You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
138 lines
4.8 KiB
138 lines
4.8 KiB
import time
|
|
|
|
from openai import OpenAI, AsyncOpenAI
|
|
|
|
from WxMini.Milvus.Config.MulvusConfig import MODELSCOPE_ACCESS_TOKEN
|
|
from WxMini.Utils.MySQLUtil import save_chat_to_mysql
|
|
|
|
|
|
async def recognize_text(client, pool, person_id, image_url):
|
|
"""
|
|
识别图片中的文字，流式输出，并将结果记录到数据库
|
|
:param client: AsyncOpenAI 客户端
|
|
:param pool: 数据库连接池
|
|
:param person_id: 用户 ID
|
|
:param image_url: 图片 URL
|
|
:return: 最终拼接的字符串
|
|
"""
|
|
completion = await client.chat.completions.create(
|
|
model="qwen-vl-ocr",
|
|
messages=[
|
|
{
|
|
"role": "user",
|
|
"content": [
|
|
{
|
|
"type": "image_url",
|
|
"image_url": image_url,
|
|
"min_pixels": 28 * 28 * 4,
|
|
"max_pixels": 28 * 28 * 1280
|
|
},
|
|
{"type": "text", "text": "Read all the text in the image."},
|
|
]
|
|
}
|
|
],
|
|
stream=True
|
|
)
|
|
|
|
full_text = "" # 用于存储最终拼接的字符串
|
|
async for chunk in completion:
|
|
if chunk.choices[0].delta.content is not None:
|
|
for char in chunk.choices[0].delta.content:
|
|
if char != ' ':
|
|
yield char # 流式输出字符
|
|
full_text += char # 拼接字符
|
|
print(char, end='')
|
|
time.sleep(0.1) # 控制输出速度
|
|
|
|
# 记录到数据库
|
|
try:
|
|
await save_chat_to_mysql(pool, person_id, f'{image_url}', full_text, "", 0, 2, 2, 1)
|
|
except Exception as e:
|
|
print(f"记录到数据库时出错：{e}")
|
|
|
|
|
|
async def recognize_content(client, pool, person_id, image_url):
|
|
"""
|
|
识别图片中的内容，流式输出
|
|
"""
|
|
completion = await client.chat.completions.create(
|
|
model="qwen-vl-plus",
|
|
messages=[{"role": "user", "content": [
|
|
{"type": "text", "text": "这是什么"},
|
|
{"type": "image_url", "image_url": {"url": image_url}}
|
|
]}],
|
|
stream=True
|
|
)
|
|
full_text = "" # 用于存储最终拼接的字符串
|
|
async for chunk in completion:
|
|
if chunk.choices[0].delta.content is not None:
|
|
for char in chunk.choices[0].delta.content:
|
|
if char != ' ':
|
|
yield char # 流式输出字符
|
|
full_text += char # 拼接字符
|
|
print(char, end='')
|
|
time.sleep(0.1) # 控制输出速度
|
|
|
|
# 记录到数据库
|
|
try:
|
|
await save_chat_to_mysql(pool, person_id, f'{image_url}', full_text, "", 0, 2, 2, 2)
|
|
except Exception as e:
|
|
print(f"记录到数据库时出错：{e}")
|
|
|
|
|
|
async def recognize_math(pool, person_id, image_url):
|
|
client = AsyncOpenAI(
|
|
api_key=MODELSCOPE_ACCESS_TOKEN,
|
|
base_url="https://api-inference.modelscope.cn/v1"
|
|
)
|
|
"""
|
|
识别图片中的数学题，流式输出，并将结果记录到数据库
|
|
:param client: AsyncOpenAI 客户端
|
|
:param pool: 数据库连接池
|
|
:param person_id: 用户 ID
|
|
:param image_url: 图片 URL
|
|
:return: 最终拼接的字符串
|
|
"""
|
|
# 提示词
|
|
prompt = "You are a helpful and harmless assistant. You are Qwen developed by Alibaba. You should think step-by-step."
|
|
|
|
completion = await client.chat.completions.create(
|
|
model="Qwen/Qwen2.5-VL-32B-Instruct",
|
|
#model="Qwen/Qwen2.5-VL-72B-Instruct",
|
|
messages=[
|
|
{
|
|
"role": "system",
|
|
"content": [
|
|
{"type": "text", "text": prompt}
|
|
],
|
|
},
|
|
{
|
|
"role": "user",
|
|
"content": [
|
|
{
|
|
"type": "image_url",
|
|
"image_url": {"url": image_url}
|
|
},
|
|
{"type": "text", "text": "请使用中文回答：如何作答?"},
|
|
],
|
|
}
|
|
],
|
|
stream=True
|
|
)
|
|
|
|
full_text = "" # 用于存储最终拼接的字符串
|
|
async for chunk in completion:
|
|
if chunk.choices[0].delta.content is not None:
|
|
for char in chunk.choices[0].delta.content:
|
|
if char != ' ':
|
|
yield char # 流式输出字符
|
|
full_text += char # 拼接字符
|
|
print(char, end='')
|
|
time.sleep(0.1) # 控制输出速度
|
|
|
|
# 记录到数据库
|
|
try:
|
|
await save_chat_to_mysql(pool, person_id, f'{image_url}', full_text, "", 0, 2, 2, 1)
|
|
except Exception as e:
|
|
print(f"记录到数据库时出错：{e}")
|