created backup for analytics

api/routers/analytics_chatbot.py

@@ -19,11 +19,11 @@ graph = agent.chatbot_graph()
 def get_analytics(msg:str):
     # user_query=process_query(msg)
     config={"configurable": {"thread_id": "analytics-chatbot-thread"},"run_name":"analytics-chatbot"}
-    try:
-        result=graph.invoke({'messages':msg},config=config)
+    result=graph.invoke({'messages':msg},config=config)
+    if result.get('backup_data') is not None:
+        return {'backup_response': result['backup_data']}
+    else:
         return {
             'response': result['response'],
             'endpoint': result['endpoint']
         }
-    except Exception as e:
-        print(e)

logs/access.log

@@ -1467,3 +1467,45 @@
 2025-10-16 13:51:59,731 | INFO | access_logger | app.py:21 | Response status: 200
 2025-10-16 13:59:15,639 | INFO | access_logger | app.py:19 | Request: GET http://127.0.0.1:8000/api/analytics-chatbot?msg=I%20want%20demographics%20analytics%20of%20munachiya
 2025-10-16 13:59:21,151 | INFO | access_logger | app.py:21 | Response status: 200
+2025-10-16 16:15:38,202 | INFO | access_logger | app.py:19 | Request: GET http://127.0.0.1:8000/api/analytics-chatbot?msg=is%20muna%20having%20strong%20engagement%20than%20divya%3F
+2025-10-16 16:15:44,816 | INFO | access_logger | app.py:21 | Response status: 200
+2025-10-16 16:16:19,370 | INFO | access_logger | app.py:19 | Request: GET http://127.0.0.1:8000/api/analytics-chatbot?msg=who%20is%20better%20for%20my%20business%3F%20muna%20or%20divya%3F
+2025-10-16 16:16:25,888 | INFO | access_logger | app.py:21 | Response status: 200
+2025-10-16 16:32:41,530 | INFO | access_logger | app.py:19 | Request: GET http://127.0.0.1:8000/api/analytics-chatbot?msg=iam%20subash
+2025-10-16 17:01:38,064 | INFO | access_logger | app.py:19 | Request: GET http://127.0.0.1:8000/api/analytics-chatbot?msg=iam%20subash
+2025-10-16 17:01:52,561 | INFO | access_logger | app.py:21 | Response status: 200
+2025-10-17 12:07:55,874 | INFO | access_logger | app.py:19 | Request: GET http://127.0.0.1:8000/
+2025-10-17 12:07:55,875 | INFO | access_logger | app.py:21 | Response status: 200
+2025-10-17 12:07:56,420 | INFO | access_logger | app.py:19 | Request: GET http://127.0.0.1:8000/favicon.ico
+2025-10-17 12:07:56,421 | INFO | access_logger | app.py:21 | Response status: 404
+2025-10-17 12:07:58,630 | INFO | access_logger | app.py:19 | Request: GET http://127.0.0.1:8000/docs
+2025-10-17 12:07:58,631 | INFO | access_logger | app.py:21 | Response status: 200
+2025-10-17 12:07:58,788 | INFO | access_logger | app.py:19 | Request: GET http://127.0.0.1:8000/openapi.json
+2025-10-17 12:07:58,795 | INFO | access_logger | app.py:21 | Response status: 200
+2025-10-17 12:08:09,324 | INFO | access_logger | app.py:19 | Request: GET http://127.0.0.1:8000/api/analytics-chatbot?msg=I%20am%20subash
+2025-10-17 12:12:39,346 | INFO | access_logger | app.py:19 | Request: GET http://127.0.0.1:8000/api/analytics-chatbot?msg=I%20am%20subash
+2025-10-17 12:14:10,002 | INFO | access_logger | app.py:19 | Request: GET http://127.0.0.1:8000/api/analytics-chatbot?msg=I%20am%20subash
+2025-10-17 12:14:13,701 | INFO | access_logger | app.py:21 | Response status: 200
+2025-10-17 12:15:09,470 | INFO | access_logger | app.py:19 | Request: GET http://127.0.0.1:8000/api/analytics-chatbot?msg=how%20are%20you%3F
+2025-10-17 12:15:14,264 | INFO | access_logger | app.py:21 | Response status: 200
+2025-10-17 12:16:44,106 | INFO | access_logger | app.py:19 | Request: GET http://127.0.0.1:8000/api/analytics-chatbot?msg=I%20want%20analytics%20of%20muna%20chiya%20of%20last%2010%20days
+2025-10-17 12:16:48,707 | INFO | access_logger | app.py:21 | Response status: 200
+2025-10-17 12:18:43,858 | INFO | access_logger | app.py:19 | Request: GET http://127.0.0.1:8000/api/analytics-chatbot?msg=I%20want%20analytics%20of%20muna%20chiya%20of%20last%2010%20days
+2025-10-17 12:18:48,714 | INFO | access_logger | app.py:21 | Response status: 200
+2025-10-17 12:27:38,479 | INFO | access_logger | app.py:19 | Request: GET http://127.0.0.1:8000/api/show-analytics
+2025-10-17 12:27:39,454 | INFO | access_logger | app.py:21 | Response status: 200
+2025-10-17 13:13:22,253 | INFO | access_logger | app.py:19 | Request: GET http://127.0.0.1:8000/api/analytics-chatbot?msg=I%20want%20analytics%20of%20muna%20chiya%20of%20last%2010%20days
+2025-10-17 13:18:41,884 | INFO | access_logger | app.py:19 | Request: GET http://127.0.0.1:8000/api/analytics-chatbot?msg=I%20want%20analytics%20of%20muna%20chiya%20of%20last%2010%20days
+2025-10-17 13:18:51,310 | INFO | access_logger | app.py:21 | Response status: 200
+2025-10-17 13:42:23,977 | INFO | access_logger | app.py:19 | Request: GET http://127.0.0.1:8000/api/analytics-chatbot?msg=I%20want%20analytics%20of%20muna%20chiya%20of%20last%2010%20days
+2025-10-17 13:42:33,362 | INFO | access_logger | app.py:21 | Response status: 200
+2025-10-17 13:46:31,111 | INFO | access_logger | app.py:19 | Request: GET http://127.0.0.1:8000/api/analytics-chatbot?msg=I%20want%20analytics%20of%20muna%20chiya%20of%20last%2010%20days
+2025-10-17 13:46:38,070 | INFO | access_logger | app.py:21 | Response status: 200
+2025-10-17 13:53:58,981 | INFO | access_logger | app.py:19 | Request: GET http://127.0.0.1:8000/
+2025-10-17 13:53:58,982 | INFO | access_logger | app.py:21 | Response status: 200
+2025-10-17 13:54:02,244 | INFO | access_logger | app.py:19 | Request: GET http://127.0.0.1:8000/docs
+2025-10-17 13:54:02,245 | INFO | access_logger | app.py:21 | Response status: 200
+2025-10-17 13:54:02,375 | INFO | access_logger | app.py:19 | Request: GET http://127.0.0.1:8000/openapi.json
+2025-10-17 13:54:02,382 | INFO | access_logger | app.py:21 | Response status: 200
+2025-10-17 13:54:26,472 | INFO | access_logger | app.py:19 | Request: GET http://127.0.0.1:8000/api/analytics-chatbot?msg=i%20want%20the%20analytics%20of%20last%2010%20days%20of%20muna%20chiya
+2025-10-17 13:54:36,464 | INFO | access_logger | app.py:21 | Response status: 200

logs/app.log

@@ -132,3 +132,4 @@
 2025-10-14 13:36:55,021 | INFO | app_logger | api/routers/ideation.py:33 | Executed the ideation pipeline.
 2025-10-14 13:55:41,999 | INFO | app_logger | api/routers/context_analysis.py:27 | Context Analysis Completed.
 2025-10-14 14:03:00,252 | INFO | app_logger | api/routers/brainstorm.py:42 | Executed brainstorming agent.
+2025-10-17 12:27:39,451 | INFO | app_logger | api/routers/show_analytics.py:14 | Influencer Analytics returned by orchestrator.

src/genai/analytics_chatbot/agent.py

@@ -1,7 +1,7 @@
 from langgraph.graph import StateGraph, START, END
 from langgraph.checkpoint.memory import MemorySaver
 from .utils.state import State
-from .utils.nodes import ChatbotNode, FetchDataNode , FetchLastMessage , RetrievePossibleEndpoints , FetchParametersNode , RetrieveExactEndpoint
+from .utils.nodes import FetchDataNode , FetchLastMessage , RetrievePossibleEndpoints , FetchParametersNode , RetrieveExactEndpoint, BackupRetrievalNode, RoutingNode
 
 class ChatbotAgent:
     def __init__(self):
@@ -14,6 +14,7 @@ class ChatbotAgent:
         graph_builder.add_node("retrieve_exact_endpoint", RetrieveExactEndpoint().run)
         graph_builder.add_node("fetch_parameters", FetchParametersNode().run)
         graph_builder.add_node("fetch_data", FetchDataNode().run)
+        graph_builder.add_node("backup_response", BackupRetrievalNode().run)
         
         graph_builder.add_edge(START, "fetch_last_message")
         graph_builder.add_edge("fetch_last_message", 'retrieve_api_endpoints')
@@ -21,6 +22,9 @@ class ChatbotAgent:
         graph_builder.add_edge("retrieve_exact_endpoint", 'fetch_parameters')
         graph_builder.add_edge("fetch_parameters", 'fetch_data')
         graph_builder.add_edge("fetch_data", END)
+        graph_builder.add_edge("backup_response", END)
 
+        graph_builder.add_conditional_edges("fetch_parameters", RoutingNode().run,{'execute_backup':'backup_response', 'go_on':"fetch_data"})
+        graph_builder.add_conditional_edges("fetch_data", RoutingNode().run,{'execute_backup':'backup_response', 'go_on':END})
 
         return graph_builder.compile(checkpointer=self.memory)

src/genai/analytics_chatbot/utils/nodes.py

@@ -1,14 +1,16 @@
 import requests
 from langchain_core.messages import SystemMessage , HumanMessage , FunctionMessage
 from .state import State
+from .tools import RetrieverBackup
 from .schemas import ResponseFormatter , CompareBodyFormatter, LatestMessageFormatter, ParameterFormatter, EndpointFormatter
-from .prompts import chatbot_prompt , get_body_prompt , fetch_last_message_prompt , fetch_parameters_prompt, fetch_endpoint_prompt
+from .prompts import chatbot_prompt , get_body_prompt , fetch_last_message_prompt , fetch_parameters_prompt, fetch_endpoint_prompt, backup_retrieval_prompt
 from .utils import generate_api_knowledge , process_query, get_endpoint_info
 from src.genai.utils.models_loader import llm_gpt
 import numpy as np
-from src.genai.utils.data_loader import api_knowledge_df, api_index
+from src.genai.utils.data_loader import api_knowledge_df, api_index, caption_df , caption_index
 from src.genai.utils.models_loader import embedding_model
 
+
 class FetchLastMessage:
     def __init__(self):
         self.llm = llm_gpt
@@ -34,7 +36,7 @@ class RetrievePossibleEndpoints:
 
     def run(self,state:State):
         query_embedding = np.array(embedding_model.embed_query(state['latest_message'])).reshape(1, -1).astype('float32')
-        distances, indices = self.index.search(query_embedding, 5)
+        distances, indices = self.index.search(query_embedding, 3)
         for idx in indices[0]:
             row = self.df.iloc[idx]
             print('Endpoint:',row['endpoint'])
@@ -71,52 +73,33 @@ class FetchParametersNode:
         self.llm = llm_gpt
     
     def run(self , state:State):
-        print('Entered to fetch parameters')
-        print(state['method'])
-        if state['method'] == 'GET':
-            print('Condition satisfied')
-            template = fetch_parameters_prompt
-            messages=[SystemMessage(content=template),
-                    HumanMessage(content=f'''The query is: {state['latest_message']}\n. The needed parameters: {str(state['needed_parameters'])}''')
-                    ]
-            print('messages:', messages)
-            result = self.llm.with_structured_output(ParameterFormatter, method='function_calling').invoke(messages)
-            parameters_values={key: process_query(value) for key, value in result.parameters_values.items()}
-            print('The parameter values:', parameters_values)
-            return {
-                'parameters_values':parameters_values
-            }
-        else:
-            return{
-                'parameters_values': {}
-            }
-        
+        try:
+            print('Entered to fetch parameters')
+            print(state['method'])
+            if state['method'] == 'GET':
+                print('Condition satisfied')
+                template = fetch_parameters_prompt
+                messages=[SystemMessage(content=template),
+                        HumanMessage(content=f'''The query is: {state['latest_message']}\n. The needed parameters: {str(state['needed_parameters'])}''')
+                        ]
+                print('messages:', messages)
+                result = self.llm.with_structured_output(ParameterFormatter, method='function_calling').invoke(messages)
+                # parameters_values={key: process_query(value) for key, value in result.parameters_values.items()}
+                parameters_values = {k: (process_query(v) if isinstance(v, str) else v) for k, v in result.parameters_values.items()}
 
+                print('The parameter values:', parameters_values)
+                return {
+                    'parameters_values':parameters_values
+                }
+            else:
+                return{
+                    'parameters_values': {}
+                }
+        except Exception as e:
+            print('Error occoured:', e)
+            return {'error_message': str(e)}
 
-class ChatbotNode:
-    def __init__(self):
-        self.llm = llm_gpt
     
-    def run(self, state:State):
-        print('Message:',state['messages'])
-        template = chatbot_prompt()
-        knowledge_base = generate_api_knowledge('https://reveltrends.vercel.app')
-        print('The knowledge base is:', knowledge_base)
-        messages = [SystemMessage(content=template), 
-                    FunctionMessage(name='analytics_chatbot',content=str(knowledge_base)),
-                    ] + state["messages"]
-        if len(state['messages'])>11:
-            state["messages"] = state["messages"][-9:]
-        print('Messages:', state['messages'])
-        print(len(state['messages']))
-        result = self.llm.with_structured_output(ResponseFormatter, method='function_calling').invoke(messages)
-        print('The result is:',result)
-        return {
-        "messages": [{"role": "assistant", "content": f'''The endpoint is: {result.endpoint}. The parameters are: {result.parameters}'''}],
-        "endpoint": result.endpoint,
-        "method": result.method,
-        "parameters": result.parameters,
-    }
 
 class FetchDataNode:
     def __init__(self):
@@ -128,30 +111,54 @@ class FetchDataNode:
             }
 
     def run(self, state:State):
-        print('Entered to fetch data')
-        url = f'''{self.base_url}{state['endpoint']}'''
-        if state['method'] == 'GET':
-            response = requests.get(url, params=state['parameters_values'],headers=self.headers)
-        elif state['endpoint'] == '/api/v1/compare/':
-            print('Condition satisfied')
-            messages = [SystemMessage(content=get_body_prompt()), 
-            HumanMessage(content=str(state['messages']))]
-            response=llm_gpt.with_structured_output(CompareBodyFormatter , method='function_calling').invoke(messages)
-            print('INF names response:', response)
-            payload = {
-            "usernames": list(map(process_query,response.names)),
-            "freq": response.frequency          
-             }
-            
-            print('The payload is:',payload)
-            
-            headers = {
-            "Content-Type": "application/json"
-             }
+        try:
+            print('Entered to fetch data')
+            url = f'''{self.base_url}{state['endpoint']}'''
+            if state['method'] == 'GET':
+                response = requests.get(url, params=state['parameters_values'],headers=self.headers)
+            elif state['endpoint'] == '/api/v1/compare/':
+                print('Condition satisfied')
+                messages = [SystemMessage(content=get_body_prompt()), 
+                HumanMessage(content=str(state['messages']))]
+                response=llm_gpt.with_structured_output(CompareBodyFormatter , method='function_calling').invoke(messages)
+                print('INF names response:', response)
+                payload = {
+                "usernames": list(map(process_query,response.names)),
+                "freq": response.frequency          
+                }
+                
+                print('The payload is:',payload)
+                
+                headers = {
+                "Content-Type": "application/json"
+                }
+
+                response = requests.post(url, json=payload, headers=headers)
+
+            return {'response':response.json()}
+        except Exception as e:
+            print('Error occoured:', e)
+            return {'error_message': str(e)}
 
-            response = requests.post(url, json=payload, headers=headers)
 
-        return {'response':response.json()}
+class BackupRetrievalNode:
+    def __init__(self):
+        self.llm = llm_gpt
+    
+    def run(self, state:State):
+        retrieval=RetrieverBackup().retrieve(state['latest_message'])
+        return {'backup_data': retrieval}
+
+class RoutingNode:
+    def __init__(self):
+        pass
+    
+    def run(self,state:State):
+        if state.get('error_message') is not None:
+            return 'execute_backup'
+        else:
+            return 'go_on'
 
 
+                
 

src/genai/analytics_chatbot/utils/prompts.py

@@ -103,4 +103,9 @@ User Query: I want weekly engagement stats of John
 Possible Endpoints: ['/api/v1/overview/buzz_trend', '/api/v1/analytics/engagement', '/api/v1/analytics/followers']  
 endpoint: /api/v1/analytics/engagement
 
+'''
+
+backup_retrieval_prompt = '''
+You are provided with the retrieved data as a function message and the user query.
+Respond to the user query only through the context of retrieved data. Don't give hallucinated responses.
 '''

src/genai/analytics_chatbot/utils/state.py

@@ -11,4 +11,5 @@ class State(TypedDict):
     response:dict
     error_message:str
     latest_message:str
-    parameters_values:str
+    parameters_values:str
+    backup_data:str

src/genai/analytics_chatbot/utils/tools.py

@@ -1,7 +1,11 @@
-
+import re
 import numpy as np
-from src.genai.utils.data_loader import api_knowledge_df, api_index
+import faiss
+from src.genai.utils.models_loader import embedding_model , encoding_model
+from src.genai.utils.utils import clean_text
+from src.genai.utils.data_loader import api_knowledge_df, api_index, caption_df , caption_index
 from src.genai.utils.models_loader import embedding_model
+import pandas as pd
 
 class APIKnowledgeRetrieveTool:
     def __init__(self):
@@ -17,6 +21,89 @@ class APIKnowledgeRetrieveTool:
                 'parameters':row['parameters']}
         return data
     
+class RetrieverBackup:
+    def __init__(self):
+        self.df = caption_df
+        self.index = caption_index
+
+    def _filter_dataset(self, query):
+        usernames = self.df["username"].dropna().unique()
+        matched_users = [u for u in usernames if re.search(rf"\b{re.escape(u)}\b", query)]
+        if matched_users:
+            filtered_df = self.df[self.df["username"].isin(matched_users)]
+            return filtered_df
+        else:
+            return self.df
+
+
+    def retrieve_old(self, query):
+        query_embedding = np.array(embedding_model.embed_query(str(query))).reshape(1, -1).astype('float32')
+        print('Embeddings Generated')
+        faiss.normalize_L2(query_embedding)
+        print('Query embedded')
+        filtered_df = self._filter_dataset(query)
+        distances, indices = self.index.search(query_embedding, len(filtered_df))
+        similarity_threshold = 0.35
+        selected = [(idx, sim) for idx, sim in zip(indices[0], distances[0]) if sim >= similarity_threshold]
+        if not selected:
+            return "No influencers found."
+
+        outer_list = []
+        for rank, (idx, sim) in enumerate(selected, 1):
+            row = filtered_df.iloc[idx]
+            inner_list = [
+                f"[{rank}]. The influencer name is: **{row['username']}** — Likes: **{row['likesCount']}**, Comments: **{row['commentCount']}**",
+                f"The branding or promotion done is:\n{row['visible_texts_or_brandings']}",
+                f"The details of product or service is:\n{row['product_or_service_details']}"
+            ]
+            outer_list.append(inner_list)
+
+        cleaned_response = clean_text(str(outer_list))
+        print('response cleaned')
+        tokens = encoding_model.encode(cleaned_response)[:500]
+        print('tokens got')
+        return encoding_model.decode(tokens)
+    
+    def retrieve(self, query):
+        query_embedding = np.array(embedding_model.embed_query(str(query))).reshape(1, -1).astype('float32')
+        faiss.normalize_L2(query_embedding)
+
+        # Search on full dataset (index is built on full df)
+        distances, indices = self.index.search(query_embedding, len(self.df))
+        similarity_threshold = 0.35
+
+        # Prepare matched usernames
+        usernames = self.df["username"].dropna().unique()
+        matched_users = [u for u in usernames if re.search(rf"\b{re.escape(u)}\b", query)]
+
+        results = []
+        rank = 1
+        for idx, sim in zip(indices[0], distances[0]):
+            if sim < similarity_threshold:
+                continue
+            
+            row = self.df.iloc[idx]
+
+            # If query mentions usernames, only keep those rows
+            if matched_users and row["username"] not in matched_users:
+                continue
+
+            results.append({
+                'url': row['videoUrl'],
+                'username': row['username'],
+                'likesCount': int(row['likesCount']) if pd.notnull(row['likesCount']) else None,
+                'commentCount': int(row['commentCount']) if pd.notnull(row['commentCount']) else None
+            })
+        results = results[:10] if len(results) > 10 else results
+        return results
+
+        
+
+
+
+    
+
+    
     
 
 

src/genai/analytics_chatbot/utils/utils.py

@@ -80,26 +80,6 @@ def generate_api_knowledge(base_url: str):
 
     return api_knowledge
 
-def influencers_name():
-    return {
-  "information_of_influencers_usernames": "From the names_dictionary, the key represents their original names and the value represents their usernames (influencer_username).",
-  "names_dictionary": {
-    "divya dhakal": "divyadhakal_",
-    "istu karki": "ishtukarkee",
-    "kano mama": "kano_mama",
-    "muna chiya": "munachiya",
-    "nepal food": "nepal_food",
-    "ggkaam": "ggkaam610",
-    "adarsh": "adars_fpv",
-    "ajay tamang": "ajaytm43",
-    "anisha kafle": "anishakafle",
-    "diwash gurung": "diwasg",
-    "eat grub food": "eatgrubfood",
-    "surakshya kc": "imsurakshyakc",
-    "jholey": "jholeyism",
-    "mrb vlogs": "mrbvlog2"
-  }
-}
 
 def process_query(user_query: str) -> str:
     # load mapping from json file

src/genai/ideation_agent/utils/prompts.py

@@ -47,6 +47,20 @@ idea_2:
     One-line_description: show how one bag works in 3 daily scenarios (work, casual, night out).
     Hook: “1 bag → 3 lifestyles.”
 
+The business details is:
+Business Type: Fitness and Gym.
+Platforms: Instagram reels, TikTok
+Target Audience: age 18–50
+Content Shooting Device: Mobile phone
+
+The ideas for it are:
+idea 1: 
+    title: Hydration Hacks for Gym Lovers
+    One_liner_description: A short informative video where a trainer shares quick, practical hydration tips.
+    hook: Feeling tired mid-workout? You might just be thirsty.
+    Unique_selling_proposition: Simple, science-backed advice from trainers that helps members train smarter, not harder.
+    niche: People looking for health, wellness, and workout education.
+
 Generate creative and attention-grabbing video ideas that stand out. Each idea must include surprising elements, be far from generic, and be memorable for its originality and catchiness.
 """