app.py

import gradio as gr
from huggingface_hub import InferenceClient

"""
For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
"""
# client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")


from google.cloud import storage
from google.oauth2 import service_account
import json

# upload image to google cloud storage
def upload_image_to_gcs_blob(image):

    google_creds = os.environ.get("GOOGLE_APPLICATION_CREDENTIALS_JSON")

    creds_json = json.loads(google_creds)
    credentials = service_account.Credentials.from_service_account_info(creds_json)

    # 现在您可以使用这些凭证对Google Cloud服务进行认证
    storage_client = storage.Client(credentials=credentials, project=creds_json['project_id'])

    bucket_name=os.environ.get('bucket_name')
    bucket = storage_client.bucket(bucket_name)
    
    destination_blob_name = os.path.basename(image)
    blob = bucket.blob(destination_blob_name)

    blob.upload_from_filename(image)

    public_url = blob.public_url
    
    return public_url



# def respond(
#     message,
#     history: list[tuple[str, str]],
#     system_message,
#     max_tokens,
#     temperature,
#     top_p,
# ):
#     messages = [{"role": "system", "content": system_message}]

#     for val in history:
#         if val[0]:
#             messages.append({"role": "user", "content": val[0]})
#         if val[1]:
#             messages.append({"role": "assistant", "content": val[1]})

#     messages.append({"role": "user", "content": message})

#     response = ""

#     for message in client.chat_completion(
#         messages,
#         max_tokens=max_tokens,
#         stream=True,
#         temperature=temperature,
#         top_p=top_p,
#     ):
#         token = message.choices[0].delta.content

#         response += token
#         yield response

def get_completion(message,history,system_message,max_tokens,temperature):
    # base64_image = encode_image(image)
    if message["text"].strip() == "" and not message["files"]:
        gr.Error("Please input a query and optionally image(s).")
    
    if message["text"].strip() == "" and message["files"]:
        gr.Error("Please input a text query along the image(s).")
    
    text = message['text']
    content = [
        {"type": "text", "text": text},
    ]
    if message['files']:
        image = message['files'][0]
        image_url = upload_image_to_gcs_blob(image)
        content_image = {
            "type": "image_url",
            "image_url": {
                "url": image_url,
            },}
        content.append(content_image)

    init_message = [{"role": "system", "content": system_message}]
    
    history_openai_format = []
    for human, assistant in history:
        history_openai_format.append({"role": "user", "content": human })
        history_openai_format.append({"role": "assistant", "content":assistant})
    history_openai_format.append({"role": "user", "content": content})
    
    
    # 请求头部信息
    openai_api_key = os.environ.get('openai_api_key')
    headers = {
      'Authorization': f'Bearer {openai_api_key}'
    }
    
    # 请求体信息
    data = {
        'model': 'gpt-4o',  # 可以根据需要更换其他模型
        'messages': init_message + history_openai_format[-5:], #system message + 最近的2次對話 + 最新一條消息
        'temperature': temperature,  # 可以根据需要调整
        'max_tokens':max_tokens,
        # 'stream':True，
    }
    
    response = requests.post('https://burn.hair/v1/chat/completions', headers=headers, json=data)
    
    # 解析响应内容
    response_data = response.json()
    response_content = response_data['choices'][0]['message']['content']
    usage = response_data['usage']
    
    return response_content




"""
For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
"""
demo = gr.ChatInterface(
    get_completion,
    multimodal=True,
    additional_inputs=[
        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
    ],
)


if __name__ == "__main__":
    demo.launch()