def create_model(project_id, dataset_id, display_name):
"""Create a model."""
# [START automl_vision_classification_create_model]
from google.cloud import automl
# TODO(developer): Uncomment and set the following variables
# project_id = "YOUR_PROJECT_ID"
# dataset_id = "YOUR_DATASET_ID"
# display_name = "your_models_display_name"
client = automl.AutoMlClient()
# A resource that represents Google Cloud Platform location.
project_location = f"projects/{project_id}/locations/us-central1"
# Leave model unset to use the default base model provided by Google
# train_budget_milli_node_hours: The actual train_cost will be equal or
# less than this value.
# https://cloud.google.com/automl/docs/reference/rpc/google.cloud.automl.v1#imageclassificationmodelmetadata
metadata = automl.ImageClassificationModelMetadata(
train_budget_milli_node_hours=24000)
model = automl.Model(
display_name=display_name,
dataset_id=dataset_id,
image_classification_model_metadata=metadata,
)
# Create a model with the model metadata in the region.
response = client.create_model(parent=project_location, model=model)
print("Training operation name: {}".format(response.operation.name))
print("Training started...")
# [END automl_vision_classification_create_model]
return response
def train_image_classification_model(
self,
display_name: (str, 'the display name for the dataset and model'),
input_paths:
(str,
'the paths to csv files describing the input data for a new dataset'
) = '',
dataset_id: (str, 'the id of an existing dataset to reuse') = '',
classification_type: (automl.ClassificationType,
'MULTICLASS or MULTILABEL'
) = automl.ClassificationType.MULTICLASS,
train_budget_milli_node_hours: int = 24000) -> Operation:
dataset = None
if len(dataset_id) == 0:
# Specify the classification type
# Types:
# MultiLabel: Multiple labels are allowed for one example.
# MultiClass: At most one label is allowed per example.
# https://cloud.google.com/automl/docs/reference/rpc/google.cloud.automl.v1#classificationtype
metadata = automl.ImageClassificationDatasetMetadata(
classification_type=classification_type)
dataset = automl.Dataset(
display_name=display_name,
image_classification_dataset_metadata=metadata,
)
# Leave model unset to use the default base model provided by Google
# train_budget_milli_node_hours: The actual train_cost will be equal or
# less than this value.
# https://cloud.google.com/automl/docs/reference/rpc/google.cloud.automl.v1#imageclassificationmodelmetadata
metadata = automl.ImageClassificationModelMetadata(
train_budget_milli_node_hours=train_budget_milli_node_hours)
model = automl.Model(
display_name=display_name,
dataset_id=dataset_id,
image_classification_model_metadata=metadata,
)
long_running_operation = self.train_automl_model(
model=model,
dataset=dataset,
dataset_id=dataset_id,
input_paths=input_paths)
return long_running_operation
# Import data from the input URI
response = client.import_data(name=dataset_full_id, input_config=input_config)
print("Processing import...")
print("Data imported. {}".format(response.result()))
start_training_time = time.time()
project_location = f"projects/{project_id}/locations/us-central1"
# Leave model unset to use the default base model provided by Google
# train_budget_milli_node_hours: The actual train_cost will be equal or
# less than this value.
# https://cloud.google.com/automl/docs/reference/rpc/google.cloud.automl.v1#imageclassificationmodelmetadata
metadata = automl.ImageClassificationModelMetadata(
train_budget_milli_node_hours=8000
)
model = automl.Model(
display_name=display_name,
dataset_id=dataset_id,
image_classification_model_metadata=metadata,
)
# Create a model with the model metadata in the region.
response = client.create_model(parent=project_location, model=model)
print("Training operation name: {}".format(response.operation.name))
print("Training started...")
model_id = response.result().name.split('/')[-1]