Spaces:
Runtime error
Runtime error
File size: 2,799 Bytes
a9f30f1 5e4742f a9f30f1 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 |
# -*- coding: utf-8 -*-
"""Fitness.ipynb
Automatically generated by Colab.
Original file is located at
https://colab.research.google.com/drive/1rF0K4Qn52EJlq0ULtyo9jFhQ8aZrgV8P
"""
#pip install replicate duckduckgo-search
import replicate
from duckduckgo_search import DDGS
# 🔐 IBM Granite on Replicate
REPLICATE_API_TOKEN = "r8_fN4twLwm1FOIgMwxX0IpFmpVtUbPxEV3f1hNd"
client = replicate.Client(api_token=REPLICATE_API_TOKEN)
# 🔎 Optional: search tool
def search_duckduckgo(query):
with DDGS() as ddgs:
results = ddgs.text(query, max_results=3)
return "\n".join([f"- {r['title']}: {r['href']}" for r in results])
# 🧠 LLM query function
def query_replicate(prompt, model="ibm-granite/granite-3.3-8b-instruct"):
output = client.run(model, input={"prompt": prompt, "max_new_tokens": 300})
return "".join(output)
# 🤖 Agent 1: Workout Recommender
def workout_agent(prompt):
system_prompt = "You are a personal workout coach. Recommend home-based fitness routines based on user goals and preferences."
return query_replicate(f"{system_prompt}\n\nUser: {prompt}")
# 🍎 Agent 2: Nutrition Coach
def nutrition_agent(prompt):
system_prompt = "You are a friendly nutritionist. Suggest healthy, affordable meal ideas based on user input. Be simple and practical."
return query_replicate(f"{system_prompt}\n\nUser: {prompt}")
# 💡 Agent 3: Motivational Buddy
def motivation_agent(prompt):
system_prompt = "You are a motivational fitness buddy. Share tips, inspiration, and strategies to stay consistent with healthy habits."
return query_replicate(f"{system_prompt}\n\nUser: {prompt}")
def fitness_buddy_router(user_prompt):
workout_keywords = ["workout", "exercise", "routine", "stretch", "home gym"]
nutrition_keywords = ["diet", "meal", "protein", "food", "nutrition", "eat"]
motivation_keywords = ["motivation", "lazy", "can't", "habit", "inspiration", "routine", "mental"]
prompt_lower = user_prompt.lower()
if any(word in prompt_lower for word in workout_keywords):
return workout_agent(user_prompt)
elif any(word in prompt_lower for word in nutrition_keywords):
return nutrition_agent(user_prompt)
elif any(word in prompt_lower for word in motivation_keywords):
return motivation_agent(user_prompt)
else:
# default fallback agent
return "🤖 Please clarify: are you asking about workouts, meals, or motivation?"
import gradio as gr
gr.Interface(
fn=fitness_buddy_router,
inputs=gr.Textbox(lines=2, placeholder="Ask about fitness, meals, or motivation..."),
outputs=gr.Textbox(label="Fitness Buddy Says"),
title="💪 Fitness Buddy AI",
description="Your personal AI coach for workouts, nutrition, and motivation!"
).launch()
|