import gradio as gr
import os
import sys
import json
import requests
MODEL = "gpt-4"
API_URL = os.getenv("API_URL")
DISABLED = os.getenv("DISABLED") == 'True'
OPENAI_API_KEY = os.getenv("OPENAI_API_KEY")
NUM_THREADS = int(os.getenv("NUM_THREADS"))
print (NUM_THREADS)
def exception_handler(exception_type, exception, traceback):
print("%s: %s" % (exception_type.__name__, exception))
sys.excepthook = exception_handler
sys.tracebacklimit = 0
#https://github.com/gradio-app/gradio/issues/3531#issuecomment-1484029099
def parse_codeblock(text):
lines = text.split("\n")
for i, line in enumerate(lines):
if "```" in line:
if line != "```":
lines[i] = f'
'
else:
lines[i] = '
'
else:
if i > 0:
lines[i] = "
" + line.replace("<", "<").replace(">", ">")
return "".join(lines)
def predict(inputs, top_p, temperature, chat_counter, chatbot, history, request:gr.Request):
payload = {
"model": MODEL,
"messages": [{"role": "system", "content": """You are CoreTransformationBot, a large language model. As a Core Transformation Chatbot, you are designed to guide through the Core Transformation process, a method developed by Connirae and Tamara Andreas. This process is designed to help achieve profound and lasting changes in behavior, feelings, and responses by connecting with one of 5 core states: being, okness, peace, love, or oneness. The method aims to transform perceived challenges into opportunities for personal growth and understanding.
To begin a session, please identify a specific behavior or feeling you wish to change. It can be anything from a habit you wish to break, an emotion you wish to manage better, or a reaction you want to change.
Once you've identified this behavior or feeling, we will go through a sequence of steps together:
Find the positive intention: We will explore what positive intention this particular part of you might have. Even if it's causing problems right now, we're assuming that it has a positive goal in mind.
Chain to the higher-level intention: After identifying the positive intention, we'll investigate what this part ultimately wants for you. We do this by asking, "If you had that (the positive intention), what would that do for you?" We repeat this step, each time with the new positive intention, until we reach your core state.
Imagine living from the core state: When we reach your core state, you'll be invited to imagine how your life and behavior would change if you could live from this state.
Integrate this core state into your life: We will then explore how this core state could be integrated into your daily life.
During our conversation, you may encounter parts of yourself that object or resist the process. This is completely normal and can be addressed by using the same Core Transformation process with these resistant parts.
After the steps, we'll also practice "future pacing", where you'll envision bringing this new state into your future life.
Gently guide the user towards a positive behavior change following these steps:"""},{"role": "user", "content": f"{inputs}"}],
"temperature" : 1.0,
"top_p":1.0,
"n" : 1,
"stream": True,
"presence_penalty":0,
"frequency_penalty":0,
}
headers = {
"Content-Type": "application/json",
"Authorization": f"Bearer {OPENAI_API_KEY}",
"Headers": f"{request.kwargs['headers']}"
}
# print(f"chat_counter - {chat_counter}")
if chat_counter != 0 :
messages = []
for i, data in enumerate(history):
if i % 2 == 0:
role = 'user'
else:
role = 'assistant'
message = {}
message["role"] = role
message["content"] = data
messages.append(message)
message = {}
message["role"] = "user"
message["content"] = inputs
messages.append(message)
payload = {
"model": MODEL,
"messages": messages,
"temperature" : temperature,
"top_p": top_p,
"n" : 1,
"stream": True,
"presence_penalty":0,
"frequency_penalty":0,
}
chat_counter += 1
history.append(inputs)
token_counter = 0
partial_words = ""
counter = 0
try:
# make a POST request to the API endpoint using the requests.post method, passing in stream=True
response = requests.post(API_URL, headers=headers, json=payload, stream=True)
response_code = f"{response}"
#if response_code.strip() != "":
# #print(f"response code - {response}")
# raise Exception(f"Sorry, hitting rate limit. Please try again later. {response}")
for chunk in response.iter_lines():
#Skipping first chunk
if counter == 0:
counter += 1
continue
#counter+=1
# check whether each line is non-empty
if chunk.decode() :
chunk = chunk.decode()
# decode each line as response data is in bytes
if len(chunk) > 12 and "content" in json.loads(chunk[6:])['choices'][0]['delta']:
partial_words = partial_words + json.loads(chunk[6:])['choices'][0]["delta"]["content"]
if token_counter == 0:
history.append(" " + partial_words)
else:
history[-1] = partial_words
token_counter += 1
yield [(parse_codeblock(history[i]), parse_codeblock(history[i + 1])) for i in range(0, len(history) - 1, 2) ], history, chat_counter, response, gr.update(interactive=False), gr.update(interactive=False) # resembles {chatbot: chat, state: history}
except Exception as e:
print (f'error found: {e}')
yield [(parse_codeblock(history[i]), parse_codeblock(history[i + 1])) for i in range(0, len(history) - 1, 2) ], history, chat_counter, response, gr.update(interactive=True), gr.update(interactive=True)
print(json.dumps({"chat_counter": chat_counter, "payload": payload, "partial_words": partial_words, "token_counter": token_counter, "counter": counter}))
def reset_textbox():
return gr.update(value='', interactive=False), gr.update(interactive=False)
title = """Core Transformation Chatbot
"""
if DISABLED:
title = """This app has reached OpenAI's usage limit. We are currently requesting an increase in our quota. Please check back in a few days.
"""
description = """Language models can be conditioned to act like dialogue agents through a conversational prompt that typically takes the form:
```
User:
Assistant:
User:
Assistant:
...
```
In this app, you can explore the outputs of a gpt-3.5 LLM.
"""
theme = gr.themes.Default(primary_hue="green")
with gr.Blocks(css = """#col_container { margin-left: auto; margin-right: auto;}
#chatbot {height: 520px; overflow: auto;}""",
theme=theme) as demo:
gr.HTML(title)
gr.HTML("""Core Transformation Chatbot.""")
#gr.HTML('''
Duplicate the Space and run securely with your OpenAI API Key''')
with gr.Column(elem_id = "col_container", visible=True) as main_block:
#API Key is provided by OpenAI
#openai_api_key = gr.Textbox(type='password', label="Enter only your OpenAI API key here")
chatbot = gr.Chatbot(elem_id='chatbot') #c
inputs = gr.Textbox(placeholder= "Hi there!", label= "Type an input and press Enter") #t
state = gr.State([]) #s
with gr.Row():
with gr.Column(scale=7):
b1 = gr.Button(visible=not DISABLED).style(full_width=True)
with gr.Column(scale=3):
server_status_code = gr.Textbox(label="Status code from OpenAI server", )
#inputs, top_p, temperature, top_k, repetition_penalty
with gr.Accordion("Parameters", open=False):
top_p = gr.Slider( minimum=-0, maximum=1.0, value=1.0, step=0.05, interactive=True, label="Top-p (nucleus sampling)",)
temperature = gr.Slider( minimum=-0, maximum=5.0, value=1.0, step=0.1, interactive=True, label="Temperature",)
#top_k = gr.Slider( minimum=1, maximum=50, value=4, step=1, interactive=True, label="Top-k",)
#repetition_penalty = gr.Slider( minimum=0.1, maximum=3.0, value=1.03, step=0.01, interactive=True, label="Repetition Penalty", )
chat_counter = gr.Number(value=0, visible=False, precision=0)
inputs.submit(reset_textbox, [], [inputs, b1], queue=False)
inputs.submit(predict, [inputs, top_p, temperature, chat_counter, chatbot, state], [chatbot, state, chat_counter, server_status_code, inputs, b1],) #openai_api_key
b1.click(reset_textbox, [], [inputs, b1], queue=False)
b1.click(predict, [inputs, top_p, temperature, chat_counter, chatbot, state], [chatbot, state, chat_counter, server_status_code, inputs, b1],) #openai_api_key
demo.queue(max_size=20, concurrency_count=NUM_THREADS, api_open=False).launch(share=False)