app.py

import streamlit as st
from transformers import pipeline
import os
import argilla as rg

import httpx
api_key = os.getenv("api_key")
api_url="https://Rajashreee-my-argilla2.hf.space"
print("Argilla response:", response.status_code, response.text)
client = rg.Argilla(
    api_url=api_url,
    api_key=api_key
)
# input_text = st.text_area("Enter a movie review")
# dataset_name = "movie-review-feedback"
# api_url = "http://localhost:6900"  # ✅ internal access between containers
# "
# api_key = os.getenv("ARGILLA_API_KEY")

# rg.init(api_url=api_url, api_key=api_key)  
# settings = rg.Settings(
#     guidelines="These are some guidelines.",
#     fields=[
#         rg.TextField(
#             name="text",
#         ),
#     ],
#     questions=[
#         rg.LabelQuestion(
#             name="label",
#             labels=['positve','negative']
#         ),
#     ],
# )

# dataset = rg.Dataset(
#     name="movie-review-feedback",
#     workspace="my_workspace",
#     settings=settings,
# )

# dataset.create()

# import streamlit as st
# from transformers import pipeline
# from datasets import Dataset
# import argilla as rg

# import os

# api_url = "http://localhost:6900"
# api_key = os.getenv("ARGILLA_API_KEY")

# # Set up Argilla (same Hugging Face Space)
# rg.configure(
#     api_url=api_url,  
#     api_key=api_key 
# )

# st.title("Movie Review Sentiment Logger")

# input_text = st.text_area("Enter a movie review")

# if st.button("Analyze and Log"):
#     if input_text:
#         classifier = pipeline("sentiment-analysis")
#         result = classifier(input_text)[0]
        
#         # Log to Argilla
#         record = rg.TextClassificationRecord(
#             text=input_text,
#             prediction=[(result["label"], result["score"])]
#         )
#         rg.log([record], name="movie-review-feedback")
        
#         st.success(f"Logged with prediction: {result['label']} ({round(result['score'],2)})")