Hugging Face's logo

Spaces:
chanhua
/
huaapp
Sleeping

App Files Community
huaapp
File size: 4,170 Bytes 
ef3d2c0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
 
# from langchain.agents import load_tools
import requests
from PIL import Image
from langchain.agents import initialize_agent
from langchain.chains.conversation.memory import ConversationBufferWindowMemory
# from langchain.agents import AgentType
# from langchain.llms import OpenAI
# from langchain.chat_models import ChatOpenAI
from langchain.chat_models import AzureChatOpenAI
from langchain.tools import BaseTool
# from transformers.models.oneformer.modeling_oneformer import OneFormerModelOutput
import os
from transformers import BlipProcessor, BlipForConditionalGeneration
from langchain import PromptTemplate, FewShotPromptTemplate
from langchain.chains import LLMChain

OPENAI_API_KEY = os.environ['OPENAI_API_KEY']
OPENAI_API_BASE = os.environ['OPENAI_API_BASE']
DEPLOYMENT_NAME = os.environ['DEPLOYMENT_NAME']

llm = AzureChatOpenAI(deployment_name=DEPLOYMENT_NAME, openai_api_base=OPENAI_API_BASE,
                      openai_api_key=OPENAI_API_KEY, openai_api_version="2023-03-15-preview",
                      model_name="gpt-3.5-turbo")

# OPENAI_API_KEY = os.environ['OPENAI_API_KEY'] or 'Your OPENAI API Key'
# OPENAI_API_KEY = "123"

# llm = ChatOpenAI(openai_api_key=OPENAI_API_KEY, temperature=0, model_name='gpt-3.5-turbo')


image_to_text_model = "Salesforce/blip-image-captioning-large"
# device = 'cuda' if torch.cuda.is_available() else 'cpu'
device = 'cpu'

processor = BlipProcessor.from_pretrained(image_to_text_model)
model = BlipForConditionalGeneration.from_pretrained(image_to_text_model).to(device)


def describeImage3(url):
    image_object = Image.open(requests.get(url, stream=True).raw).convert('RGB')
    # image
    inputs = processor(image_object, return_tensors="pt").to(device)
    outputs = model.generate(**inputs)
    return processor.decode(outputs[0], skip_special_tokens=True)


def describeImage(image_url):
    image_obj = Image.open(image_url).convert('RGB')
    inputs = processor(image_obj, return_tensors='pt').to(device)
    outputs = model.generate(**inputs)
    return processor.decode(outputs[0], skip_special_tokens=True)


def describeImage2(image_object):
    # image_object = Image.open(requests.get(url, stream=True).raw).convert('RGB')
    # image
    inputs = processor(image_object, return_tensors="pt").to(device)
    outputs = model.generate(**inputs)
    return processor.decode(outputs[0], skip_special_tokens=True)


def toChinese(en: str):
    pp = "将下面的语句翻译成中文\n{en}"
    prompt = PromptTemplate(
        input_variables=["en"],
        template=pp
    )
    llchain = LLMChain(llm=llm, prompt=prompt)
    return llchain.run(en)


# description = describeImage('https://images.unsplash.com/photo-1673207520321-c27d09eb0955?ixlib=rb-4.0.3&ixid=MnwxMjA3fDB8MHxwaG90by1wYWdlfHx8fGVufDB8fHx8&auto=format&fit=crop&w=1035&q=80')
# description = describeImage('https://alifei03.cfp.cn/creative/vcg/800/new/VCG21gic13601846.jpg')


# description


class DescribeImageTool(BaseTool):
    name = "Describe Image Tool"
    description = 'use this tool to describe an image.'

    def _run(self, url: str):
        description = describeImage(url)
        return description

    def _arun(self, query: str):
        raise NotImplementedError("Async operation not supported yet")


tools = [DescribeImageTool()]

agent = initialize_agent(
    agent='chat-conversational-react-description',
    tools=tools,
    llm=llm,
    verbose=True,
    max_iterations=3,
    early_stopping_method='generate',
    memory=ConversationBufferWindowMemory(
        memory_key='chat_history',
        k=5,
        return_messages=True
    )
)

# image_url = 'https://images.unsplash.com/photo-1673207520321-c27d09eb0955?ixlib=rb-4.0.3&ixid=MnwxMjA3fDB8MHxwaG90by1wYWdlfHx8fGVufDB8fHx8&auto=format&fit=crop&w=1035&q=80'
# image_url = 'https://alifei03.cfp.cn/creative/vcg/800/new/VCG21gic13601846.jpg'
# agent(f"Describe the following image:\n{image_url}")


# agent(f"What is the brand of car in the following image:\n{image_url}")

# image_url = 'https://alifei03.cfp.cn/creative/vcg/800/new/VCG21gic13601846.jpg'
# agent(f"Please describe the following image:\n{image_url}")


# agent.memory.buffer