Anleitung zur Optimierung

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

Dieses Tutorial hilft Ihnen beim Einstieg in die Gemini API-Abstimmung mit dem Python SDK oder der REST API unter Verwendung von curl Die Beispiele zeigen, wie das Textmodell dahinter den Textgenerierungsdienst der Gemini API.

Auf ai.google.dev ansehen In Google Colab ausführen Quelle auf GitHub ansehen

Authentifizierung einrichten

Mit der Gemini API können Sie Modelle anhand Ihrer eigenen Daten abstimmen. Da es sich um Ihre Daten und für Ihre abgestimmten Modelle benötigen, als dies mit API-Schlüsseln möglich ist.

Bevor Sie diese Anleitung ausführen können, müssen Sie OAuth für Ihr Projekt.

In Colab ist die Einrichtung am einfachsten, wenn Sie den Inhalt Ihrer client_secret.json-Datei in den „Secrets Manager“ von Colab (unter dem Schlüsselsymbol in im linken Bereich) mit dem Secret-Namen CLIENT_SECRET.

Dieser gcloud-Befehl wandelt die Datei client_secret.json in Anmeldedaten um, die kann zur Authentifizierung beim Dienst verwendet werden.

<ph type="x-smartling-placeholder">
</ph>
try:
  from google.colab import userdata
  import pathlib
  pathlib.Path('client_secret.json').write_text(userdata.get('CLIENT_SECRET'))

  # Use `--no-browser` in colab
  !gcloud auth application-default login --no-browser --client-id-file client_secret.json --scopes='https://www.googleapis.com/auth/cloud-platform,https://www.googleapis.com/auth/generative-language.tuning'
except ImportError:
  !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'
You are authorizing client libraries without access to a web browser. Please run the following command on a machine with a web browser and copy its output back here. Make sure the installed gcloud version is 372.0.0 or newer.

gcloud auth application-default login --remote-bootstrap="https://accounts.google.com/o/oauth2/auth?response_type=code&client_id=87071151422-n1a3cb6c7fvkfg4gmhdtmn5ulol2l4be.apps.googleusercontent.com&scope=https%3A%2F%2Fwww.googleapis.com%2Fauth%2Fcloud-platform+https%3A%2F%2Fwww.googleapis.com%2Fauth%2Fgenerative-language.tuning&state=QIyNibWSaTIsozjmvZEkVBo6EcoW0G&access_type=offline&code_challenge=76c1ZiGvKN8cvlYfj3BmbCwE4e7tvrlwaX3REUX25gY&code_challenge_method=S256&token_usage=remote"

Enter the output of the above command: https://localhost:8085/?state=QIyNibWSaTIsozjmvZEkVBo6EcoW0G&code=4/0AeaYSHBKrY911S466QjKQIFODoOPXlO1mWyTYYdrbELIDV6Hw2DKRAyro62BugroSvIWsA&scope=https://www.googleapis.com/auth/cloud-platform%20https://www.googleapis.com/auth/generative-language.tuning

Credentials saved to file: [/content/.config/application_default_credentials.json]

These credentials will be used by any library that requests Application Default Credentials (ADC).

Variablen festlegen

CURL

Variablen für wiederkehrende Werte zur Verwendung für den Rest der REST API festlegen Anrufe. Der Code verwendet die Python-Bibliothek os, um die Umgebung festzulegen Variablen, auf die in allen Codezellen zugegriffen werden kann.

Dies gilt speziell für die Colab-Notebookumgebung. Der Code in der nächsten der Codezelle entspricht dem Ausführen der folgenden Befehle in einer Bash .

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"

Importieren Sie die requests-Bibliothek.

import requests
import json

Abgestimmte Modelle auflisten

Überprüfen Sie Ihre Authentifizierungseinrichtung, indem Sie die verfügbaren abgestimmten Modelle auflisten.

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()

Abgestimmtes Modell erstellen

Um ein abgestimmtes Modell zu erstellen, müssen Sie Ihr Dataset an das Modell in der training_data.

In diesem Beispiel stimmen Sie ein Modell ab, um die nächste Zahl in der Sequenz hinzufügen. Wenn die Eingabe beispielsweise 1 ist, sollte das Modell 2 ausgeben. Wenn die Eingabe one hundred ist, sollte die Ausgabe one hundred one sein.

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'} }

Legen Sie eine Variable mit dem Namen Ihres abgestimmten Modells fest, die für den Rest des Anrufe.

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

Abgestimmten Modellstatus abrufen

Der Status des Modells wird während des Trainings auf CREATING gesetzt und ändert sich in ACTIVE.

CURL

Unten sehen Sie Python-Code zum Parsen des generierten Modellnamens aus der JSON-Antwort. Wenn du dies in einem Terminal ausführst, kannst du es mit einer Bash versuchen JSON-Parser zum Parsen der Antwort.

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

Führen Sie eine weitere GET-Anfrage mit dem Modellnamen aus, um die Modellmetadaten abzurufen, enthält das Feld für den Bundesstaat.


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()

Der folgende Code prüft das Statusfeld alle 5 Sekunden, bis es nicht mehr angezeigt wird hat den Status CREATING.

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'}

Inferenz ausführen

Sobald der Abstimmungsjob abgeschlossen ist, können Sie ihn verwenden, um Text mit dem Text zu generieren .

CURL

Versuchen Sie, eine römische Zahl einzugeben, z. B. 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"
    }
  ]
}
}

Die Ausgabe Ihres Modells kann korrekt sein oder nicht. Wenn das abgestimmte Modell nicht den erforderlichen Standards entspricht, können Sie Qualitätsbeispiele, das Anpassen der Hyperparameter oder das Hinzufügen einer Präambel zu Ihrem Beispiele. Sie können sogar ein weiteres abgestimmtes Modell erstellen, das auf dem ersten Modell basiert, erstellt.

Weitere Informationen finden Sie im Leitfaden zur Abstimmung. .

Python

Versuchen Sie, eine japanische Zahl einzugeben, z. B. 6 (UDP):

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'}]} }

Die Ausgabe Ihres Modells kann korrekt sein oder nicht. Wenn das abgestimmte Modell nicht den erforderlichen Standards entspricht, können Sie Qualitätsbeispiele, das Anpassen der Hyperparameter oder das Hinzufügen einer Präambel zu Ihrem Beispiele.

Fazit

Auch wenn die Trainingsdaten keine Verweise auf römische oder japanische Sprache enthielten, Zahlen, konnte das Modell nach der Feinabstimmung lange verallgemeinern. Auf diese Weise können Sie Modelle an Ihre Anwendungsfälle anpassen.

Nächste Schritte

Informationen zum Verwenden des Abstimmungsdienstes mithilfe des Python SDK für den Gemini API erhalten Sie in der Kurzanleitung zur Feinabstimmung mit Python. Weitere Informationen Informationen zur Verwendung anderer Dienste in der Gemini API finden Sie unter REST – Erste Schritte )