Skip to content

[PLT-1398] Vb/project dashboard extra data plt 1398 #1772

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 5 commits into from
Aug 21, 2024
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
60 changes: 52 additions & 8 deletions libs/labelbox/src/labelbox/schema/labeling_service_dashboard.py
Original file line number Diff line number Diff line change
Expand Up @@ -7,20 +7,24 @@
from labelbox.pydantic_compat import BaseModel, root_validator, Field
from labelbox.schema.search_filters import SearchFilter, build_search_filter
from labelbox.utils import _CamelCaseMixin
from .ontology_kind import EditorTaskType
from labelbox.schema.media_type import MediaType
from labelbox.schema.labeling_service_status import LabelingServiceStatus
from labelbox.utils import _CamelCaseMixin, sentence_case

GRAPHQL_QUERY_SELECTIONS = """
id
name
# serviceType
# createdAt
# updatedAt
# createdById
boostRequestedAt
boostUpdatedAt
boostRequestedBy
boostStatus
dataRowsCount
dataRowsInReviewCount
dataRowsInReworkCount
dataRowsDoneCount
mediaType
editorTaskType
"""


Expand All @@ -38,16 +42,16 @@ class LabelingServiceDashboard(BaseModel):
"""
id: str = Field(frozen=True)
name: str = Field(frozen=True)
service_type: Optional[str] = Field(frozen=True, default=None)
created_at: Optional[datetime] = Field(frozen=True, default=None)
updated_at: Optional[datetime] = Field(frozen=True, default=None)
created_by_id: Optional[str] = Field(frozen=True, default=None)
status: LabelingServiceStatus = Field(frozen=True,
default=LabelingServiceStatus.Missing)
status: LabelingServiceStatus = Field(frozen=True, default=None)
data_rows_count: int = Field(frozen=True)
data_rows_in_review_count: int = Field(frozen=True)
data_rows_in_rework_count: int = Field(frozen=True)
data_rows_done_count: int = Field(frozen=True)
media_type: Optional[MediaType] = Field(frozen=True, default=None)
editor_task_type: EditorTaskType = Field(frozen=True, default=None)

client: Any # type Any to avoid circular import from client

Expand All @@ -59,12 +63,43 @@ def __init__(self, **kwargs):

@property
def tasks_completed(self):
"""
Count how many data rows have been completed (i.e. in the Done queue)
"""
return self.data_rows_done_count

@property
def tasks_remaining(self):
"""
Count how many data rows have not been completed
"""
return self.data_rows_count - self.data_rows_done_count

@property
def service_type(self):
"""
Descriptive labeling service definition by media type and editor task type
"""
if self.media_type is None:
return None

if self.editor_task_type is None:
return sentence_case(self.media_type.value)

if self.editor_task_type == EditorTaskType.OfflineModelChatEvaluation and self.media_type == MediaType.Conversational:
return "Offline chat evaluation"

if self.editor_task_type == EditorTaskType.ModelChatEvaluation and self.media_type == MediaType.Conversational:
return "Live chat evaluation"

if self.editor_task_type == EditorTaskType.ResponseCreation and self.media_type == MediaType.Text:
return "Response creation"

if self.media_type == MediaType.LLMPromptCreation or self.media_type == MediaType.LLMPromptResponseCreation:
return "Prompt response creation"

return sentence_case(self.media_type.value)

class Config(_CamelCaseMixin.Config):
...

Expand Down Expand Up @@ -141,8 +176,17 @@ def convert_to_labeling_service_dashboard(client, data):
)

@root_validator(pre=True)
def convert_boost_status_to_enum(cls, data):
def convert_boost_data(cls, data):
if 'boostStatus' in data:
data['status'] = LabelingServiceStatus(data.pop('boostStatus'))

if 'boostRequestedAt' in data:
data['created_at'] = data.pop('boostRequestedAt')

if 'boostUpdatedAt' in data:
data['updated_at'] = data.pop('boostUpdatedAt')

if 'boostRequestedBy' in data:
data['created_by_id'] = data.pop('boostRequestedBy')

return data
26 changes: 22 additions & 4 deletions libs/labelbox/src/labelbox/schema/media_type.py
Original file line number Diff line number Diff line change
@@ -1,5 +1,7 @@
from enum import Enum

from labelbox.utils import camel_case


class MediaType(Enum):
Audio = "AUDIO"
Expand All @@ -23,17 +25,33 @@ class MediaType(Enum):
LLM = "LLM"

@classmethod
def _missing_(cls, name):
def _missing_(cls, value):
"""Handle missing null data types for projects
created without setting allowedMediaType
Handle upper case names for compatibility with
the GraphQL"""

if name is None:
if value is None:
return cls.Unknown

for member in cls.__members__:
if member.name == name.upper():
def matches(value, name):
"""
This will convert string values (from api) to match enum values
Some string values come as snake case (i.e. llm-prompt-creation)
Some string values come as camel case (i.e. llmPromptCreation)
etc depending on which api returns the value
"""
value_upper = value.upper()
name_upper = name.upper()
value_underscore = value.replace("-", "_")
camel_case_value = camel_case(value_underscore)

return (value_upper == name_upper or
value_underscore.upper() == name_upper or
camel_case_value.upper() == name_upper)

for name, member in cls.__members__.items():
if matches(value, name):
return member

@classmethod
Expand Down
22 changes: 14 additions & 8 deletions libs/labelbox/src/labelbox/schema/ontology_kind.py
Original file line number Diff line number Diff line change
Expand Up @@ -21,24 +21,30 @@ def get_ontology_kind_validation_error(cls, ontology_kind):
return TypeError(f"{ontology_kind}: is not a valid ontology kind. Use"
f" any of {OntologyKind.__members__.items()}"
" from OntologyKind.")

@staticmethod
def evaluate_ontology_kind_with_media_type(ontology_kind,
media_type: Optional[MediaType]) -> Union[MediaType, None]:

def evaluate_ontology_kind_with_media_type(
ontology_kind,
media_type: Optional[MediaType]) -> Union[MediaType, None]:

ontology_to_media = {
OntologyKind.ModelEvaluation: (MediaType.Conversational, "For chat evaluation, media_type must be Conversational."),
OntologyKind.ResponseCreation: (MediaType.Text, "For response creation, media_type must be Text.")
OntologyKind.ModelEvaluation:
(MediaType.Conversational,
"For chat evaluation, media_type must be Conversational."),
OntologyKind.ResponseCreation:
(MediaType.Text,
"For response creation, media_type must be Text.")
}

if ontology_kind in ontology_to_media:
expected_media_type, error_message = ontology_to_media[ontology_kind]
expected_media_type, error_message = ontology_to_media[
ontology_kind]

if media_type is None or media_type == expected_media_type:
media_type = expected_media_type
else:
raise ValueError(error_message)

return media_type


Expand Down
9 changes: 9 additions & 0 deletions libs/labelbox/src/labelbox/utils.py
Original file line number Diff line number Diff line change
Expand Up @@ -39,6 +39,15 @@ def snake_case(s):
return _convert(s, "_", lambda i: False)


def sentence_case(s: str) -> str:
""" Converts a string in [snake|camel|title]case to Sentence case. """
# Replace underscores with spaces and convert to lower case
sentence_str = s.replace("_", " ").lower()
# Capitalize the first letter of each word
sentence_str = sentence_str.capitalize()
return sentence_str


def is_exactly_one_set(*args):
return sum([bool(arg) for arg in args]) == 1

Expand Down
8 changes: 8 additions & 0 deletions libs/labelbox/tests/integration/test_labeling_dashboard.py
Original file line number Diff line number Diff line change
@@ -1,6 +1,8 @@
from datetime import datetime, timedelta
from labelbox.schema.labeling_service import LabelingServiceStatus
from labelbox.schema.search_filters import DateOperator, DateRange, DateRangeOperator, DateRangeValue, DateValue, IdOperator, OperationType, OrganizationFilter, WorkforceRequestedDateFilter, WorkforceRequestedDateRangeFilter, WorkspaceFilter
from labelbox.schema.ontology_kind import EditorTaskType
from labelbox.schema.media_type import MediaType


def test_request_labeling_service_dashboard(rand_gen,
Expand All @@ -18,6 +20,9 @@ def test_request_labeling_service_dashboard(rand_gen,
assert labeling_service_dashboard.status == LabelingServiceStatus.Missing
assert labeling_service_dashboard.tasks_completed == 0
assert labeling_service_dashboard.tasks_remaining == 0
assert labeling_service_dashboard.media_type == MediaType.Conversational
assert labeling_service_dashboard.editor_task_type == EditorTaskType.OfflineModelChatEvaluation
assert labeling_service_dashboard.service_type == "Offline chat evaluation"

labeling_service_dashboard = [
ld for ld in project.client.get_labeling_service_dashboards()
Expand Down Expand Up @@ -82,3 +87,6 @@ def test_request_labeling_service_dashboard_filters(requested_labeling_service):
]
assert len(labeling_service_dashboard) == 0
assert labeling_service_dashboard == []
labeling_service_dashboard = project.client.get_labeling_service_dashboards(
).get_one()
assert labeling_service_dashboard
10 changes: 9 additions & 1 deletion libs/labelbox/tests/unit/test_utils.py
Original file line number Diff line number Diff line change
@@ -1,5 +1,5 @@
import pytest
from labelbox.utils import format_iso_datetime, format_iso_from_string
from labelbox.utils import format_iso_datetime, format_iso_from_string, sentence_case


@pytest.mark.parametrize('datetime_str, expected_datetime_str',
Expand All @@ -11,3 +11,11 @@ def test_datetime_parsing(datetime_str, expected_datetime_str):
# NOTE I would normally not take 'expected' using another function from sdk code, but in this case this is exactly the usage in _validate_parse_datetime
assert format_iso_datetime(
format_iso_from_string(datetime_str)) == expected_datetime_str


@pytest.mark.parametrize(
'str, expected_str',
[('AUDIO', 'Audio'),
('LLM_PROMPT_RESPONSE_CREATION', 'Llm prompt response creation')])
def test_sentence_case(str, expected_str):
assert sentence_case(str) == expected_str
Loading