如何使用谷歌云 AutoML 创建数据集

Question

Silly question, but I'm new to Python and Google Cloud. I'm trying to predict customer-lifetime-value with google cloud automl_v1beta1, the code is here

 from google.cloud.automl_v1beta1 import AutoMlClient, PredictionServiceClient from google.cloud_v1beta1 import automl client = AutoMlClient.from_service_account_file(keyfile_name) location_path = f"projects/{project_id}/locations/us-central1" dataset_display_name = automl_dataset # create dataset create_dataset_response = client.create_dataset( parent=location_path, dataset={'display_name': dataset_display_name, 'tables_dataset_metadata': {}})

and I received the following errors: \

 Traceback (most recent call last): File "/home/tensorflow-lifetime-value/clv_automl/test1.py", line 220, in <module> main() File "/home/tensorflow-lifetime-value/clv_automl/test1.py", line 193, in main model_name = create_automl_model(client, File "/home/tensorflow-lifetime-value/clv_automl/test1.py", line 48, in create_automl_model create_dataset_response = client.create_dataset( File "/home/.local/lib/python3.9/site-packages/google/cloud/automl_v1/services/auto_ml/client.py", line 623, in create_dataset request.dataset = dataset File "/usr/local/lib/python3.9/dist-packages/proto/message.py", line 757, in __setattr__ pb_value = marshal.to_proto(pb_type, value) File "/usr/local/lib/python3.9/dist-packages/proto/marshal/marshal.py", line 211, in to_proto pb_value = rule.to_proto(value) File "/usr/local/lib/python3.9/dist-packages/proto/marshal/rules/message.py", line 36, in to_proto return self._descriptor(**value) ValueError: Protocol message Dataset has no "tables_dataset_metadata" field.

Answer 1

I got the same error using your code when I used google-cloud-automl<=0.1.2 . Your error will be fixed if you use google-cloud-automl>=0.1.2 . But I suggest to always install the latest version by doing pip install --upgrade google-cloud-automl to maximize the latest updates from Google Cloud.