def create_cord19_application_package():
app_package = ApplicationPackage(name="cord19")
app_package.schema.add_fields(
Field(name="id", type="string", indexing=["attribute", "summary"]),
Field(
name="title",
type="string",
indexing=["index", "summary"],
index="enable-bm25",
),
)
app_package.schema.add_field_set(FieldSet(name="default",
fields=["title"]))
app_package.schema.add_rank_profile(
RankProfile(name="bm25", first_phase="bm25(title)"))
bert_config = BertModelConfig(
model_id="pretrained_bert_tiny",
tokenizer="google/bert_uncased_L-2_H-128_A-2",
model="google/bert_uncased_L-2_H-128_A-2",
query_input_size=5,
doc_input_size=10,
)
app_package.add_model_ranking(
model_config=bert_config,
include_model_summary_features=True,
inherits="default",
first_phase="bm25(title)",
second_phase=SecondPhaseRanking(rerank_count=10, expression="logit1"),
)
return app_package
def setUp(self) -> None:
self.model_config = BertModelConfig(
model_id="bert_tiny",
query_input_size=4,
doc_input_size=8,
tokenizer=os.path.join(os.environ["RESOURCES_DIR"],
"bert_tiny_tokenizer"),
)
def setUp(self) -> None:
self.app_package = ApplicationPackage(name="test_app")
self.app_package.schema.add_fields(
Field(name="id", type="string", indexing=["attribute", "summary"]),
Field(
name="title",
type="string",
indexing=["index", "summary"],
index="enable-bm25",
),
Field(
name="body",
type="string",
indexing=["index", "summary"],
index="enable-bm25",
),
)
self.app_package.schema.add_field_set(
FieldSet(name="default", fields=["title", "body"]))
self.app_package.schema.add_rank_profile(
RankProfile(name="default", first_phase="nativeRank(title, body)"))
self.app_package.schema.add_rank_profile(
RankProfile(
name="bm25",
first_phase="bm25(title) + bm25(body)",
inherits="default",
))
self.app_package.query_profile_type.add_fields(
QueryTypeField(
name="ranking.features.query(query_bert)",
type="tensor<float>(x[768])",
))
self.app_package.query_profile.add_fields(
QueryField(name="maxHits", value=100),
QueryField(name="anotherField", value="string_value"),
)
bert_config = BertModelConfig(
model_id="bert_tiny",
query_input_size=4,
doc_input_size=8,
tokenizer=os.path.join(os.environ["RESOURCES_DIR"],
"bert_tiny_tokenizer"),
model=os.path.join(os.environ["RESOURCES_DIR"], "bert_tiny_model"),
)
self.app_package.add_model_ranking(
model_config=bert_config,
include_model_summary_features=True,
inherits="default",
first_phase="bm25(title)",
second_phase=SecondPhaseRanking(rerank_count=10,
expression="logit1"),
)
def setUp(self) -> None:
#
# Create application package
#
self.app_package = ApplicationPackage(name="cord19")
self.app_package.schema.add_fields(
Field(name="cord_uid", type="string", indexing=["attribute", "summary"]),
Field(
name="title",
type="string",
indexing=["index", "summary"],
index="enable-bm25",
),
)
self.app_package.schema.add_field_set(
FieldSet(name="default", fields=["title"])
)
self.app_package.schema.add_rank_profile(
RankProfile(name="bm25", first_phase="bm25(title)")
)
self.bert_config = BertModelConfig(
model_id="pretrained_bert_tiny",
tokenizer="google/bert_uncased_L-2_H-128_A-2",
model="google/bert_uncased_L-2_H-128_A-2",
query_input_size=5,
doc_input_size=10,
)
self.app_package.add_model_ranking(
model_config=self.bert_config,
include_model_summary_features=True,
inherits="default",
first_phase="bm25(title)",
second_phase=SecondPhaseRanking(rerank_count=10, expression="logit1"),
)
#
# Deploy on Vespa Cloud
#
self.vespa_cloud = VespaCloud(
tenant="vespa-team",
application="pyvespa-integration",
key_content=os.getenv("VESPA_CLOUD_USER_KEY").replace(r"\n", "\n"),
application_package=self.app_package,
)
self.disk_folder = os.path.join(os.getenv("WORK_DIR"), "sample_application")
self.instance_name = "test"
self.app = self.vespa_cloud.deploy(
instance=self.instance_name, disk_folder=self.disk_folder
)
def setUp(self) -> None:
#
# Create application package
#
self.app_package = ApplicationPackage(name="cord19")
self.app_package.schema.add_fields(
Field(name="cord_uid",
type="string",
indexing=["attribute", "summary"]),
Field(
name="title",
type="string",
indexing=["index", "summary"],
index="enable-bm25",
),
)
self.app_package.schema.add_field_set(
FieldSet(name="default", fields=["title"]))
self.app_package.schema.add_rank_profile(
RankProfile(name="bm25", first_phase="bm25(title)"))
self.bert_config = BertModelConfig(
model_id="pretrained_bert_tiny",
tokenizer="google/bert_uncased_L-2_H-128_A-2",
model="google/bert_uncased_L-2_H-128_A-2",
query_input_size=5,
doc_input_size=10,
)
self.app_package.add_model_ranking(
model_config=self.bert_config,
include_model_summary_features=True,
inherits="default",
first_phase="bm25(title)",
second_phase=SecondPhaseRanking(rerank_count=10,
expression="logit1"),
)
self.disk_folder = os.path.join(os.getenv("WORK_DIR"),
"sample_application")
self.vespa_docker = VespaDocker(port=8089)
self.app = self.vespa_docker.deploy(
application_package=self.app_package, disk_folder=self.disk_folder)