Report with AI +prompts

src/spatial/models/report_datafetcher.py

@@ -105,10 +105,7 @@ 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"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"  

src/spatial/requirements.txt

@@ -7,8 +7,10 @@ geojson~=3.1.0
 libpysal~=4.12.1
 requests~=2.32.3
 python-dotenv~=1.0.1
+pandas-gbq
 google-cloud-bigquery
 google-cloud-storage
+google-cloud-bigquery-storage
 google-cloud-bigquery[pandas] 
 gunicorn
 flask-cors~=5.0.0
@@ -21,13 +23,12 @@ shapely~=2.0.6
 scikit-learn~=1.5.2
 gcsfs~=2024.9.0.post1
 joblib~=1.4.2
-lightgbm~=4.1.0
-numpy~=1.25.2
+lightgbm~=4.1.0 
 numpy
 torch
 transformers
 datasets
 sentencepiece
 huggingface_hub
 google-generativeai
-openai
+openai