操作案例.ipynb · AI-Labs/ai-labs-mllm-service

加入 Gitee

与超过 1200万开发者一起发现、参与优秀开源项目，私有仓库也完全免费：）

文件

克隆/下载

操作案例.ipynb 24.72 KB

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 基础命令"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 启动服务"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "!chmod 600 /root/.ssh/id_rsa"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "!service ssh start"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "!nohup python server.py &"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "!ifconfig"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 查看显卡使用情况"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "!ixsmi"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 公共案例"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import requests, json\n",
    "\n",
    "headers = {\n",
    "    'Connection': 'keep-alive',\n",
    "    'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/109.0.0.0 Safari/537.36'\n",
    "}\n",
    "response = requests.post(url=f\"http://localhost:8000/upload\", headers=headers, files={'file': (\"person.png\", open(f\"statics/image/person.png\", 'rb'))})\n",
    "\n",
    "response.json()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import requests, json\n",
    "\n",
    "headers = {\n",
    "    'Connection': 'keep-alive',\n",
    "    'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/109.0.0.0 Safari/537.36'\n",
    "}\n",
    "response = requests.post(url=f\"http://localhost:8000/tools/query_express\", headers=headers, data={'number': 'YT8980437173209'})\n",
    "\n",
    "response.json()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 对话案例"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import requests, json\n",
    "\n",
    "requests.get(url=\"http://localhost:8000/v1/models\").json()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from openai import OpenAI\n",
    "from IPython import display\n",
    "\n",
    "base_url = \"http://localhost:8000/v1/\"\n",
    "client = OpenAI(api_key=\"EMPTY\", base_url=base_url)\n",
    "\n",
    "def simple_chat(messages, use_stream=False):\n",
    "    return client.chat.completions.create(\n",
    "        model=\"glm-4\",\n",
    "        messages=messages,\n",
    "        stream=use_stream,\n",
    "        max_tokens=4096,\n",
    "        temperature=0.7,\n",
    "        presence_penalty=1.2,\n",
    "        top_p=0.8,\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "messages = [{\n",
    "                \"role\": \"system\",\n",
    "                \"content\": \"你是一位优秀的服装商品智能销售专家，你需要推销一件服装，你需要放大商品的优点，激发用户的购买欲望\"\n",
    "            }, {\n",
    "                \"role\": \"system\",\n",
    "                \"content\": \"这幅图片展示了一件棕色的运动衫。这件运动衫的设计独特，具有明显的运动风格。它有一个V领，领口是黑色的，边缘是白色的。袖子是短袖的，袖口是黑色的，边缘是白色的。运动衫的正面有一个大号的数字“95”，数字是白色的，背景是橙色和蓝色的渐变色。数字的下方有一个白色的“DSQUARED2”标志，标志是黑色的。运动衫的袖子上有一个红色的枫叶图案，枫叶的边缘是白色的。运动衫的后背有一个黑色的\"\n",
    "            }, {\n",
    "                \"role\": \"user\",\n",
    "                \"content\": \"这件衣服的价格是123.4元，材质是氨纶，特点是有弹性，商品名称是高弹运动吸汗T恤。请写一段电商平台的种草文案。\"\n",
    "            }]\n",
    "\n",
    "use_stream=True\n",
    "\n",
    "response = simple_chat(messages, use_stream)\n",
    "\n",
    "if use_stream:\n",
    "    answer = \"\"\n",
    "    for chunk in response:\n",
    "        answer += chunk.choices[0].delta.content\n",
    "        display.clear_output(wait=True)\n",
    "        print(answer)\n",
    "else:\n",
    "    print(response.choices[0].message.content)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "messages = [{\n",
    "                \"role\": \"system\",\n",
    "                \"content\": \"你是一位优秀的服装商品智能销售专家，你熟悉电商平台的购物流程，并且你是精通Mermaid语法的专家。请根据用户的问题生成Mermaid格式的流程图，仅输出流程图，注意必须符合Mermaid的`graph LR`语法格式！别说废话。\"\n",
    "            }, {\n",
    "                \"role\": \"user\",\n",
    "                \"content\": \"如果要下单，我该怎么做\"\n",
    "            }]\n",
    "\n",
    "use_stream=True\n",
    "\n",
    "response = simple_chat(messages, use_stream)\n",
    "answer = \"\"\n",
    "if use_stream:\n",
    "    for chunk in response:\n",
    "        answer += chunk.choices[0].delta.content\n",
    "else:\n",
    "    answer = response.choices[0].message.content\n",
    "\n",
    "display.clear_output(wait=True)\n",
    "print(\"graph\" + answer.split(\"graph\")[1].replace(\"```\", \"\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "messages = [{\n",
    "                \"role\": \"user\",\n",
    "                \"content\": \"你是谁\"\n",
    "            }]\n",
    "\n",
    "use_stream=True\n",
    "\n",
    "response = simple_chat(messages, use_stream)\n",
    "\n",
    "if use_stream:\n",
    "    answer = \"\"\n",
    "    for chunk in response:\n",
    "        answer += chunk.choices[0].delta.content\n",
    "        display.clear_output(wait=True)\n",
    "        print(answer)\n",
    "else:\n",
    "    print(response.choices[0].message.content)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "messages = [{\n",
    "                \"role\": \"system\",\n",
    "                \"content\": \"你是一个文生图 Prompt 翻译器，将中文翻译为英文，翻译简洁精准，如果已经是英文就直接原文返回。你只需要直接给出英文答案, 不要废话\"\n",
    "            },\n",
    "            {\n",
    "                \"role\": \"user\",\n",
    "                \"content\": \"小丑是蝙蝠侠电影系列中的反派，他在一面破碎的镜子中看着自己，风格类似克里斯托弗·诺兰，充满动态、黑暗和恐怖的氛围，戏剧性的灯光，细节丰富，充满隐藏细节，超现实主义，壁纸肖像，色彩缤纷的现实主义，对细节的现实关注，高度细节化，色彩缤纷的现实主义，超高清，8K\"\n",
    "            }]\n",
    "\n",
    "use_stream=False\n",
    "\n",
    "response = simple_chat(messages, use_stream)\n",
    "\n",
    "if use_stream:\n",
    "    answer = \"\"\n",
    "    for chunk in response:\n",
    "        answer += chunk.choices[0].delta.content\n",
    "        display.clear_output(wait=True)\n",
    "        print(answer)\n",
    "else:\n",
    "    print(response.choices[0].message.content)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from openai import OpenAI\n",
    "from IPython import display\n",
    "\n",
    "base_url = \"http://localhost:8000/v1/\"\n",
    "client = OpenAI(api_key=\"EMPTY\", base_url=base_url)\n",
    "\n",
    "tools = [{\n",
    "            \"type\": \"function\",\n",
    "            \"function\": {\n",
    "                \"name\": \"query_express\",\n",
    "                \"description\": \"根据快递单号，查询快递信息\",\n",
    "                \"parameters\": {\n",
    "                    \"type\": \"object\",\n",
    "                    \"properties\": {\n",
    "                        \"number\": {\n",
    "                            \"description\": \"快递单号\",\n",
    "                            \"type\": \"string\"\n",
    "                        }\n",
    "                    },\n",
    "                    \"required\": [ \"number\" ]\n",
    "                },\n",
    "            }\n",
    "        }, {\n",
    "            \"type\": \"function\",\n",
    "            \"function\": {\n",
    "                \"name\": \"query_weather\",\n",
    "                \"description\": \"根据城市名称，查询城市的天气信息\",\n",
    "                \"parameters\": {\n",
    "                    \"type\": \"object\",\n",
    "                    \"properties\": {\n",
    "                        \"city\": {\n",
    "                            \"description\": \"城市名称\",\n",
    "                            \"type\": \"string\"\n",
    "                        }\n",
    "                    },\n",
    "                    \"required\": [ \"city\" ]\n",
    "                },\n",
    "            }\n",
    "        }, {\n",
    "            \"type\": \"function\",\n",
    "            \"function\": {\n",
    "                \"name\": \"draw_picture\",\n",
    "                \"description\": \"根据用户的需求，生成一幅画\",\n",
    "                \"parameters\": {\n",
    "                    \"type\": \"object\",\n",
    "                    \"properties\": {\n",
    "                        \"prompt\": {\n",
    "                            \"description\": \"用户的需求\",\n",
    "                            \"type\": \"string\"\n",
    "                        }\n",
    "                    },\n",
    "                    \"required\": [ \"prompt\" ]\n",
    "                },\n",
    "            }\n",
    "        }, {\n",
    "            \"type\": \"function\",\n",
    "            \"function\": {\n",
    "                \"name\": \"product_recommend\",\n",
    "                \"description\": \"根据用户的需求，推荐合适的服装商品\",\n",
    "                \"parameters\": {\n",
    "                    \"type\": \"object\",\n",
    "                    \"properties\": {\n",
    "                        \"requirement\": {\n",
    "                            \"description\": \"用户的需求\",\n",
    "                            \"type\": \"string\"\n",
    "                        }\n",
    "                    },\n",
    "                    \"required\": [ \"requirement\" ]\n",
    "                },\n",
    "            }\n",
    "        }, {\n",
    "            \"type\": \"function\",\n",
    "            \"function\": {\n",
    "                \"name\": \"flow_graph\",\n",
    "                \"description\": \"根据用户的需求，提供做某件事情的流程\",\n",
    "                \"parameters\": {\n",
    "                    \"type\": \"object\",\n",
    "                    \"properties\": {\n",
    "                        \"requirement\": {\n",
    "                            \"description\": \"用户的需求\",\n",
    "                            \"type\": \"string\"\n",
    "                        }\n",
    "                    },\n",
    "                    \"required\": [ \"requirement\" ]\n",
    "                },\n",
    "            }\n",
    "        }]\n",
    "\n",
    "messages = [{\n",
    "                \"role\": \"user\",\n",
    "                \"content\": \"成都的天气怎么样\"\n",
    "            }]\n",
    "\n",
    "use_stream=False\n",
    "\n",
    "response = client.chat.completions.create(\n",
    "        model=\"glm-4\",\n",
    "        messages=messages,\n",
    "        stream=use_stream,\n",
    "        max_tokens=4096,\n",
    "        temperature=0.7,\n",
    "        presence_penalty=1.2,\n",
    "        top_p=0.8,\n",
    "        tools=tools\n",
    "    )\n",
    "\n",
    "if response.choices[0].message.tool_calls is not None:\n",
    "    tool_call = response.choices[0].message.tool_calls[0]\n",
    "    function_name = tool_call.function.name\n",
    "    function_args = tool_call.function.arguments\n",
    "    print(tool_call.function)\n",
    "\n",
    "else:\n",
    "    print(response.choices[0].message.content)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 语音案例"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 语音生成案例"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import requests, json\n",
    "\n",
    "# url=\"http://localhost:8000/tts/chattts/generate_audio_file\"\n",
    "url=\"http://localhost:8000/tts/melotts/generate_audio_file\"\n",
    "\n",
    "headers = {'Content-Type': 'application/json; charset=utf-8', 'accept': 'application/json'}\n",
    "data=json.dumps({'text': '开发 AI 应用从此简单，汇聚最新最热 AI 模型，提供模型体验、推理、训练、部署和应用的一站式服务，提供充沛算力，做中国最好的 AI 社区'})\n",
    "\n",
    "response = requests.post(url=url, headers=headers, data=data)\n",
    "\n",
    "from IPython.display import Audio\n",
    "Audio(response.content, rate=24000)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 语音识别案例"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import requests, json\n",
    "\n",
    "# url=\"http://localhost:8000/asr/whisper/audio_to_text\"\n",
    "url=\"http://localhost:8000/asr/funasr/audio_to_text\"\n",
    "\n",
    "headers = {\n",
    "    'Connection': 'keep-alive',\n",
    "    'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/109.0.0.0 Safari/537.36'\n",
    "}\n",
    "files = {\n",
    "    \"audio\": (\"test.wav\", open(f\"statics/audio/test.wav\", 'rb'))\n",
    "}\n",
    "response = requests.post(url=url, headers=headers, files=files)\n",
    "\n",
    "response.json()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 图片案例"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 图片生成案例"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import requests, json\n",
    "\n",
    "# url=\"http://localhost:8000/image/stable_diffusion_3/generate_image_file\"\n",
    "url=\"http://localhost:8001/image/stable_diffusion_xl/generate_image_file\"\n",
    "\n",
    "headers = {'Content-Type': 'application/json; charset=utf-8'}\n",
    "data=json.dumps({'prompt': 'a dog, 8K'})\n",
    "\n",
    "response = requests.post(url=url, headers=headers, data=data)\n",
    "\n",
    "from IPython.display import Image\n",
    "Image(response.content)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 图片识别案例"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import requests, json\n",
    "\n",
    "headers = {\n",
    "    'Connection': 'keep-alive',\n",
    "    'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/109.0.0.0 Safari/537.36'\n",
    "}\n",
    "files = {\n",
    "    \"image\": (\"cloth.png\", open(f\"statics/image/cloth.png\", 'rb'))\n",
    "}\n",
    "data={\n",
    "    'user_input': \"请详细描述这幅图片，精准捕获图片中服装的每一个细节\"\n",
    "}\n",
    "\n",
    "response = requests.post(url=\"http://localhost:8000/vision_language/qwen2vl/chat_with_image\", headers=headers, data=data, files=files)\n",
    "\n",
    "response.json()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import requests, json\n",
    "\n",
    "headers = {\n",
    "    'Connection': 'keep-alive',\n",
    "    'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/109.0.0.0 Safari/537.36'\n",
    "}\n",
    "files = {\n",
    "    \"media\": (\"test.mp4\", open(f\"statics/video/test.mp4\", 'rb'))\n",
    "}\n",
    "data={\n",
    "    'media_type': \"video\",\n",
    "    'user_input': \"请详细介绍一下这个视频\"\n",
    "}\n",
    "\n",
    "response = requests.post(url=\"http://localhost:8000/vision_language/qwen2vl/chat_with_media\", headers=headers, data=data, files=files)\n",
    "\n",
    "response.json()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import requests, json\n",
    "\n",
    "headers = {\n",
    "    'Connection': 'keep-alive',\n",
    "    'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/109.0.0.0 Safari/537.36'\n",
    "}\n",
    "files = {\n",
    "    \"image\": (\"cloth.png\", open(f\"statics/image/cloth.png\", 'rb'))\n",
    "}\n",
    "data={\n",
    "    'user_input': \"请描述这幅图片\"\n",
    "}\n",
    "\n",
    "response = requests.post(url=\"http://localhost:8000/vision_language/internvl/chat_with_image\", headers=headers, data=data, files=files)\n",
    "\n",
    "response.json()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import requests, json\n",
    "\n",
    "headers = {\n",
    "    'Connection': 'keep-alive',\n",
    "    'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/109.0.0.0 Safari/537.36'\n",
    "}\n",
    "files = {\n",
    "    \"image\": (\"cloth.png\", open(f\"statics/image/cloth.png\", 'rb'))\n",
    "}\n",
    "data={\n",
    "    'product_name': \"短袖\",\n",
    "    'product_tags': \"\",\n",
    "    'product_gender': \"男\",\n",
    "    'product_season': \"夏季\",\n",
    "    'product_price': \"123.4\",\n",
    "    'product_style': \"简约\",\n",
    "    'product_material': \"纯棉\",\n",
    "    'product_advantage': \"\",\n",
    "    'product_description': \"这是一件纯棉T恤，适合夏季穿，清爽！\"\n",
    "}\n",
    "\n",
    "response = requests.post(url=\"http://localhost:8000/vision_language/internvl/marketing_documents\", headers=headers, data=data, files=files)\n",
    "\n",
    "response.json()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 虚拟试穿案例"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import requests, json\n",
    "\n",
    "headers = {\n",
    "    'Connection': 'keep-alive',\n",
    "    'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/109.0.0.0 Safari/537.36'\n",
    "}\n",
    "data = {\n",
    "    \"strength\": 1.0,\n",
    "    \"num_inference_steps\": 30,\n",
    "    \"guidance_scale\": 2.5,\n",
    "    \"seed\": 42,\n",
    "    \"enable_shape_control\": False\n",
    "}\n",
    "files = {\n",
    "    \"person_image\": (\"person.png\", open(f\"statics/image/person.png\", 'rb')),\n",
    "    \"person_mask\": (\"person_mask.png\", open(f\"statics/image/person_mask.png\", 'rb')),\n",
    "    \"cloth_image\": (\"cloth.png\", open(f\"statics/image/cloth.png\", 'rb')),\n",
    "    \"cloth_mask\": (\"cloth_mask.png\", open(f\"statics/image/cloth_mask.png\", 'rb'))\n",
    "}\n",
    "response = requests.post(url=f\"http://localhost:8000/try_on/anydoor/tryon_file\", headers=headers, data=data, files=files)\n",
    "\n",
    "from IPython.display import Image\n",
    "Image(response.content)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import requests, json\n",
    "\n",
    "headers = {\n",
    "    'Connection': 'keep-alive',\n",
    "    'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/109.0.0.0 Safari/537.36'\n",
    "}\n",
    "data = {\n",
    "    \"strength\": 1.0,\n",
    "    \"num_inference_steps\": 30,\n",
    "    \"guidance_scale\": 2.5,\n",
    "    \"seed\": 42,\n",
    "    \"enable_shape_control\": False\n",
    "}\n",
    "files = {\n",
    "    \"person_image\": (\"person.png\", open(f\"statics/image/person.png\", 'rb')),\n",
    "    \"person_mask\": (\"person_mask.png\", open(f\"statics/image/person_mask.png\", 'rb')),\n",
    "    \"cloth_image\": (\"cloth.png\", open(f\"statics/image/cloth.png\", 'rb')),\n",
    "    \"cloth_mask\": (\"cloth_mask.png\", open(f\"statics/image/cloth_mask.png\", 'rb'))\n",
    "}\n",
    "response = requests.post(url=f\"http://localhost:8000/try_on/catvton/tryon_file\", headers=headers, data=data, files=files)\n",
    "\n",
    "from IPython.display import Image\n",
    "Image(response.content)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 数字人案例"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import requests, json\n",
    "\n",
    "headers = {\n",
    "    'Connection': 'keep-alive',\n",
    "    'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/109.0.0.0 Safari/537.36'\n",
    "}\n",
    "data = {\n",
    "    'talker_name': '1'\n",
    "}\n",
    "files = {\n",
    "    \"audio\": (\"test.wav\", open(f\"statics/audio/test.wav\", 'rb'))\n",
    "}\n",
    "response = requests.post(url=\"http://localhost:8000/digital_human/hallo/generate_video_file\", headers=headers, data=data, files=files)\n",
    "\n",
    "with open(\"output.mp4\", \"wb\") as f:\n",
    "    f.write(response.content)\n",
    "\n",
    "from IPython.display import Video\n",
    "Video(\"output.mp4\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import requests, json\n",
    "\n",
    "headers = {\n",
    "    'Connection': 'keep-alive',\n",
    "    'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/109.0.0.0 Safari/537.36'\n",
    "}\n",
    "data = {\n",
    "    'talker_name': 'kara'\n",
    "}\n",
    "files = {\n",
    "    \"audio\": (\"test.wav\", open(f\"statics/audio/test.wav\", 'rb'))\n",
    "}\n",
    "response = requests.post(url=\"http://localhost:8000/digital_human/v_express/generate_video_file\", headers=headers, data=data, files=files)\n",
    "\n",
    "with open(\"output.mp4\", \"wb\") as f:\n",
    "    f.write(response.content)\n",
    "\n",
    "from IPython.display import Video\n",
    "Video(\"output.mp4\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import requests, json\n",
    "\n",
    "headers = {\n",
    "    'Connection': 'keep-alive',\n",
    "    'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/109.0.0.0 Safari/537.36'\n",
    "}\n",
    "files = {\n",
    "    \"audio\": (\"test.wav\", open(f\"statics/audio/test.wav\", 'rb')),\n",
    "    \"face\": (\"ref.jpg\", open(f\"statics/digital_human/v_express/kara/ref.jpg\", 'rb'))\n",
    "}\n",
    "response = requests.post(url=\"http://localhost:8000/digital_human/easy_wav2lip/generate_video_file\", headers=headers, files=files)\n",
    "\n",
    "with open(\"output.mp4\", \"wb\") as f:\n",
    "    f.write(response.content)\n",
    "\n",
    "from IPython.display import Video\n",
    "Video(\"output.mp4\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from IPython.display import Video\n",
    "Video(\"output.mp4\")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.14"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}

一键复制编辑原始数据按行查看历史

提交于 2024-11-16 18:31 . mllm-service

马建仓 AI 助手

尝试更多

代码解读

代码找茬

代码优化

https://gitee.com/AI-Labs/ai-labs-mllm-service.git

git@gitee.com:AI-Labs/ai-labs-mllm-service.git

AI-Labs

ai-labs-mllm-service

master

AI-Labs/ai-labs-mllm-service

简介

发行版

贡献者

近期动态

AI-Labs/ai-labs-mllm-service .gitee-modal { width: 500px !important; }

简介

发行版

贡献者

近期动态

搜索帮助

AI-Labs/ai-labs-mllm-service