File size: 17,089 Bytes
9e7090f |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 |
import time
import uuid
from selenium import webdriver
from selenium.webdriver.chrome.options import Options
from selenium.webdriver.common.by import By
from selenium.webdriver.support import expected_conditions as EC
from selenium.webdriver.support.ui import WebDriverWait
import click
import requests
from requests import get
from uuid import uuid4
from re import findall
from requests.exceptions import RequestException
from curl_cffi.requests import get, RequestsError
import g4f
from random import randint
from PIL import Image
import io
import re
import json
import yaml
from ..AIutel import Optimizers
from ..AIutel import Conversation
from ..AIutel import AwesomePrompts, sanitize_stream
from ..AIbase import Provider, AsyncProvider
from Helpingai_T2 import Perplexity
from webscout import exceptions
from typing import Any, AsyncGenerator, Dict
import logging
import httpx
class AsyncLLAMA2(AsyncProvider):
def __init__(
self,
is_conversation: bool = True,
max_tokens: int = 800,
temperature: float = 0.75,
presence_penalty: int = 0,
frequency_penalty: int = 0,
top_p: float = 0.9,
model: str = "meta/meta-llama-3-70b-instruct",
timeout: int = 30,
intro: str = None,
filepath: str = None,
update_file: bool = True,
proxies: dict = {},
history_offset: int = 10250,
act: str = None,
):
"""Instantiates LLAMA2
Args:
is_conversation (bool, optional): Flag for chatting conversationally. Defaults to True.
max_tokens (int, optional): Maximum number of tokens to be generated upon completion. Defaults to 800.
temperature (float, optional): Charge of the generated text's randomness. Defaults to 0.75.
presence_penalty (int, optional): Chances of topic being repeated. Defaults to 0.
frequency_penalty (int, optional): Chances of word being repeated. Defaults to 0.
top_p (float, optional): Sampling threshold during inference time. Defaults to 0.9.
model (str, optional): LLM model name. Defaults to "meta/llama-2-70b-chat".
timeout (int, optional): Http request timeout. Defaults to 30.
intro (str, optional): Conversation introductory prompt. Defaults to None.
filepath (str, optional): Path to file containing conversation history. Defaults to None.
update_file (bool, optional): Add new prompts and responses to the file. Defaults to True.
proxies (dict, optional): Http request proxies. Defaults to {}.
history_offset (int, optional): Limit conversation history to this number of last texts. Defaults to 10250.
act (str|int, optional): Awesome prompt key or index. (Used as intro). Defaults to None.
"""
self.is_conversation = is_conversation
self.max_tokens_to_sample = max_tokens
self.model = model
self.temperature = temperature
self.presence_penalty = presence_penalty
self.frequency_penalty = frequency_penalty
self.top_p = top_p
self.chat_endpoint = "https://www.llama2.ai/api"
self.stream_chunk_size = 64
self.timeout = timeout
self.last_response = {}
self.headers = {
"Content-Type": "application/json",
"Referer": "https://www.llama2.ai/",
"Content-Type": "text/plain;charset=UTF-8",
"Origin": "https://www.llama2.ai",
}
self.__available_optimizers = (
method
for method in dir(Optimizers)
if callable(getattr(Optimizers, method)) and not method.startswith("__")
)
Conversation.intro = (
AwesomePrompts().get_act(
act, raise_not_found=True, default=None, case_insensitive=True
)
if act
else intro or Conversation.intro
)
self.conversation = Conversation(
is_conversation, self.max_tokens_to_sample, filepath, update_file
)
self.conversation.history_offset = history_offset
self.session = httpx.AsyncClient(
headers=self.headers,
proxies=proxies,
)
async def ask(
self,
prompt: str,
stream: bool = False,
raw: bool = False,
optimizer: str = None,
conversationally: bool = False,
) -> dict | AsyncGenerator:
"""Chat with AI asynchronously.
Args:
prompt (str): Prompt to be send.
stream (bool, optional): Flag for streaming response. Defaults to False.
raw (bool, optional): Stream back raw response as received. Defaults to False.
optimizer (str, optional): Prompt optimizer name - `[code, shell_command]`. Defaults to None.
conversationally (bool, optional): Chat conversationally when using optimizer. Defaults to False.
Returns:
dict|AsyncGeneraror[dict] : ai content
```json
{
"text" : "How may I help you today?"
}
```
"""
conversation_prompt = self.conversation.gen_complete_prompt(prompt)
if optimizer:
if optimizer in self.__available_optimizers:
conversation_prompt = getattr(Optimizers, optimizer)(
conversation_prompt if conversationally else prompt
)
else:
raise Exception(
f"Optimizer is not one of {self.__available_optimizers}"
)
payload = {
"prompt": f"{conversation_prompt}<s>[INST] {prompt} [/INST]",
"model": self.model,
"systemPrompt": "You are a helpful assistant.",
"temperature": self.temperature,
"topP": self.top_p,
"maxTokens": self.max_tokens_to_sample,
"image": None,
"audio": None,
}
async def for_stream():
async with self.session.stream(
"POST", self.chat_endpoint, json=payload, timeout=self.timeout
) as response:
if (
not response.is_success
or not response.headers.get("Content-Type")
== "text/plain; charset=utf-8"
):
raise exceptions.FailedToGenerateResponseError(
f"Failed to generate response - ({response.status_code}, {response.reason_phrase})"
)
message_load: str = ""
async for value in response.aiter_lines():
try:
if bool(value.strip()):
message_load += value + "\n"
resp: dict = dict(text=message_load)
yield value if raw else resp
self.last_response.update(resp)
except json.decoder.JSONDecodeError:
pass
self.conversation.update_chat_history(
prompt, await self.get_message(self.last_response)
)
async def for_non_stream():
async for _ in for_stream():
pass
return self.last_response
return for_stream() if stream else await for_non_stream()
async def chat(
self,
prompt: str,
stream: bool = False,
optimizer: str = None,
conversationally: bool = False,
) -> str | AsyncGenerator:
"""Generate response `str` asynchronously.
Args:
prompt (str): Prompt to be send.
stream (bool, optional): Flag for streaming response. Defaults to False.
optimizer (str, optional): Prompt optimizer name - `[code, shell_command]`. Defaults to None.
conversationally (bool, optional): Chat conversationally when using optimizer. Defaults to False.
Returns:
str|AsyncGenerator: Response generated
"""
async def for_stream():
async_ask = await self.ask(
prompt, True, optimizer=optimizer, conversationally=conversationally
)
async for response in async_ask:
yield await self.get_message(response)
async def for_non_stream():
return await self.get_message(
await self.ask(
prompt,
False,
optimizer=optimizer,
conversationally=conversationally,
)
)
return for_stream() if stream else await for_non_stream()
async def get_message(self, response: dict) -> str:
"""Retrieves message only from response
Args:
response (str): Response generated by `self.ask`
Returns:
str: Message extracted
"""
assert isinstance(response, dict), "Response should be of dict data-type only"
return response["text"]
class LLAMA2(Provider):
def __init__(
self,
is_conversation: bool = True,
max_tokens: int = 800,
temperature: float = 0.75,
presence_penalty: int = 0,
frequency_penalty: int = 0,
top_p: float = 0.9,
model: str = "meta/meta-llama-3-70b-instruct",
timeout: int = 30,
intro: str = None,
filepath: str = None,
update_file: bool = True,
proxies: dict = {},
history_offset: int = 10250,
act: str = None,
):
"""Instantiates LLAMA2
Args:
is_conversation (bool, optional): Flag for chatting conversationally. Defaults to True.
max_tokens (int, optional): Maximum number of tokens to be generated upon completion. Defaults to 800.
temperature (float, optional): Charge of the generated text's randomness. Defaults to 0.75.
presence_penalty (int, optional): Chances of topic being repeated. Defaults to 0.
frequency_penalty (int, optional): Chances of word being repeated. Defaults to 0.
top_p (float, optional): Sampling threshold during inference time. Defaults to 0.9.
model (str, optional): LLM model name. Defaults to "meta/llama-2-70b-chat".
timeout (int, optional): Http request timeout. Defaults to 30.
intro (str, optional): Conversation introductory prompt. Defaults to None.
filepath (str, optional): Path to file containing conversation history. Defaults to None.
update_file (bool, optional): Add new prompts and responses to the file. Defaults to True.
proxies (dict, optional): Http request proxies. Defaults to {}.
history_offset (int, optional): Limit conversation history to this number of last texts. Defaults to 10250.
act (str|int, optional): Awesome prompt key or index. (Used as intro). Defaults to None.
"""
self.session = requests.Session()
self.is_conversation = is_conversation
self.max_tokens_to_sample = max_tokens
self.model = model
self.temperature = temperature
self.presence_penalty = presence_penalty
self.frequency_penalty = frequency_penalty
self.top_p = top_p
self.chat_endpoint = "https://www.llama2.ai/api"
self.stream_chunk_size = 64
self.timeout = timeout
self.last_response = {}
self.headers = {
"Content-Type": "application/json",
"Referer": "https://www.llama2.ai/",
"Content-Type": "text/plain;charset=UTF-8",
"Origin": "https://www.llama2.ai",
}
self.__available_optimizers = (
method
for method in dir(Optimizers)
if callable(getattr(Optimizers, method)) and not method.startswith("__")
)
self.session.headers.update(self.headers)
Conversation.intro = (
AwesomePrompts().get_act(
act, raise_not_found=True, default=None, case_insensitive=True
)
if act
else intro or Conversation.intro
)
self.conversation = Conversation(
is_conversation, self.max_tokens_to_sample, filepath, update_file
)
self.conversation.history_offset = history_offset
self.session.proxies = proxies
def ask(
self,
prompt: str,
stream: bool = False,
raw: bool = False,
optimizer: str = None,
conversationally: bool = False,
) -> dict:
"""Chat with AI
Args:
prompt (str): Prompt to be send.
stream (bool, optional): Flag for streaming response. Defaults to False.
raw (bool, optional): Stream back raw response as received. Defaults to False.
optimizer (str, optional): Prompt optimizer name - `[code, shell_command]`. Defaults to None.
conversationally (bool, optional): Chat conversationally when using optimizer. Defaults to False.
Returns:
dict : {}
```json
{
"text" : "How may I help you today?"
}
```
"""
conversation_prompt = self.conversation.gen_complete_prompt(prompt)
if optimizer:
if optimizer in self.__available_optimizers:
conversation_prompt = getattr(Optimizers, optimizer)(
conversation_prompt if conversationally else prompt
)
else:
raise Exception(
f"Optimizer is not one of {self.__available_optimizers}"
)
self.session.headers.update(self.headers)
payload = {
"prompt": f"{conversation_prompt}<s>[INST] {prompt} [/INST]",
"model": self.model,
"systemPrompt": "You are a helpful assistant.",
"temperature": self.temperature,
"topP": self.top_p,
"maxTokens": self.max_tokens_to_sample,
"image": None,
"audio": None,
}
def for_stream():
response = self.session.post(
self.chat_endpoint, json=payload, stream=True, timeout=self.timeout
)
if (
not response.ok
or not response.headers.get("Content-Type")
== "text/plain; charset=utf-8"
):
raise exceptions.FailedToGenerateResponseError(
f"Failed to generate response - ({response.status_code}, {response.reason})"
)
message_load: str = ""
for value in response.iter_lines(
decode_unicode=True,
delimiter="\n",
chunk_size=self.stream_chunk_size,
):
try:
if bool(value.strip()):
message_load += value + "\n"
resp: dict = dict(text=message_load)
yield value if raw else resp
self.last_response.update(resp)
except json.decoder.JSONDecodeError:
pass
self.conversation.update_chat_history(
prompt, self.get_message(self.last_response)
)
def for_non_stream():
for _ in for_stream():
pass
return self.last_response
return for_stream() if stream else for_non_stream()
def chat(
self,
prompt: str,
stream: bool = False,
optimizer: str = None,
conversationally: bool = False,
) -> str:
"""Generate response `str`
Args:
prompt (str): Prompt to be send.
stream (bool, optional): Flag for streaming response. Defaults to False.
optimizer (str, optional): Prompt optimizer name - `[code, shell_command]`. Defaults to None.
conversationally (bool, optional): Chat conversationally when using optimizer. Defaults to False.
Returns:
str: Response generated
"""
def for_stream():
for response in self.ask(
prompt, True, optimizer=optimizer, conversationally=conversationally
):
yield self.get_message(response)
def for_non_stream():
return self.get_message(
self.ask(
prompt,
False,
optimizer=optimizer,
conversationally=conversationally,
)
)
return for_stream() if stream else for_non_stream()
def get_message(self, response: dict) -> str:
"""Retrieves message only from response
Args:
response (str): Response generated by `self.ask`
Returns:
str: Message extracted
"""
assert isinstance(response, dict), "Response should be of dict data-type only"
return response["text"] |