app.py

from transformers import BertTokenizerFast,TFBertForSequenceClassification,TextClassificationPipeline
import numpy as np
import tensorflow as tf
import gradio as gr 
import openai
import os

# Sentiment Analysis Pre-Trained Model
model_path = "leadingbridge/sentiment-analysis"
tokenizer = BertTokenizerFast.from_pretrained(model_path)
model = TFBertForSequenceClassification.from_pretrained(model_path, id2label={0: 'negative', 1: 'positive'} )

def sentiment_analysis(text):
  pipe = TextClassificationPipeline(model=model, tokenizer=tokenizer)
  result = pipe(text)
  return result


# Open AI Chatbot Model
openai.api_key = "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"

start_sequence = "\nAI:"
restart_sequence = "\nHuman: "

prompt = "You can discuss any topic with the Chatbot assistant in Chinese by typing it in here"

def openai_create(prompt):

    response = openai.Completion.create(
    model="text-davinci-003",
    prompt=prompt,
    temperature=0.9,
    max_tokens=1024,
    top_p=1,
    frequency_penalty=0,
    presence_penalty=0.6,
    stop=[" Human:", " AI:"]
    )

    return response.choices[0].text

def chatgpt_clone(input, history):
    history = history or []
    s = list(sum(history, ()))
    s.append(input)
    inp = ' '.join(s)
    output = openai_create(inp)
    history.append((input, output))
    return history, history


# Open AI Chinese Translation Model
def translate_to_chinese(text_to_translate):
    response = openai.Completion.create(
      model="text-davinci-003",
      prompt=f"Translate this short sentence into Chinese:\n\n{text_to_translate}\n\n1.",
      temperature=0.3,
      max_tokens=1024,
      top_p=1.0,
      frequency_penalty=0.0,
      presence_penalty=0.0
    )
    return response.choices[0].text.strip()

# Open AI English Translation Model
def translate_to_english(text_to_translate):
    response = openai.Completion.create(
      model="text-davinci-003",
      prompt=f"Translate this short sentence into English:\n\n{text_to_translate}\n\n1.",
      temperature=0.3,
      max_tokens=1024,
      top_p=1.0,
      frequency_penalty=0.0,
      presence_penalty=0.0
    )
    return response.choices[0].text.strip()


    
# Gradio Output Model
with gr.Blocks() as demo:
    gr.Markdown("Choose the Chinese NLP model you want to use from the tabs")
    with gr.Tab("OpenAI Chatbot"):
        chatbot = gr.Chatbot()
        message = gr.Textbox(placeholder=prompt)
        state = gr.State()
        submit = gr.Button("SEND")
        submit.click(chatgpt_clone, inputs=[message, state], outputs=[chatbot, state])
    with gr.Tab("Sentiment Analysis"):
        inputs = gr.Textbox(placeholder="Enter a Chinese positive or negative sentence here")
        outputs = gr.Textbox(label="Sentiment Analysis")
        proceed_button = gr.Button("proceed")           
        proceed_button.click(fn=sentiment_analysis, inputs=inputs, outputs=outputs)
    with gr.Tab("Translation to Chinese"):
        inputs = gr.Textbox(placeholder="Enter a short English sentence to translate to Chinese here.")
        outputs = gr.Textbox(label="Translation Result")
        proceed_button = gr.Button("Translate")
        proceed_button.click(fn=translate_to_chinese, inputs=inputs, outputs=outputs)
    with gr.Tab("Translation to English"):
        inputs = gr.Textbox(placeholder="Enter a Short sentence to translate to English here.")
        outputs = gr.Textbox(label="Translation Result")
        proceed_button = gr.Button("Translate")
        proceed_button.click(fn=translate_to_english, inputs=inputs, outputs=outputs)



demo.launch(inline=False)