File size: 7,319 Bytes
c87c295 |
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 |
import gradio as gr
import random
import time, os
import copy
import re
import torch
from rich.console import Console
from rich.table import Table
from datetime import datetime
from threading import Thread
from typing import Optional
from transformers import TextIteratorStreamer
from utils.special_tok_llama2 import (
B_CODE,
E_CODE,
B_RESULT,
E_RESULT,
B_INST,
E_INST,
B_SYS,
E_SYS,
DEFAULT_PAD_TOKEN,
DEFAULT_BOS_TOKEN,
DEFAULT_EOS_TOKEN,
DEFAULT_UNK_TOKEN,
IGNORE_INDEX,
)
from finetuning.conversation_template import (
json_to_code_result_tok_temp,
msg_to_code_result_tok_temp,
)
import warnings
warnings.filterwarnings("ignore", category=UserWarning, module="transformers")
os.environ["TF_CPP_MIN_LOG_LEVEL"] = "2"
from code_interpreter.LlamaCodeInterpreter import LlamaCodeInterpreter
class StreamingLlamaCodeInterpreter(LlamaCodeInterpreter):
streamer: Optional[TextIteratorStreamer] = None
# overwirte generate function
@torch.inference_mode()
def generate(
self,
prompt: str = "[INST]\n###User : hi\n###Assistant :",
max_new_tokens=512,
do_sample: bool = True,
use_cache: bool = True,
top_p: float = 0.95,
temperature: float = 0.1,
top_k: int = 50,
repetition_penalty: float = 1.0,
) -> str:
# Get the model and tokenizer, and tokenize the user text.
self.streamer = TextIteratorStreamer(
self.tokenizer, skip_prompt=True, Timeout=5
)
input_prompt = copy.deepcopy(prompt)
inputs = self.tokenizer([prompt], return_tensors="pt")
input_tokens_shape = inputs["input_ids"].shape[-1]
eos_token_id = self.tokenizer.convert_tokens_to_ids(DEFAULT_EOS_TOKEN)
e_code_token_id = self.tokenizer.convert_tokens_to_ids(E_CODE)
kwargs = dict(
**inputs,
max_new_tokens=max_new_tokens,
do_sample=do_sample,
top_p=top_p,
temperature=temperature,
use_cache=use_cache,
top_k=top_k,
repetition_penalty=repetition_penalty,
eos_token_id=[
eos_token_id,
e_code_token_id,
], # Stop generation at either EOS or E_CODE token
streamer=self.streamer,
)
thread = Thread(target=self.model.generate, kwargs=kwargs)
thread.start()
return ""
def change_markdown_image(text: str):
modified_text = re.sub(r"!\[(.*?)\]\(\'(.*?)\'\)", r"", text)
return modified_text
def gradio_launch(model_path: str, load_in_4bit: bool = True, MAX_TRY: int = 5):
with gr.Blocks(theme=gr.themes.Monochrome()) as demo:
chatbot = gr.Chatbot(height=820, avatar_images="./assets/logo2.png")
msg = gr.Textbox()
clear = gr.Button("Clear")
interpreter = StreamingLlamaCodeInterpreter(
model_path=model_path, load_in_4bit=load_in_4bit
)
def bot(history):
user_message = history[-1][0]
interpreter.dialog.append({"role": "user", "content": user_message})
print(f"###User : [bold]{user_message}[bold]")
# print(f"###Assistant : ")
# setup
HAS_CODE = False # For now
INST_END_TOK_FLAG = False
full_generated_text = ""
prompt = interpreter.dialog_to_prompt(dialog=interpreter.dialog)
start_prompt = copy.deepcopy(prompt)
prompt = f"{prompt} {E_INST}"
_ = interpreter.generate(prompt)
history[-1][1] = ""
generated_text = ""
for character in interpreter.streamer:
history[-1][1] += character
generated_text += character
yield history
full_generated_text += generated_text
HAS_CODE, generated_code_block = interpreter.extract_code_blocks(
generated_text
)
attempt = 1
while HAS_CODE:
if attempt > MAX_TRY:
break
# if no code then doesn't have to execute it
# refine code block for history
history[-1][1] = (
history[-1][1]
.replace(f"{B_CODE}", "\n```python\n")
.replace(f"{E_CODE}", "\n```\n")
)
history[-1][1] = change_markdown_image(history[-1][1])
yield history
# replace unknown thing to none ''
generated_code_block = generated_code_block.replace(
"<unk>_", ""
).replace("<unk>", "")
(
code_block_output,
error_flag,
) = interpreter.execute_code_and_return_output(
f"{generated_code_block}"
)
code_block_output = interpreter.clean_code_output(code_block_output)
generated_text = (
f"{generated_text}\n{B_RESULT}\n{code_block_output}\n{E_RESULT}\n"
)
full_generated_text += (
f"\n{B_RESULT}\n{code_block_output}\n{E_RESULT}\n"
)
# append code output
history[-1][1] += f"\n```RESULT\n{code_block_output}\n```\n"
history[-1][1] = change_markdown_image(history[-1][1])
yield history
prompt = f"{prompt} {generated_text}"
_ = interpreter.generate(prompt)
for character in interpreter.streamer:
history[-1][1] += character
generated_text += character
history[-1][1] = change_markdown_image(history[-1][1])
yield history
HAS_CODE, generated_code_block = interpreter.extract_code_blocks(
generated_text
)
if generated_text.endswith("</s>"):
break
attempt += 1
interpreter.dialog.append(
{
"role": "assistant",
"content": generated_text.replace("<unk>_", "")
.replace("<unk>", "")
.replace("</s>", ""),
}
)
print("----------\n" * 2)
print(interpreter.dialog)
print("----------\n" * 2)
return history[-1][1]
def user(user_message, history):
return "", history + [[user_message, None]]
msg.submit(user, [msg, chatbot], [msg, chatbot], queue=False).then(
bot, chatbot, chatbot
)
clear.click(lambda: None, None, chatbot, queue=False)
demo.queue()
demo.launch()
if __name__ == "__main__":
import argparse
parser = argparse.ArgumentParser(description="Process path for LLAMA2_FINETUNEED.")
parser.add_argument(
"--path",
type=str,
required=True,
help="Path to the finetuned LLAMA2 model.",
default="./output/llama-2-7b-codellama-ci",
)
args = parser.parse_args()
gradio_launch(model_path=args.path, load_in_4bit=True)
|