### اول كود للابيلنق اشتغل بس مافرق بين ريكوند و نت ريكومند
import numpy as np
import streamlit as st
from openai import OpenAI
import os
from dotenv import load_dotenv
import random

# Load environment variables
os.environ["BROWSER_GATHERUSAGESTATS"] = "false"
load_dotenv()

# Initialize the client
client = OpenAI(
    base_url="https://api-inference.huggingface.co/v1",
    api_key=os.environ.get('GP2')  # Replace with your Huggingface token
)

# Initialize session state variables if they are not already defined
if "labels" not in st.session_state:
    st.session_state.labels = []
if "few_shot_examples" not in st.session_state:
    st.session_state.few_shot_examples = []
if "examples_to_classify" not in st.session_state:
    st.session_state.examples_to_classify = []
if "messages" not in st.session_state:
    st.session_state.messages = []

# Sidebar for model selection and temperature setting
selected_model = st.sidebar.selectbox("Select Model", ["Meta-Llama-3-8B"], key="model_select")
temp_values = st.sidebar.slider('Select a temperature value', 0.0, 1.0, 0.5, key="temp_slider")

# Reset conversation button
st.sidebar.button('Reset Chat', on_click=lambda: (st.session_state.update(conversation=[], messages=[])), key="reset_button")

# Main task selection: Data Generation or Data Labeling
task_choice = st.selectbox("Choose Task", ["Data Generation", "Data Labeling"], key="task_choice_select")

# Data Generation Section
if task_choice == "Data Generation":
    classification_type = st.selectbox(
        "Choose Classification Type",
        ["Sentiment Analysis", "Binary Classification", "Multi-Class Classification"],
        key="classification_type_select"
    )

    # Define labels based on classification type
    if classification_type == "Sentiment Analysis":
        st.session_state.labels = ["Positive", "Negative", "Neutral"]
        st.write("Sentiment Analysis: Positive, Negative, Neutral")
    elif classification_type == "Binary Classification":
        label_1 = st.text_input("Enter first class", key="binary_class_1")
        label_2 = st.text_input("Enter second class", key="binary_class_2")
        st.session_state.labels = [label_1, label_2]
    elif classification_type == "Multi-Class Classification":
        num_classes = st.slider("How many classes?", 3, 10, 3, key="num_classes_slider")
        st.session_state.labels = [st.text_input(f"Class {i+1}", key=f"class_input_{i+1}") for i in range(num_classes)]

    # Domain selection
    domain = st.selectbox("Choose Domain", ["Restaurant reviews", "E-commerce reviews", "Custom"], key="domain_select")
    if domain == "Custom":
        domain = st.text_input("Specify custom domain", key="custom_domain_input")

    # Word count selection
    min_words = st.number_input("Minimum words per example", min_value=10, max_value=90, value=10, key="min_words_input")
    max_words = st.number_input("Maximum words per example", min_value=10, max_value=90, value=90, key="max_words_input")

    # Few-shot examples option
    few_shot = st.radio("Do you want to use few-shot examples?", ["Yes", "No"], key="few_shot_radio")
    if few_shot == "Yes":
        num_examples = st.slider("How many few-shot examples?", 1, 5, 1, key="num_examples_slider")
        st.session_state.few_shot_examples = [
            {
                "content": st.text_area(f"Example {i+1} Text", key=f"example_text_{i+1}"),
                "label": st.selectbox(f"Label for Example {i+1}", st.session_state.labels, key=f"label_select_{i+1}")
            }
            for i in range(num_examples)
        ]
    else:
        st.session_state.few_shot_examples = []

    # Number of examples to generate
    num_to_generate = st.number_input("How many examples to generate?", min_value=1, max_value=100, value=10, key="num_to_generate_input")

    # User prompt text field
    user_prompt = st.text_area("Enter your prompt to guide example generation", "", key="user_prompt_text_area")

    # System prompt generation
    system_prompt = f"You are a professional {classification_type.lower()} expert. Your role is to generate data for {domain}.\n\n"
    if st.session_state.few_shot_examples:
        system_prompt += "Use the following few-shot examples as a reference:\n"
        for example in st.session_state.few_shot_examples:
            system_prompt += f"Example: {example['content']} \n Label: {example['label']}\n"
    system_prompt += f"Generate {num_to_generate} unique examples with diverse phrasing.\n"
    system_prompt += f"Each example should have between {min_words} and {max_words} words.\n"
    system_prompt += f"Use the labels specified: {', '.join(st.session_state.labels)}.\n"
    if user_prompt:
        system_prompt += f"Additional instructions: {user_prompt}\n"

    st.write("System Prompt:")
    st.code(system_prompt)

    if st.button("Generate Examples", key="generate_examples_button"):
        # Generate examples by concatenating all inputs and sending it to the model
        with st.spinner("Generating..."):
            st.session_state.messages.append({"role": "system", "content": system_prompt})

            try:
                stream = client.chat.completions.create(
                    model=selected_model,
                    messages=[
                        {"role": m["role"], "content": m["content"]}
                        for m in st.session_state.messages
                    ],
                    temperature=temp_values,
                    stream=True,
                    max_tokens=3000,
                )
                response = ""
                for chunk in stream:
                    response += chunk['choices'][0]['delta'].get('content', '')
                st.write(response)
            except Exception as e:
                st.error(f"Error during generation: {e}")

            st.session_state.messages.append({"role": "assistant", "content": response})

# Data Labeling Section
else:
    # Classification Type and Labels Setup
    classification_type = st.selectbox("Choose Classification Type", ["Sentiment Analysis", "Binary Classification", "Multi-Class Classification"], key="classification_type_labeling")
    
    if classification_type == "Sentiment Analysis":
        st.session_state.labels = ["Positive", "Negative", "Neutral"]
        st.write("Sentiment Analysis labels: Positive, Negative, Neutral")
    elif classification_type == "Binary Classification":
        label_1 = st.text_input("Enter first class", key="binary_class_1_labeling")
        label_2 = st.text_input("Enter second class", key="binary_class_2_labeling")
        st.session_state.labels = [label_1, label_2]
    elif classification_type == "Multi-Class Classification":
        num_classes = st.slider("How many classes?", 3, 10, 3, key="num_classes_labeling")
        st.session_state.labels = [st.text_input(f"Class {i+1}", key=f"class_input_labeling_{i+1}") for i in range(num_classes)]

    # Few-shot examples for labeling
    use_few_shot = st.radio("Do you want to use few-shot examples?", ["Yes", "No"], key="use_few_shot_labeling")
    if use_few_shot == "Yes":
        num_examples = st.slider("How many few-shot examples?", 1, 5, 1, key="few_shot_num_labeling")
        st.session_state.few_shot_examples = [
            {
                "content": st.text_area(f"Example {i+1} Text", key=f"example_text_labeling_{i+1}"),
                "label": st.selectbox(f"Label for Example {i+1}", st.session_state.labels, key=f"label_select_labeling_{i+1}")
            }
            for i in range(num_examples)
        ]
    else:
        st.session_state.few_shot_examples = []

    # Input Examples for Classification
    num_to_classify = st.number_input("How many examples do you want to classify?", min_value=1, max_value=100, value=5, key="num_to_classify_input")
    st.session_state.examples_to_classify = [st.text_area(f"Example {i+1} Text", key=f"example_classify_text_{i+1}") for i in range(num_to_classify)]

    # Placeholder for classification function (can be replaced with actual API call)
    def classify_examples(examples, labels):
        classified_results = [{"example": ex, "label": random.choice(labels)} for ex in examples]
        return classified_results

    # Classification results display
    if st.button("Classify Examples", key="classify_button"):
        results = classify_examples(st.session_state.examples_to_classify, st.session_state.labels)
        st.write("Classification Results:")
        for result in results:
            st.write(f"Example: {result['example']}\nLabel: {result['label']}\n")
شحح