Fixed naking convention

.github/workflows/main.yaml

@@ -18,33 +18,33 @@ jobs:
       ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
 
     steps:
-    - name: 📦 Checkout code
+    - name: Checkout code
       uses: actions/checkout@v3
 
-    - name: 🐍 Set up Python
+    - name: Set up Python
       uses: actions/setup-python@v4
       with:
         python-version: '3.13'
 
-    - name: 📦 Install dependencies
+    - name: Install dependencies
       run: |
         python -m pip install --upgrade pip
         pip install -r requirements.txt
         pip install pytest
 
-    - name: 🧪 Run tests
+    - name: Run tests
       run: pytest
 
-    - name: 🐳 Set up Docker Buildx
+    - name: Set up Docker Buildx
       uses: docker/setup-buildx-action@v3
 
-    - name: 🔐 Log in to Docker Hub
+    - name: Log in to Docker Hub
       uses: docker/login-action@v3
       with:
         username: ${{ secrets.DOCKER_USERNAME }}
         password: ${{ secrets.DOCKER_PASSWORD }}
 
-    - name: 🛠️ Build and Push Docker image
+    - name: Build and Push Docker image
       uses: docker/build-push-action@v5
       with:
         context: .

api/main.py

@@ -1,9 +1,26 @@
-from fastapi import FastAPI
+from fastapi import FastAPI, Request
+from logger_config import setup_loggers
+import logging
 from .routers import orchestration, context_analysis, ideation , human_idea_refining , brainstorm , generate_final_story , generate_image, show_analytics
 
+setup_loggers()
 app = FastAPI()
 
 
+# Get loggers
+app_logger = logging.getLogger("app_logger")
+error_logger = logging.getLogger("error_logger")
+access_logger = logging.getLogger("access_logger")
+
+
+@app.middleware("http")
+async def log_requests(request: Request, call_next):
+    access_logger.info(f"Request: {request.method} {request.url}")
+    response = await call_next(request)
+    access_logger.info(f"Response status: {response.status_code}")
+    return response
+
+
 @app.get("/")
 async def root():
     return {'response':'Hello'}

logger_config.py

@@ -0,0 +1,40 @@
+# logger_config.py
+import logging
+import os
+
+def setup_loggers():
+    os.makedirs("logs", exist_ok=True)
+
+    # === Format ===
+    formatter = logging.Formatter("%(asctime)s | %(levelname)s | %(name)s | %(message)s")
+
+    # === App Logger ===
+    app_logger = logging.getLogger("app_logger")
+    app_handler = logging.FileHandler("logs/app.log")
+    app_handler.setLevel(logging.INFO)
+    app_handler.setFormatter(formatter)
+    app_logger.setLevel(logging.INFO)
+    app_logger.addHandler(app_handler)
+
+    # === Error Logger ===
+    error_logger = logging.getLogger("error_logger")
+    error_handler = logging.FileHandler("logs/errors.log")
+    error_handler.setLevel(logging.ERROR)
+    error_handler.setFormatter(formatter)
+    error_logger.setLevel(logging.ERROR)
+    error_logger.addHandler(error_handler)
+
+    # === Access Logger === (optional for request logs)
+    access_logger = logging.getLogger("access_logger")
+    access_handler = logging.FileHandler("logs/access.log")
+    access_handler.setLevel(logging.INFO)
+    access_handler.setFormatter(formatter)
+    access_logger.setLevel(logging.INFO)
+    access_logger.addHandler(access_handler)
+
+    # Optional: also log to console
+    console_handler = logging.StreamHandler()
+    console_handler.setFormatter(formatter)
+    app_logger.addHandler(console_handler)
+    error_logger.addHandler(console_handler)
+    access_logger.addHandler(console_handler)

logs/access.log

@@ -0,0 +1,14 @@
+2025-07-30 16:38:13,871 | INFO | access_logger | Request: GET http://127.0.0.1:8000/docs
+2025-07-30 16:38:13,872 | INFO | access_logger | Response status: 200
+2025-07-30 16:38:14,821 | INFO | access_logger | Request: GET http://127.0.0.1:8000/openapi.json
+2025-07-30 16:38:14,833 | INFO | access_logger | Response status: 200
+2025-07-30 16:38:28,560 | INFO | access_logger | Request: GET http://127.0.0.1:8000/
+2025-07-30 16:38:28,561 | INFO | access_logger | Response status: 200
+2025-07-30 16:38:56,487 | INFO | access_logger | Request: POST http://127.0.0.1:8000/api/human-idea-refining
+2025-07-30 16:38:58,518 | INFO | access_logger | Response status: 200
+2025-07-30 16:39:25,313 | INFO | access_logger | Request: POST http://127.0.0.1:8000/api/brainstorm
+2025-07-30 16:39:36,441 | INFO | access_logger | Response status: 200
+2025-07-30 16:40:01,761 | INFO | access_logger | Request: POST http://127.0.0.1:8000/api/brainstorm
+2025-07-30 16:40:01,763 | INFO | access_logger | Response status: 422
+2025-07-30 16:53:21,510 | INFO | access_logger | Request: POST http://127.0.0.1:8000/api/show-analytics
+2025-07-30 16:53:23,607 | INFO | access_logger | Response status: 200

logs/app.log

@@ -0,0 +1 @@
+2025-07-30 16:53:23,605 | INFO | app_logger | Showing Analytics of the influencers after context analysis.

src/genai/brainstroming_agent/utils/tools.py

@@ -5,7 +5,7 @@ from src.genai.utils.models_loader import  embedding_model , llm
 import numpy as np
 import faiss
 import tiktoken
-from src.genai.utils.load_embeddings import  index , df
+from src.genai.utils.load_embeddings import  caption_index , caption_df
 from src.genai.utils.utils import clean_text
 
 def retrieve_tool(video_topic):
@@ -16,8 +16,8 @@ def retrieve_tool(video_topic):
     query_embedding = np.array(embedding_model.embed_query(str(video_topic))).reshape(1, -1).astype('float32')
     faiss.normalize_L2(query_embedding)
 
-    top_k = len(df)
-    distances, indices = index.search(query_embedding, top_k)
+    top_k = len(caption_df)
+    distances, indices = caption_index.search(query_embedding, top_k)
 
     similarity_threshold = 0.35
     selected = [(idx, sim) for idx, sim in zip(indices[0], distances[0]) if sim >= similarity_threshold]
@@ -28,7 +28,7 @@ def retrieve_tool(video_topic):
     # === Format results ===
     outer_list = []
     for rank, (idx, sim) in enumerate(selected, 1):
-        row = df.iloc[idx]
+        row = caption_df.iloc[idx]
         res = {
             'rank': rank,
             'username': row['username'],

src/genai/ideation_agent/utils/tools.py

@@ -7,7 +7,7 @@ import ast
 import faiss
 import tiktoken
 from src.genai.utils.models_loader import embedding_model
-from src.genai.utils.load_embeddings import embeddings , index , df
+from src.genai.utils.load_embeddings import caption_embeddings , caption_index , caption_df
 from src.genai.utils.utils import clean_text
 
 @tool("influencers_data_retrieval_tool", args_schema=QueryFormatter, return_direct=False,description="Retrieve influencer-related data for a given query.")
@@ -19,8 +19,8 @@ def retrieve_tool(business_details):
     query_embedding = np.array(embedding_model.embed_query(str(business_details))).reshape(1, -1).astype('float32')
     faiss.normalize_L2(query_embedding)
 
-    top_k = len(df)
-    distances, indices = index.search(query_embedding, top_k)
+    top_k = len(caption_df)
+    distances, indices = caption_index.search(query_embedding, top_k)
 
     similarity_threshold = 0.35
     selected = [(idx, sim) for idx, sim in zip(indices[0], distances[0]) if sim >= similarity_threshold]
@@ -31,7 +31,7 @@ def retrieve_tool(business_details):
     # === Format results ===
     outer_list = []
     for rank, (idx, sim) in enumerate(selected, 1):
-        row = df.iloc[idx]
+        row = caption_df.iloc[idx]
         res = {
             'rank': rank,
             'username': row['username'],
@@ -52,4 +52,6 @@ def retrieve_tool(business_details):
     encoding = tiktoken.encoding_for_model('gpt-4o-mini')
     tokens = encoding.encode(cleaned_response)
     trimmed_response = tokens[:1000]
-    return encoding.decode(trimmed_response)
+    return encoding.decode(trimmed_response)
+
+

src/genai/orchestration_agent/utils/tools.py

@@ -2,7 +2,7 @@ import faiss
 import ast
 import pandas as pd
 import numpy as np
-from src.genai.utils.load_embeddings import df, embeddings , index
+from src.genai.utils.load_embeddings import caption_df, caption_embeddings , caption_index
 from src.genai.utils.models_loader import embedding_model
 from src.genai.utils.utils import clean_text
 import tiktoken
@@ -17,16 +17,16 @@ def retrieve_data_for_analytics(business_details):
     # === Encode the query and search ===
     query_embedding = np.array(embedding_model.embed_query(str(business_details))).reshape(1, -1).astype('float32')
     top_k = 10
-    distances, indices = index.search(query_embedding, top_k)
+    distances, indices = caption_index.search(query_embedding, top_k)
 
     # === Format results ===
     results = []
     for i, idx in enumerate(indices[0]):
-        likes = df.iloc[idx]['likesCount']
-        comments = df.iloc[idx]['commentCount']
+        likes = caption_df.iloc[idx]['likesCount']
+        comments = caption_df.iloc[idx]['commentCount']
         res = {
-            'url': df.iloc[idx]['videoUrl'],
-            'username': df.iloc[idx]['username'],
+            'url': caption_df.iloc[idx]['videoUrl'],
+            'username': caption_df.iloc[idx]['username'],
             'likesCount': int(likes) if pd.notnull(likes) else None,
         'commentCount': int(comments) if pd.notnull(comments) else None
         }
@@ -38,8 +38,8 @@ def retrieve_data_for_orchestration(query):
     query_embedding = np.array(embedding_model.embed_query(str(query))).reshape(1, -1).astype('float32')
     faiss.normalize_L2(query_embedding)
 
-    top_k = len(df)
-    distances, indices = index.search(query_embedding, top_k)
+    top_k = len(caption_df)
+    distances, indices = caption_index.search(query_embedding, top_k)
 
     similarity_threshold = 0.35
     selected = [(idx, sim) for idx, sim in zip(indices[0], distances[0]) if sim >= similarity_threshold]
@@ -50,7 +50,7 @@ def retrieve_data_for_orchestration(query):
     # === Format results ===
     outer_list = []
     for rank, (idx, sim) in enumerate(selected, 1):
-        row = df.iloc[idx]
+        row = caption_df.iloc[idx]
         res = {
             'rank': rank,
             'username': row['username'],

src/genai/orchestration_agent/utils/utils.py

@@ -8,7 +8,8 @@ from src.genai.utils.models_loader import llm
 from langchain_core.messages import FunctionMessage , AIMessage
 from .tools import  retrieve_data_for_analytics
 import re
-
+import logging
+app_logger = logging.getLogger("app_logger")
 
 def caption_image(image_base64,user_input):
     if len(image_base64)>0:
@@ -42,6 +43,7 @@ def caption_image(image_base64,user_input):
 
 def show_analytics(business_details):
     tool_response = retrieve_data_for_analytics(str(business_details))
+    app_logger.info('Showing Analytics of the influencers after context analysis.')
     return tool_response
 
 def extract_latest_response_block(response):

src/genai/utils/load_embeddings.py

@@ -5,7 +5,7 @@ import pandas as pd
 from datasets import load_dataset
 
 
-def load_index_once():
+def load_caption_index():
     dataset = load_dataset("DvorakInnovationAI/rt-genai-dataset-v1", revision="openai-embeddings")
     df = dataset["train"]
     df= df.to_pandas()
@@ -16,6 +16,18 @@ def load_index_once():
     index.add(embeddings)
     return df, embeddings, index
 
+def load_imdb_ideas_index():
+    dataset = load_dataset("DvorakInnovationAI/rt-genai-imdb-ideas-v1", revision='openai_embeddings')
+    df = dataset['train']
+    df= df.to_pandas()
+    df['embeddings'] = df['embeddings'].apply(lambda x: ast.literal_eval(x) if isinstance(x,str) else x)
+    embeddings = np.vstack(df['embeddings'].values).astype('float32')
+    faiss.normalize_L2(embeddings)
+    index = faiss.IndexFlatIP(embeddings.shape[1])
+    index.add(embeddings)
+    return df , embeddings , index
+
 print('Loading Embeddings...........')
-# Load once on script start
-df, embeddings, index = load_index_once()
+caption_df, caption_embeddings, caption_index = load_caption_index()
+ideas_df , ideas_embeddings , ideas_index = load_imdb_ideas_index()
+print('Embeddings Loaded.................')