File size: 7,315 Bytes
eb3516b
 
 
379a33b
eb3516b
b74ae79
 
eb3516b
62335f3
5d247de
 
eb3516b
b74ae79
 
3a2b82e
46c0344
 
 
eb3516b
 
 
 
 
379a33b
46c0344
eb3516b
bfcdf7c
 
47a10f0
 
bfcdf7c
 
 
 
eb3516b
b74ae79
 
 
eb3516b
bfcdf7c
46c0344
b74ae79
eb3516b
 
 
 
 
 
bfcdf7c
 
 
 
2b2be0b
 
 
 
379a33b
2b2be0b
379a33b
2b2be0b
 
eb3516b
bfcdf7c
 
 
 
 
 
 
 
eb3516b
 
46c0344
 
eb3516b
379a33b
 
bfcdf7c
2b2be0b
 
 
 
 
 
 
 
379a33b
2b2be0b
41ec48f
 
2b2be0b
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
379a33b
2b2be0b
 
b74ae79
eb3516b
379a33b
eb3516b
 
2b2be0b
eb3516b
1d1c0df
eb3516b
b74ae79
eb3516b
 
 
 
 
 
 
 
b74ae79
eb3516b
 
379a33b
eb3516b
 
 
 
 
b74ae79
 
 
e723c27
379a33b
b74ae79
eb3516b
 
 
 
 
b74ae79
eb3516b
 
379a33b
eb3516b
 
 
 
 
 
 
9043986
e723c27
2b2be0b
 
 
 
379a33b
 
5f52d2c
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
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
import gradio as gr
import os
import sys
import json
import requests
import random


MODEL = "gpt-4o-mini"
API_URL = os.getenv("API_URL")
#API_URL = "https://api.openai.com/v1/chat/completions"
DISABLED = os.getenv("DISABLED") == 'True'
OPENAI_API_KEYS = os.getenv("OPENAI_API_KEYS").split(',')
print (API_URL)
#print (OPENAI_API_KEYS)
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

def predict(inputs, top_p, temperature, chat_counter, chatbot, history, request:gr.Request):
    payload = {
        "model": MODEL,
        "messages": [{"role": "user", "content": f"{inputs}"}],
        "temperature": temperature,
        "top_p": top_p,
        "n" : 1,
        "stream": True,
        "presence_penalty":0,
        "frequency_penalty":0,
    }
    OPENAI_API_KEY = random.choice(OPENAI_API_KEYS)
    print (OPENAI_API_KEY)
    headers_dict = {key.decode('utf-8'): value.decode('utf-8') for key, value in request.headers.raw}
    headers = {
        "Content-Type": "application/json",
        "Authorization": f"Bearer {OPENAI_API_KEY}",
        "Headers": f"{headers_dict}"
    }

    # 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() != "<Response [200]>":
        #    #print(f"response code - {response}")
        #    raise Exception(f"Sorry, hitting rate limit. Please try again later. {response}")

        for chunk in response.iter_lines():
            #print (chunk)
            #sys.stdout.flush()
            #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 [(history[i], 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 [(history[i], 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 = """<h1 align="center">GPT-4o Mini: Research Preview (Short-Term Availability)</h1>"""
if DISABLED:
    title = """<h1 align="center" style="color:red">This app has reached OpenAI's usage limit. Please check back tomorrow.</h1>"""
description = """Language models can be conditioned to act like dialogue agents through a conversational prompt that typically takes the form:
```
User: <utterance>
Assistant: <utterance>
User: <utterance>
Assistant: <utterance>
...
```
In this app, you can explore the outputs of a gpt-4 turbo 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("""<h3 align="center">This app provides you full access to GPT-4o mini (128K token limit). You don't need any OPENAI API key.</h3>""")

    #gr.HTML('''<center><a href="https://huggingface.co/spaces/ysharma/ChatGPT4?duplicate=true"><img src="https://bit.ly/3gLdBN6" alt="Duplicate Space"></a>Duplicate the Space and run securely with your OpenAI API Key</center>''')
    with gr.Column(elem_id = "col_container", visible=True) as main_block:
        #GPT4 API Key is provided by Huggingface
        #openai_api_key = gr.Textbox(type='password', label="Enter only your GPT4 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=10, default_concurrency_limit=NUM_THREADS, api_open=False).launch(share=False)