Dev

.gitignore

@@ -6,4 +6,5 @@ test_venv/
 .vscode
 hit.py
 try.ipynb
-db/
+db/
+data/

Dockerfile

@@ -1,25 +1,19 @@
-FROM python:3.9-slim
-
+FROM python:3.12.7-slim-bullseye
 
 WORKDIR /app
 
-
 COPY requirements.txt .
 
-RUN apt-get update && apt-get install -y libmagic-dev
+RUN apt-get update && apt-get install -y \
+    build-essential \
+    libmagic-dev \
+    gcc \
+    g++ \
+    && rm -rf /var/lib/apt/lists/*
 
 RUN pip install --upgrade pip
-
 RUN pip install --no-cache-dir -r requirements.txt
 
-
 COPY . .
 
-
-EXPOSE 8000
-
-
-ENV PORT=8000
-
-# Command to run the application
-# CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "8000"]
+CMD ["uvicorn", "api:app", "--host", "0.0.0.0", "--port", "10000"]

README.md

@@ -292,6 +292,17 @@ percentage, missing keywords, and profile summary.
   - **Starter Code Generation:**
     - The starter code generator agent produces starter Python code for the project, including data loading, model definition, and a basic training loop based on the findings from the problem definition, data assessment, and model recommendation.
 
+### 30. Agrilens
+
+- **Route:** `/agrilens`  
+- **Description:** This API endpoint allows users to upload agricultural images and receive AI-powered insights into crops, pests, diseases, and overall health assessment.
+The analysis also includes actionable recommendations for improving agricultural practices and yields.
+- **Features:**
+  - **Image Upload:** Users can upload agricultural images in JPG, JPEG, or PNG formats.  
+  - **Advanced Detection:** Identifies crops, pests, diseases, and farming equipment within the uploaded image.  
+  - **Health Assessment:** Evaluates crop health, identifying stress factors or signs of disease.  
+  - **Smart Recommendations:** Provides actionable insights for enhancing agricultural productivity, such as fertilizer, pesticide, or irrigation advice.  
+
 ## Usage
 
 Each endpoint accepts specific parameters as described in the respective endpoint documentation. Users can make POST requests to these endpoints with the required parameters to perform the desired tasks.

agents/agent_doc/agents.py

@@ -1,13 +1,11 @@
 from crewai import Agent
 from .tools import scrape_tool, search_tool
-from langchain_google_genai import ChatGoogleGenerativeAI
 import os
 import settings
 
-llm=ChatGoogleGenerativeAI(model=settings.GEMINI_FLASH,
-                           verbose=True,
-                           temperature=0.7,
-                           google_api_key=os.getenv("GOOGLE_API_KEY"))
+os.environ["GEMINI_API_KEY"] = os.environ.get('GEMINI_API_KEY')
+
+llm = "gemini/gemini-1.5-flash-8b"
 
 
 diagnostician = Agent(

agents/agent_doc/crew.py

@@ -5,13 +5,12 @@
 
 doc_crew = Crew(
     agents=[diagnostician, treatment_advisor],
-    tasks=[diagnose_task, treatment_task],
-    verbose=2
+    tasks=[diagnose_task, treatment_task]
 )
 
 def run_doc_crew(input_data):
     result = doc_crew.kickoff(inputs=input_data)
-    return result
+    return str(result)
 
 if __name__=='__main__':
     doc_agent_input ={

agents/investment_risk_analyst_agent/agents.py

@@ -2,14 +2,12 @@
 from .tools import search_tool,scrape_tool
 from dotenv import load_dotenv
 load_dotenv()
-from langchain_google_genai import ChatGoogleGenerativeAI
 import os
 import settings
 
-llm=ChatGoogleGenerativeAI(model=settings.GEMINI_FLASH,
-                           verbose=True,
-                           temperature=0.5,
-                           google_api_key=os.getenv("GOOGLE_API_KEY"))
+os.environ["GEMINI_API_KEY"] = os.environ.get('GEMINI_API_KEY')
+
+llm = "gemini/gemini-1.5-flash-8b"
 
 data_analyst_agent = Agent(
     role="Data Analyst",

agents/investment_risk_analyst_agent/crew.py

@@ -7,10 +7,9 @@
 from .agents import data_analyst_agent,trading_strategy_agent,execution_agent,risk_management_agent
 from .tasks import data_analysis_task,strategy_development_task,risk_assessment_task,execution_planning_task
 
-llm=ChatGoogleGenerativeAI(model="gemini-1.5-flash-latest",
-                           verbose=True,
-                           temperature=0.7,
-                           google_api_key=os.getenv("GOOGLE_API_KEY"))
+os.environ["GEMINI_API_KEY"] = os.environ.get('GEMINI_API_KEY')
+
+llm = "gemini/gemini-1.5-flash-8b"
 
 
 financial_trading_crew = Crew(
@@ -31,7 +30,7 @@
 
 def run_investment_crew(input_data):
     result = financial_trading_crew.kickoff(inputs=input_data)
-    return result
+    return str(result)
 
 if __name__=='__main__':
     financial_trading_inputs ={

agents/job_posting_agent/agents.py

@@ -2,12 +2,10 @@
 from .tools import web_search_tool, serper_search_tool
 import os
 import settings
-from langchain_google_genai import ChatGoogleGenerativeAI
 
-llm=ChatGoogleGenerativeAI(model=settings.GEMINI_FLASH,
-                           verbose=True,
-                           temperature=0.7,
-                           google_api_key=os.getenv("GOOGLE_API_KEY"))
+os.environ["GEMINI_API_KEY"] = os.environ.get('GEMINI_API_KEY')
+
+llm = "gemini/gemini-1.5-flash-8b"
 
 
 class JobAgents():

agents/job_posting_agent/crew.py

@@ -32,7 +32,7 @@
 
 def run_job_crew(input_data):
     result = job_crew.kickoff(input_data)
-    return result
+    return str(result)
 
 if __name__=='__main__':
     job_agent_input = {

agents/ml_assistant/agents.py

@@ -6,9 +6,8 @@
 
 class MLAgents():
     def __init__(self,model) -> None:
-        self.llm = ChatGroq(
-            api_key=os.environ['GROQ_API_KEY'],
-            model_name=model)
+        os.environ["GROQ_API_KEY"] = os.environ.get('GROQ_API_KEY')
+        self.llm = f"groq/{model}"
 
     def problem_definition_agent(self):
         return Agent(

agents/ml_assistant/crew.py

@@ -28,7 +28,7 @@ def run_ml_crew(file_path, user_question, model="llama3-70b-8192"):
     # Format the input data for agents
     input_data = {
         "ml_problem": user_question,
-        "df": df.head(),
+        "df": df.head().to_json(orient="records"),
         "file_name": file_path
     }
 
@@ -40,7 +40,7 @@ def run_ml_crew(file_path, user_question, model="llama3-70b-8192"):
     )
 
     result = ml_crew.kickoff(input_data)
-    return result
+    return str(result)
 
 if __name__=="__main__":
     print(run_ml_crew(file_path="data/iris.csv",

agents/tech_news_agent/agents.py

@@ -2,15 +2,12 @@
 from .tools import tool
 from dotenv import load_dotenv
 load_dotenv()
-from langchain_google_genai import ChatGoogleGenerativeAI
 import os
 
 
-## call the gemini models
-llm=ChatGoogleGenerativeAI(model="gemini-1.5-flash",
-                           verbose=True,
-                           temperature=0.5,
-                           google_api_key=os.getenv("GOOGLE_API_KEY"))
+os.environ["GEMINI_API_KEY"] = os.environ.get('GEMINI_API_KEY')
+
+llm = "gemini/gemini-1.5-flash-8b"
 
 # Creating a senior researcher agent with memory and verbose mode
 

agents/tech_news_agent/crew.py

@@ -11,7 +11,7 @@
 
 def run_crew(topic):
     result = crew.kickoff(inputs={'topic': topic})
-    return result
+    return str(result)
 
 if __name__=='__main__':
     print(run_crew(topic="AI in Constructions"))

api.py

@@ -15,7 +15,8 @@
 from mongo import MongoDB
 from helper_functions import get_qa_chain,get_gemini_response,get_url_doc_qa,extract_transcript_details,\
     get_gemini_response_health,get_gemini_pdf,read_sql_query,remove_substrings,questions_generator,groq_pdf,\
-    summarize_audio,chatbot_send_message,extraxt_pdf_text,advance_rag_llama_index,parse_sql_response, extract_video_id
+    summarize_audio,chatbot_send_message,extraxt_pdf_text,advance_rag_llama_index,parse_sql_response, extract_video_id,\
+    encode_image
 from langchain_groq import ChatGroq
 from langchain.chains.conversation.base import ConversationChain
 from langchain.chains.conversation.memory import ConversationBufferWindowMemory
@@ -47,12 +48,21 @@
 from slowapi.util import get_remote_address
 from slowapi.errors import RateLimitExceeded
 from groq import Groq
+import openai
+from dotenv import load_dotenv
+
+load_dotenv()
 
 os.environ["LANGCHAIN_TRACING_V2"]="true"
 os.environ["LANGCHAIN_API_KEY"]=os.getenv("LANGCHAIN_API_KEY")
 os.environ["LANGCHAIN_PROJECT"]="genify"
 os.environ["LANGCHAIN_ENDPOINT"]="https://api.smith.langchain.com"
 
+openai_compatible_client = openai.OpenAI(
+    base_url="https://api.groq.com/openai/v1",
+    api_key=os.getenv("GROQ_API_KEY")
+)
+
 redis = Redis(host=os.getenv("REDIS_HOST"), port=settings.REDIS_PORT, password=os.getenv("REDIS_PASSWORD"))
 client = Groq()
 mongo_client = MongoDB(collection_name=os.getenv("MONGO_COLLECTION_USER"))
@@ -314,7 +324,7 @@ async def blogs(topic: str = Form("Generative AI")):
             print("Retrieving response from Redis cache")
             return ResponseText(response=cached_response.decode("utf-8"))
 
-        model = genai.GenerativeModel(settings.GEMINI_FLASH)
+        model = genai.GenerativeModel(settings.GEMINI_FLASH_8B)
         blog_prompt = f""" You are expert in blog writing. Write a blog on the topic {topic}. Use a friendly and informative tone, and include examples and tips to encourage readers to get started with the topic provided. """
         response = model.generate_content(blog_prompt)
         redis.set(cache_key, response.text, ex=60)
@@ -523,7 +533,7 @@ async def ats(resume_pdf: UploadFile = File(...), job_description: str = Form(..
             return ResponseText(response=cached_response.decode("utf-8"))
 
         text = extraxt_pdf_text(resume_pdf.file)
-        model = genai.GenerativeModel(settings.GEMINI_PRO_1_5)
+        model = genai.GenerativeModel(settings.GEMINI_FLASH_8B)
         ats_prompt = f"""
                 Hey Act Like a skilled or very experienced ATS (Application Tracking System)
                 with a deep understanding of the tech field, software engineering, data science, data analysis,
@@ -1032,7 +1042,7 @@ async def ml_crew(file: UploadFile = File(...),user_question: str = Form(...),mo
         db = MongoDB()
         payload = {
                 "endpoint": "/ml_assistant",
-                "propmt" : user_question,
+                "prompt" : user_question,
                 "Model" : model,
                 "Output" : output
             }
@@ -1043,6 +1053,62 @@ async def ml_crew(file: UploadFile = File(...),user_question: str = Form(...),mo
 
     except Exception as e:
         return {"error": str(e)}
+
+@app.post("/agrilens")
+async def analyze_image(file: UploadFile,custom_prompt: str = Form(""),token: str = Depends(oauth2_scheme)):
+    try:
+        payload = jwt.decode(token, os.getenv("TOKEN_SECRET_KEY"), algorithms=[settings.ALGORITHM])
+        email = payload.get("sub")
+        if email is None:
+            raise HTTPException(status_code=status.HTTP_401_UNAUTHORIZED, detail="Invalid token")
+        user = users_collection.find_one({"email": email})
+        if user is None:
+            raise HTTPException(status_code=status.HTTP_401_UNAUTHORIZED, detail="User not found")
+    except JWTError:
+        raise HTTPException(status_code=status.HTTP_401_UNAUTHORIZED, detail="Invalid token")
+
+
+    if not file.filename.lower().endswith(('.jpg', '.jpeg', '.png')):
+        raise HTTPException(status_code=400, detail="Invalid file type. Only JPG, JPEG, and PNG are supported.")
+
+    image_data = await file.read()
+    base64_image = encode_image(image_data)
+
+    final_prompt = custom_prompt if custom_prompt and custom_prompt.strip() else settings.AGRILENS_DEFAULT_PROMPT
+
+    try:
+        chat_completion = openai_compatible_client.chat.completions.create(
+            messages=[
+                {
+                    "role": "user",
+                    "content": [
+                        {
+                            "type": "image_url",
+                            "image_url": {
+                                "url": f"data:image/jpeg;base64,{base64_image}"
+                            },
+                        },
+                        {"type": "text", "text": final_prompt},
+                    ],
+                }
+            ],
+            model="llama-3.2-90b-vision-preview"
+        )
+
+        response_content = chat_completion.choices[0].message.content
+        db = MongoDB()
+        payload = {
+                "endpoint": "/agrilens",
+                "custom_prompt" : final_prompt,
+                "Output" : response_content
+            }
+        mongo_data = {"Document": payload}
+        result = db.insert_data(mongo_data)
+        print(result)
+        return ResponseText(response=response_content)
+
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=f"Error during analysis: {e}")
 
 if __name__ == '__main__':
     import uvicorn