From ae736779a63d43d62ead77d3ddafab3306b2acb4 Mon Sep 17 00:00:00 2001
From: wabinyai <rajacastro@gmail.com>
Date: Thu, 14 Nov 2024 11:23:39 +0300
Subject: [PATCH 01/17] auto reporting **_WHAT DOES THIS PR DO?_** This PR
 implements a new API endpoint for generating comprehensive air quality
 reports with spatial analysis. - [ ] Item 1 - [ ] Item 2

**_WHAT ISSUES ARE RELATED TO THIS PR?_**

- Jira cards
    - []
- GitHub issues
    - Closes #<enter GitHub issue number here>

**_HOW DO I TEST OUT THIS PR?_**

- [ ] Analytics: [Link to Analytics README](https://github.com/airqo-platform/AirQo-api/tree/staging/src/analytics#readme)
- [ ] Auth Service: [Link to Auth Service README](https://github.com/airqo-platform/AirQo-api/tree/staging/src/auth-service#readme)
- [ ] Calibrate: [Link to Calibrate README](https://github.com/airqo-platform/AirQo-api/tree/staging/src/calibrate#readme)
- [ ] Data Proxy: [Link to Data Proxy README](https://github.com/airqo-platform/AirQo-api/tree/staging/src/data-proxy#readme)
- [ ] Device Monitoring: [Link to Device Monitoring README](https://github.com/airqo-platform/AirQo-api/tree/staging/src/device-monitoring#readme)
- [ ] Device Registry: [Link to Device Registry README](https://github.com/airqo-platform/AirQo-api/tree/staging/src/device-registry#readme)
- [ ] Device Status: [Link to Device Status README](https://github.com/airqo-platform/AirQo-api/tree/staging/src/device-status#readme)
- [ ] Device Uptime: [Link to Device Uptime README](https://github.com/airqo-platform/AirQo-api/tree/staging/src/device-uptime#readme)
- [ ] Exceedances: [Link to Exceedances README](https://github.com/airqo-platform/AirQo-api/tree/staging/src/exceedances#readme)
- [ ] Firebase: [Link to Firebase README](https://github.com/airqo-platform/AirQo-api/tree/staging/src/firebase)
- [ ] GP Model: [Link to GP Model README](https://github.com/airqo-platform/AirQo-api/tree/staging/src/gp-model#readme)
- [ ] Incentives: [Link to Incentives README](https://github.com/airqo-platform/AirQo-api/tree/staging/src/incentives#readme)
- [ ] Kafka Connectors: [Link to Kafka Connectors](https://github.com/airqo-platform/AirQo-api/tree/staging/src/kafka-connectors)
- [ ] Locate: [Link to Locate README](https://github.com/airqo-platform/AirQo-api/tree/staging/src/locate#readme)
- [ ] Metadata: [Link to Metadata README](https://github.com/airqo-platform/AirQo-api/tree/staging/src/meta-data#readme)
- [ ] Predict: [Link to Predict README](https://github.com/airqo-platform/AirQo-api/tree/staging/src/predict#readme)
- [ ] Spatial: [Link to Spatial](https://github.com/airqo-platform/AirQo-api/tree/staging/src/spatial)
- [ ] View: [Link to View README](https://github.com/airqo-platform/AirQo-api/tree/staging/src/view#readme)
- [ ] Workflows: [Link to Workflows README](https://github.com/airqo-platform/AirQo-api/tree/staging/src/workflows#readme)

**_WHICH ENDPOINTS SHOULD BE READY FOR TESTING?:_**

- [ ] Endpoint 1
- [ ] Endpoint 2

**_ARE THERE ANY RELATED PRs?_**

- [ ] Related PR 1
- [ ] Related PR 2
---
 src/spatial/configure.py                 |   4 +-
 src/spatial/controllers/controllers.py   |   5 +
 src/spatial/models/report_datafetcher.py | 214 +++++++++++++++++++++++
 src/spatial/requirements.txt             |   8 +-
 src/spatial/views/report_view.py         |  42 +++++
 5 files changed, 271 insertions(+), 2 deletions(-)
 create mode 100644 src/spatial/models/report_datafetcher.py
 create mode 100644 src/spatial/views/report_view.py

diff --git a/src/spatial/configure.py b/src/spatial/configure.py
index 561196e70b..b09ccc2033 100644
--- a/src/spatial/configure.py
+++ b/src/spatial/configure.py
@@ -25,7 +25,9 @@ class Config:
         "BIGQUERY_SATELLITE_MODEL_PREDICTIONS"
     )
 
-
+    HUGGING_FACE_TOKEN = os.getenv("HUGGING_FACE_TOKEN")
+    GOOGLE_API_KEY = os.getenv("GOOGLE_API_KEY")
+    OPENAI_API_KEY = os.getenv("OPENAI_API_KEY")
 class ProductionConfig(Config):
     DEBUG = False
     TESTING = False
diff --git a/src/spatial/controllers/controllers.py b/src/spatial/controllers/controllers.py
index df1570f519..26ebc4b86d 100644
--- a/src/spatial/controllers/controllers.py
+++ b/src/spatial/controllers/controllers.py
@@ -12,6 +12,7 @@
 from views.satellite_predictions import SatellitePredictionView
 from views.site_category_view import SiteCategorizationView
 from views.site_selection_views import SiteSelectionView
+from views.report_view import ReportView
 
 
 controller_bp = Blueprint("controller", __name__)
@@ -66,3 +67,7 @@ def site_selection():
 @controller_bp.route("/satellite_prediction", methods=["POST"])
 def get_satellite_prediction():
     return SatellitePredictionView.make_predictions()
+
+@controller_bp.route("/air_quality_report", methods=["POST"])
+def fetch_air_quality():
+    return ReportView.generate_air_quality_report()
\ No newline at end of file
diff --git a/src/spatial/models/report_datafetcher.py b/src/spatial/models/report_datafetcher.py
new file mode 100644
index 0000000000..52317650ae
--- /dev/null
+++ b/src/spatial/models/report_datafetcher.py
@@ -0,0 +1,214 @@
+import requests
+import openai
+from transformers import AutoModelForCausalLM, AutoTokenizer
+from huggingface_hub import login
+from configure import Config
+import google.generativeai as genai 
+
+
+# Configure API keys
+GOOGLE_API_KEY = Config.GOOGLE_API_KEY
+genai.configure(api_key=GOOGLE_API_KEY)
+hf_token = Config.HUGGING_FACE_TOKEN
+
+if hf_token:
+    login(hf_token)
+else:
+    print("Hugging Face token is missing. Set the 'HUGGING_FACE_TOKEN' environment variable.")
+
+class DataFetcher:
+    @staticmethod
+    def fetch_air_quality_data_a(grid_id, start_time, end_time):
+        token = Config.AIRQO_API_TOKEN  
+        if token is None:
+            print("Error: AIRQO_API_TOKEN environment variable is not set.")
+            return None
+
+        url = f"https://platform.airqo.net/api/v2/analytics/grid/report?token={token}"
+        payload = {"grid_id": grid_id, "start_time": start_time, "end_time": end_time}
+
+        try:
+            response = requests.post(url, json=payload)
+            response.raise_for_status()
+            return response.json()
+        except requests.exceptions.HTTPError as http_err:
+            print(f"HTTP error occurred: {http_err}")
+        except requests.exceptions.RequestException as req_err:
+            print(f"Request error occurred: {req_err}")
+        except ValueError as json_err:
+            print(f"JSON decoding error: {json_err}")
+
+        return None
+
+class AirQualityReport:  
+    def __init__(self, data): 
+        self.data = data
+        self.grid_name = data.get('airquality', {}).get('sites', {}).get('grid name', [None])
+        self.annual_data = data.get('airquality', {}).get('annual_pm', [None])[0]
+        self.daily_mean_data = data.get('airquality', {}).get('daily_mean_pm', [])
+        self.diurnal = data.get('airquality', {}).get('diurnal', [])
+        self.monthly_data = data.get('airquality', {}).get('site_monthly_mean_pm', [])
+        self.monthly_name_data = data.get('airquality', {}).get('pm_by_month_name', [])
+        main_site_info = self.monthly_data[0] if self.monthly_data else {}
+        self.main_site = main_site_info.get('site_name')
+        self.site_names = [item.get('site_name', None) for item in self.data.get('airquality', {}).get('site_annual_mean_pm', [])]
+        self.site_latitude = main_site_info.get('site_latitude')
+        self.site_longitude = main_site_info.get('site_longitude')
+        self.num_sites = data.get('airquality', {}).get('sites', {}).get('number_of_sites') 
+        self.starttime = data.get('airquality', {}).get('period', {}).get('startTime', '')[:10]
+        self.endtime = data.get('airquality', {}).get('period', {}).get('endTime', '')[:10] 
+
+        self.annual_pm2_5_calibrated_value = self.annual_data.get("pm2_5_calibrated_value")
+        self.annual_pm10_calibrated_value = self.annual_data.get("pm10_calibrated_value")
+
+        # Finding the minimum and maximum values
+        self.daily_min_pm2_5 = min(self.daily_mean_data, key=lambda x: x['pm2_5_calibrated_value'])
+        self.daily_max_pm2_5 = max(self.daily_mean_data, key=lambda x: x['pm2_5_calibrated_value'])
+        
+
+
+        # Initialize models once in the constructor
+        self.gemini_model = genai.GenerativeModel('gemini-pro')
+        openai.api_key = Config.OPENAI_API_KEY
+
+    def _prepare_base_info(self):
+        return (
+            f"The air quality report is for {self.grid_name} for the period of {self.starttime} to {self.endtime}. "
+            f"These air quality monitoring sites are {self.site_names} and measure PM2.5 and PM10, "
+            f"at coordinates {self.site_latitude}°N, {self.site_longitude}°E. "
+            f"The annual PM2.5 concentration averages {self.annual_data} µg/m³."
+        )
+
+    def _generate_prompt(self, audience):
+        base_info = self._prepare_base_info()
+        if audience == "researcher":
+            return (
+                f"{audience}"
+                
+                f"Generate a comprehensive air quality assessment report for {self.grid_name} for the period of {self.starttime} to {self.endtime}. Begin with a detailed introduction (100-130 words) covering the city's geographical location, climate characteristics, population density, and major pollution sources. "
+                f"{base_info} include the period under review."
+                f"Daily mean measurements show: {self.daily_mean_data}. "
+                f"Diurnal patterns indicate: {self.diurnal}. Monthly trends reveal: {self.monthly_data}. "
+                f"Provide a thorough analysis of spatial and temporal air quality variations, identify pollution hotspots and clean zones, examine seasonal patterns, and assess compliance with WHO guidelines. "
+                f"Conclude with actionable recommendations for air quality improvement and public health protection. Data source: AirQo monitoring network."
+            )
+        elif audience == "policymaker":
+            return (
+                f"{audience}"
+                f"Create an executive summary of air quality conditions in {self.grid_name} for the period of {self.starttime} to {self.endtime}. for policy decision-making. Begin with key findings and their policy implications (50-75 words). "
+                f"{base_info} include the period under review."
+                f"Highlight critical trends: {self.monthly_data}. Diurnal patterns indicate: {self.diurnal}. "
+                f"Focus on: 1) Areas exceeding air quality standards, 2) Population exposure risk assessment, "
+                f"3) Economic implications of poor air quality. Present clear, actionable policy recommendations with expected outcomes and implementation timeframes. "
+                f"Include cost-benefit considerations and potential regulatory measures. Data source: AirQo monitoring network."
+            )
+        elif audience == "general public":
+            return (
+                f"{audience}"
+                f"{base_info} include the period under review."
+                f"Create a clear, easy-to-understand report about air quality in {self.grid_name} for the period of {self.starttime} to {self.endtime}. Start with a simple explanation of why air quality matters for public health. "
+                f"We have {self.num_sites} air quality monitors in your area. The average PM2.5 level this year is {self.annual_data} µg/m³. "
+                f"Diurnal patterns indicate: {self.diurnal}. Monthly trends reveal: {self.monthly_data}. "
+                f"Explain what these numbers mean for daily activities. Include: 1) When air quality is best and worst during the day, "
+                f"2) Which areas have better or worse air quality, 3) Simple steps people can take to protect their health, "
+                f"4) How to access daily air quality updates. Use plain language and avoid technical terms. "
+                f"Add practical tips for reducing exposure to air pollution. Data source: AirQo monitoring network."
+            )
+        else:
+            raise ValueError("Invalid audience type. Please specify 'researcher', 'policymaker', or 'general public'.")
+
+    def generate_report_with_gemini(self, audience):
+        prompt = self._generate_prompt(audience)
+        response = self.gemini_model.generate_content(prompt)
+        gemini_output = response.text
+        return self._prepare_report_json(gemini_output)
+
+    def generate_report_with_openai(self, audience):
+        prompt = self._generate_prompt(audience)
+        response = openai.ChatCompletion.create(
+            model="gpt-3.5-turbo",
+            messages=[{"role": "user", "content": prompt}]
+        )
+        openai_output = response.choices[0].message['content']
+        return self._prepare_report_json(openai_output)
+
+    # Use non-LLM template text as report content
+    def generate_report_template_witout_LLM(self, audience):
+        prompt = self._generate_prompt(audience)
+        report_content = prompt
+        return self._prepare_report_json(report_content)
+    
+    def generate_report_without_llm(self):
+    # Determine peak time and least PM2.5 values
+        peak_data = max(self.diurnal, key=lambda x: x['pm2_5_calibrated_value'])
+        peak_time = peak_data['hour']
+        peak_pm2_5 = peak_data['pm2_5_calibrated_value']
+
+        least_data = min(self.diurnal, key=lambda x: x['pm2_5_calibrated_value'])
+        least_pm2_5 = least_data['pm2_5_calibrated_value']
+        least_pm2_5_time = least_data['hour']
+
+
+        
+        introduction = (
+            f"The air quality report for {self.grid_name} covers the period from {self.starttime} to {self.endtime}. "
+            f"The {self.num_sites} monitored sites include: {', '.join(self.site_names)}. "
+            f"Measurements are taken for PM2.5 and PM10 concentrations. " 
+            f"The annual average PM2.5 concentration is {self.annual_pm2_5_calibrated_value} µg/m³."
+        )
+        
+        diurnal_description = (
+            f"Diurnal patterns observed include the following: {self.diurnal}. "
+            f"These patterns provide insight into air quality fluctuations throughout the day. "
+            f"The peak in PM2.5 {peak_pm2_5} levels occurs around {peak_time}:00 hr, indicating a period of higher pollution, often associated with increased activity or traffic. "
+            f"Conversely, the period with the least PM2.5 {least_pm2_5}  µg/m³ levels is around {least_pm2_5_time} :00 hr , "
+            f"which usually represents a period of lower activity or better atmospheric dispersion."
+            f"Understanding the patterns of pollution and their impacts on public health is crucial for effective environmental management and policy-making. "
+            f"Throughout this report, we will explore key trends in PM2.5 and PM10 concentrations, the diurnal variations, and the impact of these levels on air quality across the region."
+
+        )
+
+        daily_mean_description = (
+            f"Daily mean PM2.5 measurements during the period were recorded as follows: {self.daily_mean_data}. "
+            f"This data reveals variations in air quality on a day-to-day basis."
+        )
+
+        site_pm25_description = (
+            f"The concentration of PM2.5 across different sites shows variability: "
+            f"{', '.join([f'{site} with PM2.5 levels' for site in self.site_names])}. "
+            f"These variations indicate site-specific air quality differences for the known grids."
+        )
+        conclusion = (
+            f"Overall, the air quality report highlights the importance of monitoring and understanding the patterns of PM2.5 and PM10 concentrations in the {self.grid_name} "
+            f"The analysis of the data reveals that air quality varies significantly over time, with periods of both moderate and unhealthy conditions. "
+            f"It’s observed that these fluctuations may be influenced by various factors, including seasonal changes. For instance, the washout effect during the rainy"
+            f" season could potentially contribute to these variations. Specifically, for the period from   {self.starttime} to {self.endtime},"
+            f" the PM2.5 raw values ranged from {self.daily_min_pm2_5['pm2_5_raw_value']} µg/m³ on {self.daily_min_pm2_5['date']} to {self.daily_max_pm2_5['pm2_5_raw_value']} µg/m³ on {self.daily_max_pm2_5['date']}. respectively."
+            f"This pattern underscores the importance of continuous monitoring and the implementation of"
+            f"effective interventions to maintain air quality within safe limits. Ensuring good air quality is crucial for "
+            f"the well-being of both residents and visitors. Therefore, it’s imperative to adopt long-term"
+            f"strategies and measures that can effectively mitigate the impact of factors leading to poor airquality."
+            f"In conclusion, continuous monitoring, timely intervention, and effective policies are key to maintaining good air quality and safeguarding public health. "
+        )
+
+        report_content = (
+            f"{introduction}\n\n"
+            f"{diurnal_description}\n\n"
+            f"{daily_mean_description}\n\n"
+            f"{site_pm25_description}\n\n"
+            f"{conclusion}"
+        )
+
+
+        return self._prepare_report_json(report_content)
+
+    def _prepare_report_json(self, report_content):
+        return {
+            "grid_name": self.grid_name,
+            "main_site": self.main_site, 
+            "annual_data": self.annual_data,
+            "daily_mean_data": self.daily_mean_data,
+            "diurnal": self.diurnal,
+            "monthly_data": self.monthly_data,
+            "report": report_content
+        }
\ No newline at end of file
diff --git a/src/spatial/requirements.txt b/src/spatial/requirements.txt
index 8f4fda8f10..d584220be0 100644
--- a/src/spatial/requirements.txt
+++ b/src/spatial/requirements.txt
@@ -22,4 +22,10 @@ scikit-learn~=1.5.2
 gcsfs~=2024.9.0.post1
 joblib~=1.4.2
 lightgbm~=4.1.0
-numpy~=1.25.2
\ No newline at end of file
+numpy~=1.25.2
+torch
+transformers
+datasets
+sentencepiece
+huggingface_hub
+google-generativeai
\ No newline at end of file
diff --git a/src/spatial/views/report_view.py b/src/spatial/views/report_view.py
new file mode 100644
index 0000000000..c8015ea963
--- /dev/null
+++ b/src/spatial/views/report_view.py
@@ -0,0 +1,42 @@
+from flask import request, jsonify
+from models.report_datafetcher import DataFetcher, AirQualityReport
+
+class ReportView:
+    @staticmethod
+    def generate_air_quality_report():
+        """Fetch air quality data and generate a report based on grid_id, start_time, end_time, and audience."""
+        print('Processing request to generate air quality report...')
+
+        # Extract data from the request
+        data = request.json
+        grid_id = data.get("grid_id")
+        start_time = data.get("start_time")
+        end_time = data.get("end_time")
+        audience = data.get("audience", "general public")  # Default to "general public" if audience is not provided
+
+        # Validate input parameters
+        if not all([grid_id, start_time, end_time, audience]):
+            return jsonify({"error": "Missing required parameters: grid_id, start_time, end_time, audience"}), 400
+
+        # Fetch air quality data
+        air_quality_data = DataFetcher.fetch_air_quality_data_a(grid_id, start_time, end_time)
+
+        if air_quality_data is None:
+            return jsonify({"error": "No data found for the given parameters"}), 404
+
+        # Create an air quality report
+        report = AirQualityReport(air_quality_data)
+        
+        # Generate the report with the specified audience
+    #    json_report = report.generate_report_with_gemini(audience)          # using google gemini
+        json_report = report.generate_report_without_llm()
+    #    json_report = report.generate_report_template_witout_LLM(audience) # without LLM
+    #    json_report = report.generate_report_with_openai(audience)         # Using openai api 
+    
+        
+        if json_report is None:
+            return jsonify({"error": "Failed to generate report"}), 500
+
+        return jsonify({"report": json_report}), 200
+ 
+ 
\ No newline at end of file

From 780463a7e4b2b7de16a908468c6a2de1f5aae5a6 Mon Sep 17 00:00:00 2001
From: wabinyai <rajacastro@gmail.com>
Date: Thu, 14 Nov 2024 11:45:35 +0300
Subject: [PATCH 02/17] review

---
 src/spatial/views/report_view.py | 91 ++++++++++++++++++++++----------
 1 file changed, 62 insertions(+), 29 deletions(-)

diff --git a/src/spatial/views/report_view.py b/src/spatial/views/report_view.py
index c8015ea963..85eb11bfc0 100644
--- a/src/spatial/views/report_view.py
+++ b/src/spatial/views/report_view.py
@@ -1,42 +1,75 @@
 from flask import request, jsonify
 from models.report_datafetcher import DataFetcher, AirQualityReport
+import logging
 
 class ReportView:
     @staticmethod
     def generate_air_quality_report():
         """Fetch air quality data and generate a report based on grid_id, start_time, end_time, and audience."""
-        print('Processing request to generate air quality report...')
+        logging.info('Processing request to generate air quality report...')
+        
+        # Extract and validate request data
+        data = ReportView._get_request_data()
+        if isinstance(data, dict) and data.get("error"):
+            return jsonify(data), 400
 
-        # Extract data from the request
-        data = request.json
-        grid_id = data.get("grid_id")
-        start_time = data.get("start_time")
-        end_time = data.get("end_time")
-        audience = data.get("audience", "general public")  # Default to "general public" if audience is not provided
+        grid_id = data["grid_id"]
+        start_time = data["start_time"]
+        end_time = data["end_time"]
+        audience = data.get("audience", "general public")  # Default value if not provided
 
-        # Validate input parameters
-        if not all([grid_id, start_time, end_time, audience]):
-            return jsonify({"error": "Missing required parameters: grid_id, start_time, end_time, audience"}), 400
+        try:
+            # Fetch air quality data
+            air_quality_data = DataFetcher.fetch_air_quality_data_a(grid_id, start_time, end_time)
+            if not air_quality_data:
+                logging.warning(f"No data found for grid_id: {grid_id}, start_time: {start_time}, end_time: {end_time}")
+                return jsonify({"error": "No data found for the given parameters"}), 404
 
-        # Fetch air quality data
-        air_quality_data = DataFetcher.fetch_air_quality_data_a(grid_id, start_time, end_time)
+            # Generate the air quality report
+            report = AirQualityReport(air_quality_data)
+            json_report = ReportView._generate_report(report, audience)
+            if not json_report:
+                logging.error("Failed to generate the air quality report.")
+                return jsonify({"error": "Failed to generate report"}), 500
 
-        if air_quality_data is None:
-            return jsonify({"error": "No data found for the given parameters"}), 404
+            return jsonify({"report": json_report}), 200
 
-        # Create an air quality report
-        report = AirQualityReport(air_quality_data)
-        
-        # Generate the report with the specified audience
-    #    json_report = report.generate_report_with_gemini(audience)          # using google gemini
-        json_report = report.generate_report_without_llm()
-    #    json_report = report.generate_report_template_witout_LLM(audience) # without LLM
-    #    json_report = report.generate_report_with_openai(audience)         # Using openai api 
-    
-        
-        if json_report is None:
-            return jsonify({"error": "Failed to generate report"}), 500
+        except Exception as e:
+            logging.exception("Unexpected error occurred during report generation")
+            return jsonify({"error": "An unexpected error occurred"}), 500
 
-        return jsonify({"report": json_report}), 200
- 
- 
\ No newline at end of file
+    @staticmethod
+    def _get_request_data():
+        """Helper function to extract and validate request data."""
+        try:
+            data = request.get_json()
+            if not data:
+                logging.error("Invalid input: JSON data expected.")
+                return {"error": "Invalid input. JSON data expected."}
+            
+            required_params = ["grid_id", "start_time", "end_time"]
+            missing_params = [param for param in required_params if not data.get(param)]
+            if missing_params:
+                logging.error(f"Missing required parameters: {', '.join(missing_params)}")
+                return {"error": f"Missing required parameters: {', '.join(missing_params)}"}
+            
+            return data
+        except Exception as e:
+            logging.exception("Error parsing input data.")
+            return {"error": "Error parsing input data."}
+
+    @staticmethod
+    def _generate_report(report, audience):
+        """Helper function to generate report based on the audience type."""
+        try:
+            # Generate the report with the specified audience
+            # You can extend this logic to switch between different report generation methods
+            json_report = report.generate_report_without_llm()  # Example of one option
+            # json_report = report.generate_report_with_gemini(audience)  # Using Google Gemini (if applicable)
+            # json_report = report.generate_report_with_openai(audience)  # Using OpenAI API (if applicable)
+            # json_report = report.generate_report_template_without_llm(audience)  # Without LLM, audience-specific
+            
+            return json_report
+        except Exception as e:
+            logging.exception("Error generating the report.")
+            return None

From c9c7530d7f0609e1b40d9e7be7c0f69bbf246b26 Mon Sep 17 00:00:00 2001
From: wabinyai <rajacastro@gmail.com>
Date: Thu, 14 Nov 2024 12:14:43 +0300
Subject: [PATCH 03/17] report

---
 src/spatial/views/report_view.py | 91 ++++++++++----------------------
 1 file changed, 29 insertions(+), 62 deletions(-)

diff --git a/src/spatial/views/report_view.py b/src/spatial/views/report_view.py
index 85eb11bfc0..c8015ea963 100644
--- a/src/spatial/views/report_view.py
+++ b/src/spatial/views/report_view.py
@@ -1,75 +1,42 @@
 from flask import request, jsonify
 from models.report_datafetcher import DataFetcher, AirQualityReport
-import logging
 
 class ReportView:
     @staticmethod
     def generate_air_quality_report():
         """Fetch air quality data and generate a report based on grid_id, start_time, end_time, and audience."""
-        logging.info('Processing request to generate air quality report...')
-        
-        # Extract and validate request data
-        data = ReportView._get_request_data()
-        if isinstance(data, dict) and data.get("error"):
-            return jsonify(data), 400
-
-        grid_id = data["grid_id"]
-        start_time = data["start_time"]
-        end_time = data["end_time"]
-        audience = data.get("audience", "general public")  # Default value if not provided
+        print('Processing request to generate air quality report...')
 
-        try:
-            # Fetch air quality data
-            air_quality_data = DataFetcher.fetch_air_quality_data_a(grid_id, start_time, end_time)
-            if not air_quality_data:
-                logging.warning(f"No data found for grid_id: {grid_id}, start_time: {start_time}, end_time: {end_time}")
-                return jsonify({"error": "No data found for the given parameters"}), 404
+        # Extract data from the request
+        data = request.json
+        grid_id = data.get("grid_id")
+        start_time = data.get("start_time")
+        end_time = data.get("end_time")
+        audience = data.get("audience", "general public")  # Default to "general public" if audience is not provided
 
-            # Generate the air quality report
-            report = AirQualityReport(air_quality_data)
-            json_report = ReportView._generate_report(report, audience)
-            if not json_report:
-                logging.error("Failed to generate the air quality report.")
-                return jsonify({"error": "Failed to generate report"}), 500
+        # Validate input parameters
+        if not all([grid_id, start_time, end_time, audience]):
+            return jsonify({"error": "Missing required parameters: grid_id, start_time, end_time, audience"}), 400
 
-            return jsonify({"report": json_report}), 200
+        # Fetch air quality data
+        air_quality_data = DataFetcher.fetch_air_quality_data_a(grid_id, start_time, end_time)
 
-        except Exception as e:
-            logging.exception("Unexpected error occurred during report generation")
-            return jsonify({"error": "An unexpected error occurred"}), 500
+        if air_quality_data is None:
+            return jsonify({"error": "No data found for the given parameters"}), 404
 
-    @staticmethod
-    def _get_request_data():
-        """Helper function to extract and validate request data."""
-        try:
-            data = request.get_json()
-            if not data:
-                logging.error("Invalid input: JSON data expected.")
-                return {"error": "Invalid input. JSON data expected."}
-            
-            required_params = ["grid_id", "start_time", "end_time"]
-            missing_params = [param for param in required_params if not data.get(param)]
-            if missing_params:
-                logging.error(f"Missing required parameters: {', '.join(missing_params)}")
-                return {"error": f"Missing required parameters: {', '.join(missing_params)}"}
-            
-            return data
-        except Exception as e:
-            logging.exception("Error parsing input data.")
-            return {"error": "Error parsing input data."}
+        # Create an air quality report
+        report = AirQualityReport(air_quality_data)
+        
+        # Generate the report with the specified audience
+    #    json_report = report.generate_report_with_gemini(audience)          # using google gemini
+        json_report = report.generate_report_without_llm()
+    #    json_report = report.generate_report_template_witout_LLM(audience) # without LLM
+    #    json_report = report.generate_report_with_openai(audience)         # Using openai api 
+    
+        
+        if json_report is None:
+            return jsonify({"error": "Failed to generate report"}), 500
 
-    @staticmethod
-    def _generate_report(report, audience):
-        """Helper function to generate report based on the audience type."""
-        try:
-            # Generate the report with the specified audience
-            # You can extend this logic to switch between different report generation methods
-            json_report = report.generate_report_without_llm()  # Example of one option
-            # json_report = report.generate_report_with_gemini(audience)  # Using Google Gemini (if applicable)
-            # json_report = report.generate_report_with_openai(audience)  # Using OpenAI API (if applicable)
-            # json_report = report.generate_report_template_without_llm(audience)  # Without LLM, audience-specific
-            
-            return json_report
-        except Exception as e:
-            logging.exception("Error generating the report.")
-            return None
+        return jsonify({"report": json_report}), 200
+ 
+ 
\ No newline at end of file

From 2fd6d9049c35084600b4db2006016fd7e532fd34 Mon Sep 17 00:00:00 2001
From: wabinyai <rajacastro@gmail.com>
Date: Thu, 14 Nov 2024 13:04:51 +0300
Subject: [PATCH 04/17] required

---
 src/spatial/requirements.txt | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/src/spatial/requirements.txt b/src/spatial/requirements.txt
index d584220be0..acdb1a952e 100644
--- a/src/spatial/requirements.txt
+++ b/src/spatial/requirements.txt
@@ -23,9 +23,11 @@ gcsfs~=2024.9.0.post1
 joblib~=1.4.2
 lightgbm~=4.1.0
 numpy~=1.25.2
+numpy
 torch
 transformers
 datasets
 sentencepiece
 huggingface_hub
-google-generativeai
\ No newline at end of file
+google-generativeai
+openai
\ No newline at end of file

From ac0bcae43b1a6df5ab496dfe89439accc19b7c9e Mon Sep 17 00:00:00 2001
From: wabinyai <rajacastro@gmail.com>
Date: Thu, 14 Nov 2024 20:28:54 +0300
Subject: [PATCH 05/17] RE

---
 src/spatial/models/report_datafetcher.py | 9 +++------
 src/spatial/views/report_view.py         | 6 +++---
 2 files changed, 6 insertions(+), 9 deletions(-)

diff --git a/src/spatial/models/report_datafetcher.py b/src/spatial/models/report_datafetcher.py
index 52317650ae..e501f30dce 100644
--- a/src/spatial/models/report_datafetcher.py
+++ b/src/spatial/models/report_datafetcher.py
@@ -73,6 +73,7 @@ def __init__(self, data):
 
     def _prepare_base_info(self):
         return (
+            
             f"The air quality report is for {self.grid_name} for the period of {self.starttime} to {self.endtime}. "
             f"These air quality monitoring sites are {self.site_names} and measure PM2.5 and PM10, "
             f"at coordinates {self.site_latitude}°N, {self.site_longitude}°E. "
@@ -82,9 +83,7 @@ def _prepare_base_info(self):
     def _generate_prompt(self, audience):
         base_info = self._prepare_base_info()
         if audience == "researcher":
-            return (
-                f"{audience}"
-                
+            return (   
                 f"Generate a comprehensive air quality assessment report for {self.grid_name} for the period of {self.starttime} to {self.endtime}. Begin with a detailed introduction (100-130 words) covering the city's geographical location, climate characteristics, population density, and major pollution sources. "
                 f"{base_info} include the period under review."
                 f"Daily mean measurements show: {self.daily_mean_data}. "
@@ -94,7 +93,6 @@ def _generate_prompt(self, audience):
             )
         elif audience == "policymaker":
             return (
-                f"{audience}"
                 f"Create an executive summary of air quality conditions in {self.grid_name} for the period of {self.starttime} to {self.endtime}. for policy decision-making. Begin with key findings and their policy implications (50-75 words). "
                 f"{base_info} include the period under review."
                 f"Highlight critical trends: {self.monthly_data}. Diurnal patterns indicate: {self.diurnal}. "
@@ -103,8 +101,7 @@ def _generate_prompt(self, audience):
                 f"Include cost-benefit considerations and potential regulatory measures. Data source: AirQo monitoring network."
             )
         elif audience == "general public":
-            return (
-                f"{audience}"
+            return ( 
                 f"{base_info} include the period under review."
                 f"Create a clear, easy-to-understand report about air quality in {self.grid_name} for the period of {self.starttime} to {self.endtime}. Start with a simple explanation of why air quality matters for public health. "
                 f"We have {self.num_sites} air quality monitors in your area. The average PM2.5 level this year is {self.annual_data} µg/m³. "
diff --git a/src/spatial/views/report_view.py b/src/spatial/views/report_view.py
index c8015ea963..b31556c91f 100644
--- a/src/spatial/views/report_view.py
+++ b/src/spatial/views/report_view.py
@@ -28,9 +28,9 @@ def generate_air_quality_report():
         report = AirQualityReport(air_quality_data)
         
         # Generate the report with the specified audience
-    #    json_report = report.generate_report_with_gemini(audience)          # using google gemini
-        json_report = report.generate_report_without_llm()
-    #    json_report = report.generate_report_template_witout_LLM(audience) # without LLM
+        json_report = report.generate_report_with_gemini(audience)          # using google gemini
+    #    json_report = report.generate_report_without_llm()
+    #    json_report = report.generate_text_with_gpt_neo(audience) # without LLM
     #    json_report = report.generate_report_with_openai(audience)         # Using openai api 
     
         

From 6ec42596be4bda350f83b6df069286eefc1a45ff Mon Sep 17 00:00:00 2001
From: wabinyai <rajacastro@gmail.com>
Date: Thu, 14 Nov 2024 20:39:54 +0300
Subject: [PATCH 06/17] LOGGING

---
 src/spatial/views/report_view.py | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/src/spatial/views/report_view.py b/src/spatial/views/report_view.py
index b31556c91f..745131104b 100644
--- a/src/spatial/views/report_view.py
+++ b/src/spatial/views/report_view.py
@@ -1,11 +1,15 @@
 from flask import request, jsonify
 from models.report_datafetcher import DataFetcher, AirQualityReport
+import logging
+# Set up a basic logger configuration
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
 
 class ReportView:
     @staticmethod
     def generate_air_quality_report():
         """Fetch air quality data and generate a report based on grid_id, start_time, end_time, and audience."""
-        print('Processing request to generate air quality report...')
+        logger.info('Received request to generate air quality report.')
 
         # Extract data from the request
         data = request.json

From e8b5760dfb6e87db16e72223f343c8e54b9b73c1 Mon Sep 17 00:00:00 2001
From: wabinyai <rajacastro@gmail.com>
Date: Thu, 14 Nov 2024 21:51:32 +0300
Subject: [PATCH 07/17] Report

---
 src/spatial/views/report_view.py | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)

diff --git a/src/spatial/views/report_view.py b/src/spatial/views/report_view.py
index 745131104b..fbd100764e 100644
--- a/src/spatial/views/report_view.py
+++ b/src/spatial/views/report_view.py
@@ -1,15 +1,11 @@
 from flask import request, jsonify
-from models.report_datafetcher import DataFetcher, AirQualityReport
-import logging
-# Set up a basic logger configuration
-logging.basicConfig(level=logging.INFO)
-logger = logging.getLogger(__name__)
+from models.report_datafetcher import DataFetcher, AirQualityReport 
 
 class ReportView:
     @staticmethod
     def generate_air_quality_report():
         """Fetch air quality data and generate a report based on grid_id, start_time, end_time, and audience."""
-        logger.info('Received request to generate air quality report.')
+    #    print('Processing request to generate air quality report...')
 
         # Extract data from the request
         data = request.json

From 7bdb63e801525d55e1b079e5f4d34bd71aab0f06 Mon Sep 17 00:00:00 2001
From: wabinyai <rajacastro@gmail.com>
Date: Thu, 14 Nov 2024 22:13:39 +0300
Subject: [PATCH 08/17] url

---
 src/spatial/configure.py                 | 2 +-
 src/spatial/models/report_datafetcher.py | 4 +++-
 2 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/src/spatial/configure.py b/src/spatial/configure.py
index b09ccc2033..f07eb81cff 100644
--- a/src/spatial/configure.py
+++ b/src/spatial/configure.py
@@ -24,7 +24,7 @@ class Config:
     BIGQUERY_SATELLITE_MODEL_PREDICTIONS = os.getenv(
         "BIGQUERY_SATELLITE_MODEL_PREDICTIONS"
     )
-
+    ANALTICS_URL = os.getenv("ANALTICS_URL")
     HUGGING_FACE_TOKEN = os.getenv("HUGGING_FACE_TOKEN")
     GOOGLE_API_KEY = os.getenv("GOOGLE_API_KEY")
     OPENAI_API_KEY = os.getenv("OPENAI_API_KEY")
diff --git a/src/spatial/models/report_datafetcher.py b/src/spatial/models/report_datafetcher.py
index e501f30dce..f36ed169db 100644
--- a/src/spatial/models/report_datafetcher.py
+++ b/src/spatial/models/report_datafetcher.py
@@ -11,6 +11,7 @@
 genai.configure(api_key=GOOGLE_API_KEY)
 hf_token = Config.HUGGING_FACE_TOKEN
 
+
 if hf_token:
     login(hf_token)
 else:
@@ -20,11 +21,12 @@ class DataFetcher:
     @staticmethod
     def fetch_air_quality_data_a(grid_id, start_time, end_time):
         token = Config.AIRQO_API_TOKEN  
+        analtics_url = Config.ANALTICS_URL
         if token is None:
             print("Error: AIRQO_API_TOKEN environment variable is not set.")
             return None
 
-        url = f"https://platform.airqo.net/api/v2/analytics/grid/report?token={token}"
+        url= f"{analtics_url}?token={token}"
         payload = {"grid_id": grid_id, "start_time": start_time, "end_time": end_time}
 
         try:

From 9d8d3aba7c113301f0cf200870024859a183c5ee Mon Sep 17 00:00:00 2001
From: wabinyai <rajacastro@gmail.com>
Date: Thu, 14 Nov 2024 23:17:37 +0300
Subject: [PATCH 09/17] report apis

---
 src/spatial/controllers/controllers.py |  6 +-
 src/spatial/views/report_view.py       | 96 ++++++++++++++++++--------
 2 files changed, 74 insertions(+), 28 deletions(-)

diff --git a/src/spatial/controllers/controllers.py b/src/spatial/controllers/controllers.py
index 26ebc4b86d..a1f2895c71 100644
--- a/src/spatial/controllers/controllers.py
+++ b/src/spatial/controllers/controllers.py
@@ -70,4 +70,8 @@ def get_satellite_prediction():
 
 @controller_bp.route("/air_quality_report", methods=["POST"])
 def fetch_air_quality():
-    return ReportView.generate_air_quality_report()
\ No newline at end of file
+    return ReportView.generate_air_quality_report_with_gemini()
+
+@controller_bp.route("/air_quality_report_without_llm", methods=["POST"])
+def fetch_air_quality_without_llm():
+    return ReportView.generate_air_quality_report_without_llm()
\ No newline at end of file
diff --git a/src/spatial/views/report_view.py b/src/spatial/views/report_view.py
index fbd100764e..5c9b483b71 100644
--- a/src/spatial/views/report_view.py
+++ b/src/spatial/views/report_view.py
@@ -1,42 +1,84 @@
 from flask import request, jsonify
-from models.report_datafetcher import DataFetcher, AirQualityReport 
+from models.report_datafetcher import DataFetcher, AirQualityReport
 
 class ReportView:
     @staticmethod
-    def generate_air_quality_report():
-        """Fetch air quality data and generate a report based on grid_id, start_time, end_time, and audience."""
-    #    print('Processing request to generate air quality report...')
-
-        # Extract data from the request
+    def _fetch_and_validate_request_data():
+        """Extract and validate request data."""
         data = request.json
         grid_id = data.get("grid_id")
         start_time = data.get("start_time")
         end_time = data.get("end_time")
-        audience = data.get("audience", "general public")  # Default to "general public" if audience is not provided
+        audience = data.get("audience", "general public")
 
         # Validate input parameters
         if not all([grid_id, start_time, end_time, audience]):
-            return jsonify({"error": "Missing required parameters: grid_id, start_time, end_time, audience"}), 400
+            return None, jsonify({
+                "error": "Missing required parameters: grid_id, start_time, end_time, audience"
+            }), 400
 
         # Fetch air quality data
         air_quality_data = DataFetcher.fetch_air_quality_data_a(grid_id, start_time, end_time)
-
         if air_quality_data is None:
-            return jsonify({"error": "No data found for the given parameters"}), 404
-
-        # Create an air quality report
-        report = AirQualityReport(air_quality_data)
-        
-        # Generate the report with the specified audience
-        json_report = report.generate_report_with_gemini(audience)          # using google gemini
-    #    json_report = report.generate_report_without_llm()
-    #    json_report = report.generate_text_with_gpt_neo(audience) # without LLM
-    #    json_report = report.generate_report_with_openai(audience)         # Using openai api 
-    
-        
-        if json_report is None:
-            return jsonify({"error": "Failed to generate report"}), 500
-
-        return jsonify({"report": json_report}), 200
- 
- 
\ No newline at end of file
+            return None, jsonify({
+                "error": "No data found for the given parameters"
+            }), 404
+
+        return data, air_quality_data, None
+    @staticmethod
+    def _handle_error(exception):
+        return jsonify({
+            "error": str(exception)
+        }), 500
+
+    @staticmethod
+    def generate_air_quality_report_with_gemini():
+        """Generate a report using Gemini model."""
+        data, air_quality_data, error_response = ReportView._fetch_and_validate_request_data()
+        if error_response:
+            return error_response
+
+        try:
+            # Create an air quality report
+            report = AirQualityReport(air_quality_data)
+            # Generate the report with Gemini
+            json_report = report.generate_report_with_gemini(data.get("audience", "general public"))
+
+            if json_report is None:
+                return jsonify({
+                    "error": "Failed to generate report with Gemini"
+                }), 500
+
+            return jsonify({
+                "report": json_report,
+                "model": "gemini"
+            }), 200
+
+        except Exception as e:
+            return ReportView._handle_error(e)
+
+    @staticmethod
+    def generate_air_quality_report_without_llm():
+        """Generate a report without using LLM."""
+        data, air_quality_data, error_response = ReportView._fetch_and_validate_request_data()
+        if error_response:
+            return error_response
+
+        try:
+            # Create an air quality report
+            report = AirQualityReport(air_quality_data)
+            # Generate the report without LLM
+            json_report = report.generate_report_without_llm()
+
+            if json_report is None:
+                return jsonify({
+                    "error": "Failed to generate report"
+                }), 500
+
+            return jsonify({
+                "report": json_report,
+                "model": "rule_based"
+            }), 200
+
+        except Exception as e:
+            return ReportView._handle_error(e)

From 054371fa00cb23ee8e268050ebc5c02e2b3b3613 Mon Sep 17 00:00:00 2001
From: wabinyai <rajacastro@gmail.com>
Date: Thu, 14 Nov 2024 23:24:02 +0300
Subject: [PATCH 10/17] logging

---
 src/spatial/views/report_view.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/src/spatial/views/report_view.py b/src/spatial/views/report_view.py
index 5c9b483b71..ca36581cbf 100644
--- a/src/spatial/views/report_view.py
+++ b/src/spatial/views/report_view.py
@@ -28,7 +28,8 @@ def _fetch_and_validate_request_data():
     @staticmethod
     def _handle_error(exception):
         return jsonify({
-            "error": str(exception)
+        #    "error": str(exception)
+            "error": "An internal error has occurred!"
         }), 500
 
     @staticmethod

From 4b7f80f33561a3929abc3eac86665c851696cb36 Mon Sep 17 00:00:00 2001
From: wabinyai <rajacastro@gmail.com>
Date: Thu, 14 Nov 2024 23:55:01 +0300
Subject: [PATCH 11/17] llm

---
 src/spatial/models/report_datafetcher.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/spatial/models/report_datafetcher.py b/src/spatial/models/report_datafetcher.py
index f36ed169db..55ee0874f2 100644
--- a/src/spatial/models/report_datafetcher.py
+++ b/src/spatial/models/report_datafetcher.py
@@ -132,7 +132,7 @@ def generate_report_with_openai(self, audience):
         return self._prepare_report_json(openai_output)
 
     # Use non-LLM template text as report content
-    def generate_report_template_witout_LLM(self, audience):
+    def generate_report_template_without_LLM(self, audience):
         prompt = self._generate_prompt(audience)
         report_content = prompt
         return self._prepare_report_json(report_content)

From 569975d2b43b27155258647ea0e6b3d28dbe0a4e Mon Sep 17 00:00:00 2001
From: wabinyai <rajacastro@gmail.com>
Date: Fri, 15 Nov 2024 00:22:03 +0300
Subject: [PATCH 12/17] report

---
 src/spatial/views/report_view.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/src/spatial/views/report_view.py b/src/spatial/views/report_view.py
index ca36581cbf..06860ec93f 100644
--- a/src/spatial/views/report_view.py
+++ b/src/spatial/views/report_view.py
@@ -83,3 +83,4 @@ def generate_air_quality_report_without_llm():
 
         except Exception as e:
             return ReportView._handle_error(e)
+        

From 66199f1aac4a030f9305415dec2f7a9d64306431 Mon Sep 17 00:00:00 2001
From: wabinyai <rajacastro@gmail.com>
Date: Thu, 21 Nov 2024 13:49:42 +0300
Subject: [PATCH 13/17] ai report

---
 src/spatial/models/report_datafetcher.py | 78 +++++++++++++++---------
 1 file changed, 50 insertions(+), 28 deletions(-)

diff --git a/src/spatial/models/report_datafetcher.py b/src/spatial/models/report_datafetcher.py
index 55ee0874f2..73b502b017 100644
--- a/src/spatial/models/report_datafetcher.py
+++ b/src/spatial/models/report_datafetcher.py
@@ -4,6 +4,7 @@
 from huggingface_hub import login
 from configure import Config
 import google.generativeai as genai 
+import logging
 
 
 # Configure API keys
@@ -21,12 +22,12 @@ class DataFetcher:
     @staticmethod
     def fetch_air_quality_data_a(grid_id, start_time, end_time):
         token = Config.AIRQO_API_TOKEN  
-        analtics_url = Config.ANALTICS_URL
+        analytics_url = Config.ANALTICS_URL
         if token is None:
             print("Error: AIRQO_API_TOKEN environment variable is not set.")
             return None
 
-        url= f"{analtics_url}?token={token}"
+        url= f"{analytics_url}?token={token}"
         payload = {"grid_id": grid_id, "start_time": start_time, "end_time": end_time}
 
         try:
@@ -35,10 +36,13 @@ def fetch_air_quality_data_a(grid_id, start_time, end_time):
             return response.json()
         except requests.exceptions.HTTPError as http_err:
             print(f"HTTP error occurred: {http_err}")
+            logging.error(f"HTTP error occurred: {http_err}")
         except requests.exceptions.RequestException as req_err:
             print(f"Request error occurred: {req_err}")
+            logging.error(f"Request error occurred: {req_err}")
         except ValueError as json_err:
             print(f"JSON decoding error: {json_err}")
+            logging.error(f"JSON decoding error: {json_err}")
 
         return None
 
@@ -64,10 +68,13 @@ def __init__(self, data):
         self.annual_pm10_calibrated_value = self.annual_data.get("pm10_calibrated_value")
 
         # Finding the minimum and maximum values
-        self.daily_min_pm2_5 = min(self.daily_mean_data, key=lambda x: x['pm2_5_calibrated_value'])
-        self.daily_max_pm2_5 = max(self.daily_mean_data, key=lambda x: x['pm2_5_calibrated_value'])
-        
-
+        if self.daily_mean_data:
+            self.daily_min_pm2_5 = min(self.daily_mean_data, key=lambda x: x['pm2_5_calibrated_value'])
+            self.daily_max_pm2_5 = max(self.daily_mean_data, key=lambda x: x['pm2_5_calibrated_value'])
+        else:
+            self.daily_min_pm2_5 = None
+            self.daily_max_pm2_5 = None
+          
 
         # Initialize models once in the constructor
         self.gemini_model = genai.GenerativeModel('gemini-pro')
@@ -86,13 +93,19 @@ def _generate_prompt(self, audience):
         base_info = self._prepare_base_info()
         if audience == "researcher":
             return (   
-                f"Generate a comprehensive air quality assessment report for {self.grid_name} for the period of {self.starttime} to {self.endtime}. Begin with a detailed introduction (100-130 words) covering the city's geographical location, climate characteristics, population density, and major pollution sources. "
-                f"{base_info} include the period under review."
-                f"Daily mean measurements show: {self.daily_mean_data}. "
-                f"Diurnal patterns indicate: {self.diurnal}. Monthly trends reveal: {self.monthly_data}. "
-                f"Provide a thorough analysis of spatial and temporal air quality variations, identify pollution hotspots and clean zones, examine seasonal patterns, and assess compliance with WHO guidelines. "
-                f"Conclude with actionable recommendations for air quality improvement and public health protection. Data source: AirQo monitoring network."
-            )
+-                f"Generate a comprehensive air quality assessment report for {self.grid_name} for the period of {self.starttime} to {self.endtime}. Begin with a detailed introduction (100-130 words) covering the city's geographical location, climate characteristics, population density, and major pollution sources. "  
+-                f"{base_info} include the period under review."  
+-                f"Daily mean measurements show: {self.daily_mean_data}. "  
+-                f"Diurnal patterns indicate: {self.diurnal}. Monthly trends reveal: {self.monthly_data}. "  
++                f"Generate a comprehensive air quality assessment report for {self.grid_name} for the period of {self.starttime} to {self.endtime}. Begin with a detailed introduction (100-130 words) covering the city's geographical location, climate characteristics, population density, and major pollution sources.\n"  
++                f"{base_info}\n"  
++                f"Daily mean measurements show values ranging from {self.daily_min_pm2_5['pm2_5_calibrated_value']} to {self.daily_max_pm2_5['pm2_5_calibrated_value']} µg/m³.\n"  
++                f"Diurnal patterns indicate peak pollution levels at {self._format_diurnal_peak()}.\n"  
++                f"Monthly trends reveal fluctuations correlated with seasonal changes.\n"  
+                 f"Provide a thorough analysis of spatial and temporal air quality variations, identify pollution hotspots and clean zones, examine seasonal patterns, and assess compliance with WHO guidelines. "  
+                 f"Conclude with actionable recommendations for air quality improvement and public health protection. Data source: AirQo monitoring network."  
+             ) 
+
         elif audience == "policymaker":
             return (
                 f"Create an executive summary of air quality conditions in {self.grid_name} for the period of {self.starttime} to {self.endtime}. for policy decision-making. Begin with key findings and their policy implications (50-75 words). "
@@ -106,7 +119,7 @@ def _generate_prompt(self, audience):
             return ( 
                 f"{base_info} include the period under review."
                 f"Create a clear, easy-to-understand report about air quality in {self.grid_name} for the period of {self.starttime} to {self.endtime}. Start with a simple explanation of why air quality matters for public health. "
-                f"We have {self.num_sites} air quality monitors in your area. The average PM2.5 level this year is {self.annual_data} µg/m³. "
+                f"We have {self.num_sites} air quality monitors in your area. The average PM2.5 level this year is {self.annual_pm2_5_calibrated_value} µg/m³. "
                 f"Diurnal patterns indicate: {self.diurnal}. Monthly trends reveal: {self.monthly_data}. "
                 f"Explain what these numbers mean for daily activities. Include: 1) When air quality is best and worst during the day, "
                 f"2) Which areas have better or worse air quality, 3) Simple steps people can take to protect their health, "
@@ -124,12 +137,17 @@ def generate_report_with_gemini(self, audience):
 
     def generate_report_with_openai(self, audience):
         prompt = self._generate_prompt(audience)
-        response = openai.ChatCompletion.create(
-            model="gpt-3.5-turbo",
-            messages=[{"role": "user", "content": prompt}]
-        )
-        openai_output = response.choices[0].message['content']
-        return self._prepare_report_json(openai_output)
+        try:
+            response = openai.ChatCompletion.create(
+                model="gpt-3.5-turbo",
+                messages=[{"role": "user", "content": prompt}]
+            )
+            openai_output = response.choices[0].message['content']
+            return self._prepare_report_json(openai_output)
+        except Exception as e:
+            print(f"Error: {e}")
+            return None
+
 
     # Use non-LLM template text as report content
     def generate_report_template_without_LLM(self, audience):
@@ -139,14 +157,18 @@ def generate_report_template_without_LLM(self, audience):
     
     def generate_report_without_llm(self):
     # Determine peak time and least PM2.5 values
-        peak_data = max(self.diurnal, key=lambda x: x['pm2_5_calibrated_value'])
-        peak_time = peak_data['hour']
-        peak_pm2_5 = peak_data['pm2_5_calibrated_value']
-
-        least_data = min(self.diurnal, key=lambda x: x['pm2_5_calibrated_value'])
-        least_pm2_5 = least_data['pm2_5_calibrated_value']
-        least_pm2_5_time = least_data['hour']
-
+        if self.diurnal:
+            peak_data = max(self.diurnal, key=lambda x: x['pm2_5_calibrated_value'])
+            peak_time = peak_data['hour']
+            peak_pm2_5 = peak_data['pm2_5_calibrated_value']
+            least_data = min(self.diurnal, key=lambda x: x['pm2_5_calibrated_value'])
+            least_pm2_5 = least_data['pm2_5_calibrated_value']
+            least_pm2_5_time = least_data['hour']
+        else:
+            peak_time = None
+            peak_pm2_5 = None
+            least_pm2_5 = None
+        least_pm2_5_time = None
 
         
         introduction = (

From e2b0f70cc0a6b6a16f24e5d52af1cceae76c7407 Mon Sep 17 00:00:00 2001
From: wabinyai <rajacastro@gmail.com>
Date: Thu, 21 Nov 2024 15:18:42 +0300
Subject: [PATCH 14/17] gemini

---
 src/spatial/models/report_datafetcher.py | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/src/spatial/models/report_datafetcher.py b/src/spatial/models/report_datafetcher.py
index 73b502b017..49bd9e054d 100644
--- a/src/spatial/models/report_datafetcher.py
+++ b/src/spatial/models/report_datafetcher.py
@@ -131,9 +131,14 @@ def _generate_prompt(self, audience):
 
     def generate_report_with_gemini(self, audience):
         prompt = self._generate_prompt(audience)
-        response = self.gemini_model.generate_content(prompt)
-        gemini_output = response.text
-        return self._prepare_report_json(gemini_output)
+        try:
+            response = self.gemini_model.generate_content(prompt)
+            gemini_output = response.text
+            return self._prepare_report_json(gemini_output)
+        except Exception as e:
+            print(f"Error: {e}")
+            return None
+
 
     def generate_report_with_openai(self, audience):
         prompt = self._generate_prompt(audience)

From 4178f4a1f5256754fa407c3c3ece85a249e03519 Mon Sep 17 00:00:00 2001
From: wabinyai <rajacastro@gmail.com>
Date: Thu, 21 Nov 2024 15:19:09 +0300
Subject: [PATCH 15/17] Gemini

---
 src/spatial/models/report_datafetcher.py | 1 -
 1 file changed, 1 deletion(-)

diff --git a/src/spatial/models/report_datafetcher.py b/src/spatial/models/report_datafetcher.py
index 49bd9e054d..842ee4b446 100644
--- a/src/spatial/models/report_datafetcher.py
+++ b/src/spatial/models/report_datafetcher.py
@@ -139,7 +139,6 @@ def generate_report_with_gemini(self, audience):
             print(f"Error: {e}")
             return None
 
-
     def generate_report_with_openai(self, audience):
         prompt = self._generate_prompt(audience)
         try:

From c14792f8a0d549b014d830e5837f3e4a6b98c4a3 Mon Sep 17 00:00:00 2001
From: wabinyai <rajacastro@gmail.com>
Date: Tue, 26 Nov 2024 17:53:28 +0300
Subject: [PATCH 16/17] chat

---
 src/spatial/controllers/controllers.py   |  6 ++++-
 src/spatial/models/report_datafetcher.py | 32 +++++++++++++++++++++++-
 src/spatial/views/report_view.py         | 26 +++++++++++++++++++
 3 files changed, 62 insertions(+), 2 deletions(-)

diff --git a/src/spatial/controllers/controllers.py b/src/spatial/controllers/controllers.py
index a1f2895c71..4318936d92 100644
--- a/src/spatial/controllers/controllers.py
+++ b/src/spatial/controllers/controllers.py
@@ -74,4 +74,8 @@ def fetch_air_quality():
 
 @controller_bp.route("/air_quality_report_without_llm", methods=["POST"])
 def fetch_air_quality_without_llm():
-    return ReportView.generate_air_quality_report_without_llm()
\ No newline at end of file
+    return ReportView.generate_air_quality_report_without_llm()
+
+@controller_bp.route("/air_quality_report_with_customised_prompt", methods=["POST"])
+def fetch_air_quality_with_customised_prompt():
+    return ReportView.generate_air_quality_report_with_customised_prompt_gemini()
\ No newline at end of file
diff --git a/src/spatial/models/report_datafetcher.py b/src/spatial/models/report_datafetcher.py
index 842ee4b446..70583433fc 100644
--- a/src/spatial/models/report_datafetcher.py
+++ b/src/spatial/models/report_datafetcher.py
@@ -55,6 +55,8 @@ def __init__(self, data):
         self.diurnal = data.get('airquality', {}).get('diurnal', [])
         self.monthly_data = data.get('airquality', {}).get('site_monthly_mean_pm', [])
         self.monthly_name_data = data.get('airquality', {}).get('pm_by_month_name', [])
+        self.site_annual_mean_pm = data.get('airquality', {}).get('site_annual_mean_pm', [])
+        self.site_mean_pm = data.get('airquality', {}).get('site_mean_pm', [])
         main_site_info = self.monthly_data[0] if self.monthly_data else {}
         self.main_site = main_site_info.get('site_name')
         self.site_names = [item.get('site_name', None) for item in self.data.get('airquality', {}).get('site_annual_mean_pm', [])]
@@ -74,7 +76,7 @@ def __init__(self, data):
         else:
             self.daily_min_pm2_5 = None
             self.daily_max_pm2_5 = None
-          
+        
 
         # Initialize models once in the constructor
         self.gemini_model = genai.GenerativeModel('gemini-pro')
@@ -138,6 +140,27 @@ def generate_report_with_gemini(self, audience):
         except Exception as e:
             print(f"Error: {e}")
             return None
+    # Generate report with customised prompt
+    def generate_report_with_customised_prompt_gemini(self, custom_prompt):
+        """
+        Generate an air quality report using a customised user-provided prompt.
+        """
+        base_info = self._prepare_base_info() 
+        full_prompt = (
+
+            f"{base_info} include the period under review."
+            f"diurnal patterns indicate: {self.diurnal}. "
+            f"number of sites or devices or airqo binos: {self.num_sites}. "
+            f"{self.daily_mean_data}"
+            f"site mean{self.site_mean_pm}"
+            f"{custom_prompt}" 
+        )
+        try:
+            response = self.gemini_model.generate_content(full_prompt)
+            gemini_output = response.text
+            return self._prepare_customised_report_json(gemini_output)
+        except Exception as e:
+            print(f"Error: {e}")
 
     def generate_report_with_openai(self, audience):
         prompt = self._generate_prompt(audience)
@@ -236,4 +259,11 @@ def _prepare_report_json(self, report_content):
             "diurnal": self.diurnal,
             "monthly_data": self.monthly_data,
             "report": report_content
+        }
+    
+    def _prepare_customised_report_json(self, report_content):
+        return {
+            "grid_name": self.grid_name,
+            "start_end_time": self.starttime + " to " + self.endtime,
+            "report": report_content
         }
\ No newline at end of file
diff --git a/src/spatial/views/report_view.py b/src/spatial/views/report_view.py
index 06860ec93f..2a4199cb19 100644
--- a/src/spatial/views/report_view.py
+++ b/src/spatial/views/report_view.py
@@ -10,6 +10,7 @@ def _fetch_and_validate_request_data():
         start_time = data.get("start_time")
         end_time = data.get("end_time")
         audience = data.get("audience", "general public")
+        customise_prompt = data.get("customise_prompt", "")  # Optional customization
 
         # Validate input parameters
         if not all([grid_id, start_time, end_time, audience]):
@@ -84,3 +85,28 @@ def generate_air_quality_report_without_llm():
         except Exception as e:
             return ReportView._handle_error(e)
         
+
+    @staticmethod   
+    def generate_air_quality_report_with_customised_prompt_gemini():
+        data, air_quality_data, error_response = ReportView._fetch_and_validate_request_data()
+        if error_response:
+            return error_response
+
+        try:
+            # Create an air quality report
+            report = AirQualityReport(air_quality_data)
+            # Generate the report with Gemini
+            json_report = report.generate_report_with_customised_prompt_gemini(data.get("customise_prompt", ""))
+
+            if json_report is None:
+                return jsonify({
+                    "error": "Failed to generate report with Gemini"
+                }), 500
+
+            return jsonify({
+                "report": json_report,
+                "model": "gemini"
+            }), 200
+
+        except Exception as e:
+            return ReportView._handle_error(e)

From e6544e049800de41369d6563e06aae574f12b6ca Mon Sep 17 00:00:00 2001
From: wabinyai <rajacastro@gmail.com>
Date: Tue, 26 Nov 2024 23:12:00 +0300
Subject: [PATCH 17/17] daily mean

---
 src/spatial/models/report_datafetcher.py | 16 ++++++++++++++--
 1 file changed, 14 insertions(+), 2 deletions(-)

diff --git a/src/spatial/models/report_datafetcher.py b/src/spatial/models/report_datafetcher.py
index 70583433fc..07e8232270 100644
--- a/src/spatial/models/report_datafetcher.py
+++ b/src/spatial/models/report_datafetcher.py
@@ -5,6 +5,7 @@
 from configure import Config
 import google.generativeai as genai 
 import logging
+from functools import lru_cache
 
 
 # Configure API keys
@@ -20,6 +21,7 @@
 
 class DataFetcher:
     @staticmethod
+    @lru_cache(maxsize=128)  # Cache up to 128 most recent queries
     def fetch_air_quality_data_a(grid_id, start_time, end_time):
         token = Config.AIRQO_API_TOKEN  
         analytics_url = Config.ANALTICS_URL
@@ -71,8 +73,16 @@ def __init__(self, data):
 
         # Finding the minimum and maximum values
         if self.daily_mean_data:
-            self.daily_min_pm2_5 = min(self.daily_mean_data, key=lambda x: x['pm2_5_calibrated_value'])
-            self.daily_max_pm2_5 = max(self.daily_mean_data, key=lambda x: x['pm2_5_calibrated_value'])
+            filtered_data = [
+                    item for item in self.daily_mean_data 
+                    if 'pm2_5_calibrated_value' in item and isinstance(item['pm2_5_calibrated_value'], (int, float))
+                ]
+            if filtered_data:
+                self.daily_min_pm2_5 = min(filtered_data, key=lambda x: x['pm2_5_calibrated_value'])
+                self.daily_max_pm2_5 = max(filtered_data, key=lambda x: x['pm2_5_calibrated_value'])
+            else:
+                self.daily_min_pm2_5 = None
+                self.daily_max_pm2_5 = None
         else:
             self.daily_min_pm2_5 = None
             self.daily_max_pm2_5 = None
@@ -141,6 +151,7 @@ def generate_report_with_gemini(self, audience):
             print(f"Error: {e}")
             return None
     # Generate report with customised prompt
+    @lru_cache(maxsize=64)  # Cache up to 64 most recent reports
     def generate_report_with_customised_prompt_gemini(self, custom_prompt):
         """
         Generate an air quality report using a customised user-provided prompt.
@@ -153,6 +164,7 @@ def generate_report_with_customised_prompt_gemini(self, custom_prompt):
             f"number of sites or devices or airqo binos: {self.num_sites}. "
             f"{self.daily_mean_data}"
             f"site mean{self.site_mean_pm}"
+            f" daily {self.daily_mean_data}"
             f"{custom_prompt}" 
         )
         try: