Skip to content

sample.md

Latest commit

 

History

History
149 lines (119 loc) · 4.19 KB

sample.md

File metadata and controls

149 lines (119 loc) · 4.19 KB
Raw

Perspective API documentation > Get Started > Sample requests

Sample requests

Make an AnalyzeComment request

Run one of the sample API calls below to get scores directly from Perspective API models. You don’t need to build a model locally.

The AnalyzeComment command issues an API request to analyze the comment.text field for the requestedAttributes, in this case the TOXICITY model. Use the API key you generated above as the key parameter.

Leverage the DoNotStore flag to ensure that all submitted comments are automatically deleted after scores are returned and/or the SuggestCommentScore method to submit corrections to improve Perspective over time.

Read the API reference documentation for details on all of the request and response fields, as well as the available values for requestedAttributes. There are experimental attributes, such as "obscene", "attack on a commenter", "spam", etc., that you may also use.

Using cURL

Make an AnalyzeComment request with cURL. The following command should work for most Mac and Linux users. You may need to install cURL to run this command.

Replace YOUR_KEY_HERE with your API key.

$ curl -H "Content-Type: application/json" --data \
    '{comment: {text: "what kind of idiot name is foo?"},
       languages: ["en"],
       requestedAttributes: {TOXICITY:{}} }' \
    https://commentanalyzer.googleapis.com/v1alpha1/comments:analyze?key=YOUR_KEY_HERE

In the following response, the field attributeScores.TOXICITY.summaryScore.value gives the toxicity model's score for the comment. The comment received a score of 0.9 out of 1.0.

{
   "attributeScores": {
      "TOXICITY": {
         "summaryScore": {
            "value": 0.9208521,
            "type": "PROBABILITY"
         }
      }
   },
   "languages": [
      "en"
   ]
}

Learn more about model attribute scores the "Key concepts" section of our API reference.

Using Python

Make an AnalyzeComment request with Python.

import json 
import requests 
api_key = 'YOUR_KEY_HERE'
url = ('https://commentanalyzer.googleapis.com/v1alpha1/comments:analyze' +    
    '?key=' + api_key)
data_dict = {
    'comment': {'text': 'what kind of idiot name is foo?'},
    'languages': ['en'],
    'requestedAttributes': {'TOXICITY': {}}
}
response = requests.post(url=url, data=json.dumps(data_dict)) 
response_dict = json.loads(response.content) 
print(json.dumps(response_dict, indent=2))

You should see output similar to this:

{
   "attributeScores": {
      "TOXICITY": {
         "summaryScore": {
            "value": 0.9208521,
            "type": "PROBABILITY"
         }
      }
   },
   "languages": [
      "en"
   ]
}

Using Node.js

To use Node.js, you must have installed npm and the googleapis client:

npm install googleapis

Make an AnalyzeComment with Node.js:

var googleapis = require('googleapis');

API_KEY = 'YOUR_KEY_HERE'
DISCOVERY_URL = 'https://commentanalyzer.googleapis.com/$discovery/rest?version=v1alpha1'

googleapis.discoverAPI(DISCOVERY_URL, (err, client) => {
  if (err) throw err;
  var analyzeRequest = {
    comment: {text: 'what kind of idiot name is foo?'},
    requestedAttributes: {'TOXICITY': {}}
  };
  client.comments.analyze({key: API_KEY, resource: analyzeRequest}, (err, response) => {
    if (err) throw err;
    console.log(JSON.stringify(response, null, 2));
  });
});

You should see output similar to this:

{
  "attributeScores": {
    "TOXICITY": {
      "spanScores": [
        {
          "score": {
            "value": 0.4445836,
            "type": "PROBABILITY"
          }
        }
      ],
      "summaryScore": {
        "value": 0.4445836,
        "type": "PROBABILITY"
      }
    }
  },
  "languages": [
    "en"
  ]
}

Get started