File size: 2,245 Bytes
c09cb0e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
import torch
import gradio as gr
from transformers import AutoTokenizer, pipeline, logging
from auto_gptq import AutoGPTQForCausalLM, BaseQuantizeConfig

model_name_or_path = "TheBloke/WizardCoder-Guanaco-15B-V1.1-GPTQ"
model_basename = "gptq_model-4bit-128g"

use_triton = False

device =  "cuda:0" if torch.cuda.is_available() else "cpu"

tokenizer = AutoTokenizer.from_pretrained(model_name_or_path, use_fast=True)

quantize_config = BaseQuantizeConfig(
    bits=4,  # quantize model to 4-bit
    group_size=128,  # it is recommended to set the value to 128
    desc_act=False,  # set to False can significantly speed up inference but the perplexity may slightly bad 
)

model = AutoGPTQForCausalLM.from_quantized(model_name_or_path,
                                           model_basename=model_basename,
                                           use_safetensors=True,
                                           trust_remote_code=False,
                                           device=device,
                                           use_triton=use_triton,
                                           quantize_config=quantize_config,
                                           cache_dir="models/"
                                           )

"""
To download from a specific branch, use the revision parameter, as in this example:

model = AutoGPTQForCausalLM.from_quantized(model_name_or_path,
        revision="gptq-4bit-32g-actorder_True",
        model_basename=model_basename,
        use_safetensors=True,
        trust_remote_code=False,
        device="cuda:0",
        quantize_config=None)
"""


def code_gen(text):
    logging.set_verbosity(logging.CRITICAL)

    print("*** Pipeline:")
    pipe = pipeline(
        "text-generation",
        model=model,
        tokenizer=tokenizer,
        max_new_tokens=124,
        temperature=0.7,
        top_p=0.95,
        repetition_penalty=1.15
    )

    response = pipe(text)
    print(response)
    
    return response[0]['generated_text']


iface = gr.Interface(fn=code_gen,
                     inputs=gr.inputs.Textbox(
                         label="Input Source Code"),
                     outputs="text",
                     title="Code Generation")

iface.launch()