Question

我制作了一个Tensorflow模型并将文件上传到GCP存储。

我试图从Google AI平台获得预测结果。但是它显示了错误。

我为具有以下角色的模型“ vgg_16”创建了一个serviceAccount。

roles/iam.securityAdmin
roles/iam.securityReviewer
roles/ml.admin
roles/ml.developer
roles/ml.modelOwner
roles/ml.modelUser
roles/ml.viewer
roles/owner
roles/viewer

此外，我的项目已激活计费。

我的代码：

from googleapiclient import discovery
from oauth2client.client import GoogleCredentials
from oauth2client.service_account import ServiceAccountCredentials
import numpy
import base64
import logging
from IPython.display import display, Image

from google.cloud import storage

cropped_image = "8-android.widget.TextView.png"
display(Image(cropped_image))

PROJECT_ID = '6111116488'
MODEL_NAME = 'vgg_16'
VERSION_NAME ='v1'
def explicit():
    # Explicitly use service account credentials by specifying the private key
    # file.
    storage_client = storage.Client.from_service_account_json(
        'D://setup//Development//sfsfc1111.json')

    # Make an authenticated API request
    buckets = list(storage_client.list_buckets())
    print(buckets)

def call_ml_service(img_str):
    parent = 'projects/{}/models/{}/versions/{}'.format(PROJECT_ID, MODEL_NAME,VERSION_NAME)
    pred = None
    request_dict = {
        "instances": [
            {
                "image": {
                    "b64": "%s"%img_str
                }
            }
        ]
    }
    try:
        credentials = ServiceAccountCredentials.from_json_keyfile_name('D://setup//Development//sfsfc1111.json')
        #credentials = GoogleCredentials.get_application_default()
        cloudml_svc = discovery.build('ml', 'v1', credentials=credentials)
        request = cloudml_svc.projects().predict(name=parent, body=request_dict)
        response = request.execute()
        print(response)
        #pred = response['predictions'][0]['scores']
        #pred = numpy.asarray(pred)
    except:
        logging.exception("Something went wrong!")
    return pred

# base64 encode the same image
with open(cropped_image, 'rb') as image_file:
    encoded_string = base64.b64encode(image_file.read())
#print(encoded_string)
explicit()
# See what ML Engine thinks
online_prediction = call_ml_service(encoded_string)
print(online_prediction)

但是403“无法访问模型。”发生了。错误代码：

(tensorflow1.14) D:\work\models\research\slim>python inference_google_aiplatform.py
<IPython.core.display.Image object>
[<Bucket: android-ios-new-vcm>]
ERROR:root:Something went wrong!

Traceback (most recent call last):
  File "inference_google_aiplatform.py", line 48, in call_ml_service
    response = request.execute()
  File "C:\Users\admin\Anaconda3\envs\tensorflow1.14\lib\site-packages\googleapiclient\_helpers.py", line 134, in positional_wrapper
    return wrapped(*args, **kwargs)
  File "C:\Users\admin\Anaconda3\envs\tensorflow1.14\lib\site-packages\googleapiclient\http.py", line 907, in execute
    raise HttpError(resp, content, uri=self.uri)
googleapiclient.errors.HttpError: <HttpError 403 when requesting https://ml.googleapis.com/v1/projects/6111116488/models/vgg_16/versions/v1:predict?alt=json returned "Access to model denied.">
None

有人可以帮助我吗？

返回“拒绝访问模型”。来自Google AI平台

0 个答案: