Add DesiredOutcome class and integrate into analysis workflow

classes/response_desired_outcome.py

@@ -0,0 +1,89 @@
+import streamlit as st
+import requests
+from dotenv import load_dotenv
+import os
+from helper.upload_response import upload_response
+from helper.upload_File import uploadFile
+from pymongo import MongoClient
+import json
+
+class DesiredOutcome:
+    def __init__(self, model_url):
+        self.uploaded_files = []
+        self.file_dict = {}
+        self.model_url = model_url
+        #self.analyst_name = analyst_name
+        #self.data_src = data_src
+        #self.analyst_description = analyst_description
+        self.initialize()
+        self.row1()
+
+    def initialize(self):
+        # FOR ENV
+        load_dotenv()
+
+        # AGENT NAME
+        #st.header(self.analyst_name)
+    
+    def request_model(self, payload_txt, headers):
+        response = requests.post(self.model_url, json=payload_txt, headers=headers)
+        response.raise_for_status()
+        print(response)
+        output = response.json()
+        #st.write(output)
+        text = output["outputs"][0]["outputs"][0]["results"]["text"]["data"]["text"]
+        #text = json.loads(text)
+        #st.write(text)
+        return text
+    
+    def fetch_data(self, data_field):
+        mongodb_uri = os.getenv("MONGODB_URI")
+        myclient = MongoClient(mongodb_uri)
+        mydb = myclient.get_database()
+        mycol = mydb["df_data"]
+        
+        # Sort by timestamp field in descending order
+        x = mycol.find_one(
+            {"data_field": data_field},
+            sort=[("timestamp", -1)]  
+        )
+        
+        x = x["result"]
+        return x
+    
+    def process(self):
+                with st.spinner('Desired Outcome Analyst...', show_time=True):
+                        st.write('')
+                        headers = {"Content-Type": "application/json", "x-api-key": f"{os.getenv('x-api-key')}"}         
+                        try:
+                                payload_txt = {"input_value": self.payload, "output_type": "text", "input_type": "chat"}
+                                payload_txt_model = self.request_model(payload_txt, headers)
+                                debug_info = {'data_field' : 'Desired Outcomes Analyst', 'result': payload_txt_model}
+                                upload_response(debug_info)
+
+                                st.session_state['client_summary'] = ''
+                                count = 0
+                        except Exception as e:
+                            pass
+                        st.session_state['analyzing'] = False    
+                      
+    def row1(self):
+            st.session_state['analyzing'] = False
+            self.payload = ""  
+            count = 0
+            try:
+                session_client_summary = st.session_state['client_summary']
+                if session_client_summary == 'uploaded':
+                    count += 1
+                    self.payload += self.fetch_data("Client Summary")
+            except Exception as e:
+                pass
+            
+            if count >= 1:
+                self.process()
+                                       
+
+if __name__ == "__main__":
+    st.set_page_config(layout="wide")
+
+upload = uploadFile()

pages/analyzing_page.py

@@ -15,6 +15,7 @@ from classes.response_sem_ppc import Sem_PPC
 from classes.response_marketplace import Marketplace
 from classes.response_target_market import TargetMarket
 from classes.response_df_overview import dfOverview
+from classes.response_desired_outcome import DesiredOutcome
 from classes.response_executive_summary import ExecutiveSummary
 from classes.response_snapshot import Snapshot
 
@@ -68,8 +69,11 @@ def run_analysis():
         "marketplace": st.empty(),
         "target_market": st.empty(),
         "df_overview": st.empty(),
+        "desired_outcome": st.empty(),
+        "snapshot": st.empty(),
         "executive_summary": st.empty(),
-        "snapshot": st.empty()
+        
+        
     }
     
     # Create thread-safe handlers for each analysis type
@@ -208,6 +212,17 @@ def run_analysis():
         except Exception as e:
             handler.update_error(f"DF Overview Analysis failed: {str(e)}")
             return None
+    
+    def run_desired_outcomes_analysis():
+        handler = handlers["desired_outcome"]
+        try:
+            handler.update_info("Running Desired Outcomes Analysis...")
+            result = DesiredOutcome(os.getenv('Model_Desired_Outcomes_DM_Analyst'))
+            handler.update_success("Desired Outcomes Analysis completed successfully.")
+            return result
+        except Exception as e:
+            handler.update_error(f"Desired Outcomes Analysis failed: {str(e)}")
+            return None
             
     def run_snapshot_analysis():
         handler = handlers["snapshot"]
@@ -244,7 +259,8 @@ def run_analysis():
         (run_sem_ppc_analysis, "sem_ppc"),
         (run_marketplace_analysis, "marketplace"),
         (run_target_market_analysis, "target_market"),
-        (run_df_overview_analysis, "df_overview")
+        (run_df_overview_analysis, "df_overview"),
+        (run_desired_outcomes_analysis, "desired_outcome")
     ]
     
     # Create and start first batch threads with small delays to prevent UI conflicts