-bTkgU&ab_channel=MeriameBoulaghrous

H2O-NOVA/admin/test_water_consumtion_api.py

@@ -0,0 +1,38 @@
+import requests
+import matplotlib.pyplot as plt
+from datetime import datetime
+
+# Define the API endpoint
+api_url = "https://8791-34-40-149-146.ngrok-free.app/predict"
+
+payload = {
+    "days_to_predict": 7  # Predict the next 7 days
+}
+
+try:
+    # Send a POST request to the API
+    response = requests.post(api_url, json=payload)
+    response.raise_for_status()  # Raise an error for bad status codes
+    data = response.json()
+
+    # Extract predictions and timestamps from the response
+    predictions = data["predictions"]
+    timestamps = data["timestamps"]
+
+    # Convert timestamps to datetime objects
+    timestamps = [datetime.strptime(ts, "%Y-%m-%d") for ts in timestamps]
+
+    # Plot the predictions
+    plt.figure(figsize=(10, 6))
+    plt.plot(timestamps, predictions, marker="o", label="Predicted Meter Reading", color="orange")
+    plt.title("Forecasted Meter Reading")
+    plt.xlabel("Date")
+    plt.ylabel("Meter Reading")
+    plt.xticks(rotation=45)
+    plt.grid(True)
+    plt.legend()
+    plt.tight_layout()
+    plt.show()
+
+except requests.exceptions.RequestException as e:
+    print(f"An error occurred while fetching predictions: {e}")

H2O-NOVA/user_app/README.md

@@ -0,0 +1,19 @@
+# User App
+
+## Description
+
+Streamlit app to track water consumption
+
+## Installation
+
+1. **Install requirements:**
+   ```bash
+   python -m venv venv
+   source venv/bin/activate
+   pip install -r requirements.txt
+    ```
+
+2. **Run the app:**
+   ```bash
+   streamlit run app.py
+    ```

H2O-NOVA/user_app/app.py

@@ -0,0 +1,95 @@
+import altair as alt
+import pandas as pd
+import streamlit as st
+
+THRESHOLD = 10
+materials = ['Shower', 'Bathtub', 'Faucet', 'Washing Machine', 'Dishwasher',
+             'Water Heater', 'Irrigation System', 'Water Filter', 'Other']
+
+
+df = pd.read_csv("data/user_data.csv")
+df_melted = df.drop(columns=["Price"]).melt("Month", var_name="Category", value_name="Liters")
+
+user_name = "John Bob"
+current_water_usage = df.iloc[-1]['User Water Usage (liters)']
+city_threshold = f"{THRESHOLD} liters"
+
+
+# App layout
+st.set_page_config(layout="wide")
+st.title("Water Usage")
+
+# User details
+col1, col2 = st.columns(2)
+with col1:
+    st.subheader("User Details")
+    st.info(f"**First Name:** {user_name.split()[0]}")
+    st.info(f"**Last Name:** {user_name.split()[1]}")
+
+with col2:
+    st.subheader("Last month Water Usage")
+    st.warning(f"**Water Usage:** {current_water_usage} liters")
+    st.warning(f"**City Threshold:** {city_threshold}")
+
+
+# Users material
+st.subheader("Select Water-Related Materials/Devices")
+
+if "selected_materials" not in st.session_state:
+    st.session_state.selected_materials = []
+
+# Create checkboxes for each material
+cols = st.columns(3)
+
+for i, material in enumerate(materials):
+    col_idx = i % 3
+    with cols[col_idx]:
+        if st.checkbox(material, value=(material in st.session_state.selected_materials)):
+            # Add to session state if selected
+            if material not in st.session_state.selected_materials:
+                st.session_state.selected_materials.append(material)
+        else:
+            # Remove from session state if unchecked
+            if material in st.session_state.selected_materials:
+                st.session_state.selected_materials.remove(material)
+
+# Plots
+col1, col2 = st.columns(2)
+
+st.header("Track your water usage")
+
+chart = (
+    alt.Chart(df_melted)
+    .mark_line(point=True)
+    .encode(
+        x=alt.X("Month:T", title="Month"),
+        y=alt.Y("Liters:Q", title="Liters"),
+        color=alt.Color("Category:N", title="Category"),
+        tooltip=["Month:T", "Category:N", "Liters:Q"],
+    )
+    .properties(
+        width=800,
+        height=400,
+        title="Water Usage Evolution"
+    )
+    .interactive()
+)
+
+price_chart = (
+    alt.Chart(df)
+    .mark_line(color="red", point=True)
+    .encode(
+        x=alt.X("Month:T", title="Month"),
+        y=alt.Y("Price:Q", title="Price (in MAD)"),
+        tooltip=["Month:T", "Price:Q"],
+    )
+    .properties(
+        width=800,
+        height=200,
+        title="Monthly Price Paid (MAD)",
+    )
+    .interactive()
+)
+
+st.altair_chart(price_chart, use_container_width=True)
+st.altair_chart(chart, use_container_width=True)

H2O-NOVA/user_app/constants.py

@@ -0,0 +1,2 @@
+DATA_PATH = "data/user_data.csv"
+THRESHOLD = 10

H2O-NOVA/user_app/data/user_data.csv

@@ -0,0 +1,13 @@
+Month,User Water Usage (liters),City Threshold (liters),Price
+2024-01-31,51,60,190.8
+2024-02-29,36,65,175.8
+2024-03-31,67,55,202.14
+2024-04-30,42,60,173.45
+2024-05-31,55,70,193.2
+2024-06-30,73,55,205.45
+2024-07-31,79,65,210.15
+2024-08-31,64,60,200.2
+2024-09-30,46,45,177.8
+2024-10-31,51,50,190.8
+2024-11-30,36,60,175.8
+2024-12-31,21,45,144.8

H2O-NOVA/user_app/data_utils.py

@@ -0,0 +1,30 @@
+import calendar
+
+import pandas as pd
+from constants import DATA_PATH, THRESHOLD
+
+
+def update_data(price, water_usage, city_threshold=THRESHOLD):
+    df = pd.read_csv(DATA_PATH)
+    # Extract the last row's month (assuming the 'Month' column is in YYYY-MM-DD format)
+    last_row_date = pd.to_datetime(df['Month'].iloc[-1])
+
+    # Calculate the next month and handle year change if necessary
+    next_month = (last_row_date.month % 12) + 1
+    next_year = last_row_date.year if next_month != 1 else last_row_date.year + 1
+
+    _, last_day = calendar.monthrange(next_year, next_month)
+    next_month_last_day = f"{next_year}-{next_month:02d}-{last_day:02d}"
+
+    next_month_data = pd.DataFrame({
+        'Month': [next_month_last_day],
+        'User Water Usage (liters)': [water_usage],
+        'City Threshold (liters)': [city_threshold],
+        'Price': [price]
+    })
+
+    # Concatenate the new row to the existing DataFrame
+    df = pd.concat([df, next_month_data], ignore_index=True)
+    df.to_csv(DATA_PATH, index=False)
+
+    return df

H2O-NOVA/user_app/extractor.py

@@ -0,0 +1,65 @@
+import re
+
+import pytesseract
+from PyPDF2 import PdfReader
+
+pytesseract.pytesseract.tesseract_cmd = '/usr/bin/tesseract'
+
+
+class PpdExtractor:
+    @staticmethod
+    def extract(pdf_path, page_number=2):
+        reader = PdfReader(pdf_path)
+
+        assert page_number >= 1 or page_number <= len(reader.pages)
+
+        extracted_text = reader.pages[page_number - 1].extract_text()
+        price_match = re.search(r'Total TTC\s+(\d+,\d+)', extracted_text)
+        price = None
+        if price_match:
+            # Replace the comma with a dot and convert to float
+            price = float(price_match.group(1).replace(",", "."))
+
+        water_consumption = re.findall(r'(\d+)\s+Tranche', extracted_text)
+        if len(water_consumption) == 0:
+            water_consumption = 0
+
+        elif len(water_consumption) >= 2:
+            # Take only the first which corresponds to water consumption, other is for electricity
+            water_consumption = int(water_consumption[0])
+
+        return price, water_consumption
+
+
+class ImageExtractor:
+    def __init__(self, image):
+        self.width, self.height = image.size
+        self.image = image
+
+    def extract_water_usage(self, text):
+        # Extract water usage
+        try:
+            water_usage = re.findall(r"\d{2}/\d{2}/\d{4}\s+(\d+)", text.split("N°")[1])[-1]
+            water_usage = int(water_usage)
+        except Exception:
+            water_usage = None
+
+        return water_usage
+
+    def extract_price(self, text):
+        # Number that comes after "Sous total TTC"
+        price = re.search(r"Sous total TTC[:\s]+([\d,]+(?:\.\d{1,2})?)", text)
+        try:
+            price = price.group(1)
+            price = float(price.replace(",", "."))
+        except Exception:
+            price = None
+
+        return price
+
+    def extract(self):
+        text = pytesseract.image_to_string(self.image)
+        price = self.extract_price(text)
+        water_usage = self.extract_water_usage(text)
+
+        return price, water_usage

H2O-NOVA/user_app/packages.txt

@@ -0,0 +1,3 @@
+tesseract-ocr
+libtk8.6
+poppler-utils