微调教程

<ph type="x-smartling-placeholder"></ph>

本教程将帮助你开始进行 Gemini API 调参 使用 Python SDK 或 REST API(使用 curl。这些示例展示了如何调优背后的文本模型, Gemini API 文本生成服务。

前往 ai.google.dev 查看 试用 Colab 笔记本 在 GitHub 上查看笔记本

设置身份验证

借助 Gemini API,您可以使用自己的数据对模型进行调参。由于是您的数据 经过调优的模型会使用比 API 密钥更严格的访问权限控制。

在运行本教程之前,您需要为自己的 项目,然后下载 “OAuth 客户端 ID”为“client_secret.json”。

此 gcloud 命令可将 client_secret.json 文件转换为 可用于向该服务进行身份验证。

gcloud auth application-default login \
    --client-id-file client_secret.json \
    --scopes='https://www.googleapis.com/auth/cloud-platform,https://www.googleapis.com/auth/generative-language.tuning'

设置变量

CURL

为重复值设置变量,以用于 REST API 的其余部分 调用。代码使用 Python os 库设置环境 变量。

这特定于 Colab 笔记本环境。接下来的代码 代码单元相当于在 bash 中运行以下命令 终端。

export access_token=$(gcloud auth application-default print-access-token)
export project_id=my-project-id
export base_url=https://generativelanguage.googleapis.com
import os

access_token = !gcloud auth application-default print-access-token
access_token = '\n'.join(access_token)

os.environ['access_token'] = access_token
os.environ['project_id'] = "[Enter your project-id here]"
os.environ['base_url'] = "https://generativelanguage.googleapis.com"

Python

access_token = !gcloud auth application-default print-access-token
access_token = '\n'.join(access_token)

project = '[Enter your project-id here]'
base_url = "https://generativelanguage.googleapis.com"

导入 requests 库。

import requests
import json

列出经调参的模型

列出可用的已调参模型,验证您的身份验证设置。

CURL


curl -X GET ${base_url}/v1beta/tunedModels \
    -H "Content-Type: application/json" \
    -H "Authorization: Bearer ${access_token}" \
    -H "x-goog-user-project: ${project_id}"

Python

headers={
  'Authorization': 'Bearer ' + access_token,
  'Content-Type': 'application/json',
  'x-goog-user-project': project
}

result = requests.get(
  url=f'{base_url}/v1beta/tunedModels',
  headers = headers,
)
result.json()

创建已调参模型

如需创建经调整的模型,您需要将数据集传递给 training_data 字段。

在此示例中,您将对模型进行调参,以生成 序列。例如,如果输入为 1,则模型应输出 2。如果 输入为 one hundred,输出应为 one hundred one

CURL


curl -X POST $base_url/v1beta/tunedModels \
    -H 'Content-Type: application/json' \
    -H "Authorization: Bearer ${access_token}" \
    -H "x-goog-user-project: ${project_id}" \
    -d '
      {
        "display_name": "number generator model",
        "base_model": "models/gemini-1.0-pro-001",
        "tuning_task": {
          "hyperparameters": {
            "batch_size": 2,
            "learning_rate": 0.001,
            "epoch_count":5,
          },
          "training_data": {
            "examples": {
              "examples": [
                {
                    "text_input": "1",
                    "output": "2",
                },{
                    "text_input": "3",
                    "output": "4",
                },{
                    "text_input": "-3",
                    "output": "-2",
                },{
                    "text_input": "twenty two",
                    "output": "twenty three",
                },{
                    "text_input": "two hundred",
                    "output": "two hundred one",
                },{
                    "text_input": "ninety nine",
                    "output": "one hundred",
                },{
                    "text_input": "8",
                    "output": "9",
                },{
                    "text_input": "-98",
                    "output": "-97",
                },{
                    "text_input": "1,000",
                    "output": "1,001",
                },{
                    "text_input": "10,100,000",
                    "output": "10,100,001",
                },{
                    "text_input": "thirteen",
                    "output": "fourteen",
                },{
                    "text_input": "eighty",
                    "output": "eighty one",
                },{
                    "text_input": "one",
                    "output": "two",
                },{
                    "text_input": "three",
                    "output": "four",
                },{
                    "text_input": "seven",
                    "output": "eight",
                }
              ]
            }
          }
        }
      }' | tee tunemodel.json
{
"name": "tunedModels/number-generator-model-dzlmi0gswwqb/operations/bvl8dymw0fhw",
"metadata": {
  "@type": "type.googleapis.com/google.ai.generativelanguage.v1beta.CreateTunedModelMetadata",
  "totalSteps": 38,
  "tunedModel": "tunedModels/number-generator-model-dzlmi0gswwqb"
}
}
% Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                              Dload  Upload   Total   Spent    Left  Speed
100  2280    0   296  100  1984    611   4098 --:--:-- --:--:-- --:--:--  4720

Python

operation = requests.post(
    url = f'{base_url}/v1beta/tunedModels',
    headers=headers,
    json= {
        "display_name": "number generator",
        "base_model": "models/gemini-1.0-pro-001",
        "tuning_task": {
          "hyperparameters": {
            "batch_size": 4,
            "learning_rate": 0.001,
            "epoch_count":5,
          },
          "training_data": {
            "examples": {
              "examples": [
                {
                    'text_input': '1',
                    'output': '2',
                },{
                    'text_input': '3',
                    'output': '4',
                },{
                    'text_input': '-3',
                    'output': '-2',
                },{
                    'text_input': 'twenty two',
                    'output': 'twenty three',
                },{
                    'text_input': 'two hundred',
                    'output': 'two hundred one',
                },{
                    'text_input': 'ninety nine',
                    'output': 'one hundred',
                },{
                    'text_input': '8',
                    'output': '9',
                },{
                    'text_input': '-98',
                    'output': '-97',
                },{
                    'text_input': '1,000',
                    'output': '1,001',
                },{
                    'text_input': '10,100,000',
                    'output': '10,100,001',
                },{
                    'text_input': 'thirteen',
                    'output': 'fourteen',
                },{
                    'text_input': 'eighty',
                    'output': 'eighty one',
                },{
                    'text_input': 'one',
                    'output': 'two',
                },{
                    'text_input': 'three',
                    'output': 'four',
                },{
                    'text_input': 'seven',
                    'output': 'eight',
                }
              ]
            }
          }
        }
      }
)
operation
<Response [200]>
operation.json()
{'name': 'tunedModels/number-generator-wl1qr34x2py/operations/41vni3zk0a47',
'metadata': {'@type': 'type.googleapis.com/google.ai.generativelanguage.v1beta.CreateTunedModelMetadata',
  'totalSteps': 19,
  'tunedModel': 'tunedModels/number-generator-wl1qr34x2py'} }

使用经调整的模型的名称设置一个变量,以用于 调用。

name=operation.json()["metadata"]["tunedModel"]
name
'tunedModels/number-generator-wl1qr34x2py'

周期数、批次大小和学习速率的最佳值取决于 以及应用场景的其他限制条件。要详细了解 这些值,请参阅 高级调参设置超参数

获取经调参的模型状态

模型的状态在训练期间设置为 CREATING,并更改为 完成后ACTIVE

CURL

下面是一些 Python 代码,用于从 响应 JSON。如果是在终端中运行它,可以尝试使用 bash 用于解析响应的 JSON 解析器。

import json

first_page = json.load(open('tunemodel.json'))
os.environ['modelname'] = first_page['metadata']['tunedModel']

print(os.environ['modelname'])
tunedModels/number-generator-model-dzlmi0gswwqb

使用模型名称执行另一个 GET 请求,以获取模型元数据, 包含状态字段


curl -X GET ${base_url}/v1beta/${modelname} \
    -H 'Content-Type: application/json' \
    -H "Authorization: Bearer ${access_token}" \
    -H "x-goog-user-project: ${project_id}" | grep state
"state": "ACTIVE",
% Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                              Dload  Upload   Total   Spent    Left  Speed
100  5921    0  5921    0     0  13164      0 --:--:-- --:--:-- --:--:-- 13157

Python

tuned_model = requests.get(
    url = f'{base_url}/v1beta/{name}',
    headers=headers,
)
tuned_model.json()

以下代码每 5 秒检查一次状态字段,直到它不再 处于 CREATING 状态。

import time
import pprint

op_json = operation.json()
response = op_json.get('response')
error = op_json.get('error')

while response is None and error is None:
    time.sleep(5)

    operation = requests.get(
        url = f'{base_url}/v1/{op_json["name"]}',
        headers=headers,
    )

    op_json = operation.json()
    response = op_json.get('response')
    error = op_json.get('error')

    percent = op_json['metadata'].get('completedPercent')
    if percent is not None:
      print(f"{percent:.2f}% - {op_json['metadata']['snapshots'][-1]}")
      print()

if error is not None:
    raise Exception(error)
100.00% - {'step': 19, 'epoch': 5, 'meanLoss': 1.402067, 'computeTime': '2024-03-14T15:11:23.766989274Z'}

运行推断

调优作业完成后,您可以使用它来生成包含 服务。

CURL

尝试输入罗马数字,例如 63 (LXIII):


curl -X POST $base_url/v1beta/$modelname:generateContent \
    -H 'Content-Type: application/json' \
    -H "Authorization: Bearer ${access_token}" \
    -H "x-goog-user-project: ${project_id}" \
    -d '{
        "contents": [{
        "parts": [{
          "text": "LXIII"
          }]
        }]
        }' 2> /dev/null
{
"candidates": [
  {
    "content": {
      "parts": [
        {
          "text": "LXIV"
        }
      ],
      "role": "model"
    },
    "finishReason": "STOP",
    "index": 0,
    "safetyRatings": [
      {
        "category": "HARM_CATEGORY_SEXUALLY_EXPLICIT",
        "probability": "NEGLIGIBLE"
      },
      {
        "category": "HARM_CATEGORY_HATE_SPEECH",
        "probability": "NEGLIGIBLE"
      },
      {
        "category": "HARM_CATEGORY_HARASSMENT",
        "probability": "NEGLIGIBLE"
      },
      {
        "category": "HARM_CATEGORY_DANGEROUS_CONTENT",
        "probability": "NEGLIGIBLE"
      }
    ]
  }
],
"promptFeedback": {
  "safetyRatings": [
    {
      "category": "HARM_CATEGORY_SEXUALLY_EXPLICIT",
      "probability": "NEGLIGIBLE"
    },
    {
      "category": "HARM_CATEGORY_HATE_SPEECH",
      "probability": "NEGLIGIBLE"
    },
    {
      "category": "HARM_CATEGORY_HARASSMENT",
      "probability": "NEGLIGIBLE"
    },
    {
      "category": "HARM_CATEGORY_DANGEROUS_CONTENT",
      "probability": "NEGLIGIBLE"
    }
  ]
}
}

模型的输出不一定正确。如果经过调参的模型 未达到您所要求的标准,您可以尝试添加更多的 例如,调整超参数或 示例。您甚至可以根据自己创建的第一个模型创建另一个经过调参的模型 创建。

请参阅调参指南 ,进一步了解如何提升广告效果。

Python

尝试输入日语数字,例如 6 (6):

import time

m = requests.post(
    url = f'{base_url}/v1beta/{name}:generateContent',
    headers=headers,
    json= {
        "contents": [{
            "parts": [{
                "text": "六"
            }]
          }]
    })
import pprint
pprint.pprint(m.json())
{'candidates': [{'content': {'parts': [{'text': '七'}], 'role': 'model'},
                'finishReason': 'STOP',
                'index': 0,
                'safetyRatings': [{'category': 'HARM_CATEGORY_SEXUALLY_EXPLICIT',
                                    'probability': 'NEGLIGIBLE'},
                                  {'category': 'HARM_CATEGORY_HATE_SPEECH',
                                    'probability': 'NEGLIGIBLE'},
                                  {'category': 'HARM_CATEGORY_HARASSMENT',
                                    'probability': 'LOW'},
                                  {'category': 'HARM_CATEGORY_DANGEROUS_CONTENT',
                                    'probability': 'NEGLIGIBLE'}]}],
'promptFeedback': {'safetyRatings': [{'category': 'HARM_CATEGORY_SEXUALLY_EXPLICIT',
                                      'probability': 'NEGLIGIBLE'},
                                      {'category': 'HARM_CATEGORY_HATE_SPEECH',
                                      'probability': 'NEGLIGIBLE'},
                                      {'category': 'HARM_CATEGORY_HARASSMENT',
                                      'probability': 'NEGLIGIBLE'},
                                      {'category': 'HARM_CATEGORY_DANGEROUS_CONTENT',
                                      'probability': 'NEGLIGIBLE'}]} }

模型的输出不一定正确。如果经过调参的模型 未达到您所要求的标准,您可以尝试添加更多的 例如,调整超参数或向文本中 示例。

总结

尽管训练数据不包含任何对罗马数字或日语的引用 因此模型在微调后便能够很好地进行泛化。这样,您就可以 可以根据您的使用场景微调模型。

后续步骤

了解如何在 Python SDK 的帮助下使用调整服务进行 Gemini API,请参阅调参快速入门 Python。了解 如需使用 Gemini API 中的其他服务,请访问 REST 使用入门 教程