gitea_admin/sientia-dataops-model-manager
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

SIENTIAPDE-1243: Refactor and enhance model manager activities and workflows

Browse Source 
This commit includes several changes:

- Reorganized imports and class inheritance in activities.py, gates.py and mlflow.py for better readability and maintainability.
- Improved error handling and logging in gates.py and mlflow.py.
- Added input validation and filtering in gates.py to ensure data quality.
- Enhanced prediction formatting and storage policy management in gates.py.
- Updated metrics.py to use consistent naming conventions and labels.
- Refactored connectors_config.py to use type hints and improve code clarity.
- Updated conditional and MLFlow filters for better data quality checks.
- Improved model repository logic for retraining and updating models.
- Enhanced worker.py to include SDK metrics and improved error handling.
- Refactored workflows for better modularity and error handling.
- Updated tests to reflect the changes and improve test coverage.
This commit is contained in:
Bruno Domingues
2025-10-01 17:28:57 -03:00
parent b102f79087
commit dfc190c818
24 changed files with 1482 additions and 1399 deletions
Download Patch File Download Diff File Expand all files Collapse all files

37
tests/laborious/activities/test_activities.py
View File

@@ -1,16 +1,17 @@
from unittest.mock import ANY, MagicMock, patch
from pytest import mark
from unittest.mock import patch, MagicMock, ANY
from sientia_do.temporal.activities.postgres import Postgres
from model_manager.activities.activities import Activities
from model_manager.activities.mlflow import MLFlow
from model_manager.activities.gates import Gates
from model_manager.activities.mlflow import MLFlow
@patch('model_manager.activities.activities.Postgres.__init__')
@patch('model_manager.activities.activities.MLFlow.__init__')
@patch('model_manager.activities.activities.Gates.__init__')
def test___init__(mock_gates_init, mock_mlflow_init, mock_postgres_init):
postgres_config = {
'host': 'localhost',
'port': 5432,
@@ -18,15 +19,10 @@ def test___init__(mock_gates_init, mock_mlflow_init, mock_postgres_init):
'password': 'postgres',
'dbname': 'postgres',
'min_connections': 1,
'max_connections': 10
'max_connections': 10,
}
mlflow_config = {
'host': 'localhost',
'port': 5000,
'username': 'mlflow',
'password': 'mlflow'
}
mlflow_config = {'host': 'localhost', 'port': 5000, 'username': 'mlflow', 'password': 'mlflow'}
logger = MagicMock()
notification_handler = MagicMock()
@@ -35,7 +31,7 @@ def test___init__(mock_gates_init, mock_mlflow_init, mock_postgres_init):
postgres_config=postgres_config,
mlflow_config=mlflow_config,
logger=logger,
notification_handler=notification_handler
notification_handler=notification_handler,
)
assert isinstance(activities, Activities)
@@ -53,7 +49,7 @@ def test___init__(mock_gates_init, mock_mlflow_init, mock_postgres_init):
min_connections=postgres_config['min_connections'],
max_connections=postgres_config['max_connections'],
logger=logger,
notification_handler=notification_handler
notification_handler=notification_handler,
)
mock_mlflow_init.assert_called_once_with(
@@ -63,13 +59,11 @@ def test___init__(mock_gates_init, mock_mlflow_init, mock_postgres_init):
mlflow_username=mlflow_config['username'],
mlflow_password=mlflow_config['password'],
logger=logger,
notification_handler=notification_handler
notification_handler=notification_handler,
)
mock_gates_init.assert_called_once_with(
ANY,
logger=logger,
notification_handler=notification_handler
ANY, logger=logger, notification_handler=notification_handler
)
@@ -84,15 +78,10 @@ async def test_shutdown(_mock_mlflow_init, mock_postgres_init):
'password': 'postgres',
'dbname': 'postgres',
'min_connections': 1,
'max_connections': 10
'max_connections': 10,
}
mlflow_config = {
'host': 'localhost',
'port': 5000,
'username': 'mlflow',
'password': 'mlflow'
}
mlflow_config = {'host': 'localhost', 'port': 5000, 'username': 'mlflow', 'password': 'mlflow'}
logger = MagicMock()
notification_handler = MagicMock()
@@ -101,7 +90,7 @@ async def test_shutdown(_mock_mlflow_init, mock_postgres_init):
postgres_config=postgres_config,
mlflow_config=mlflow_config,
logger=logger,
notification_handler=notification_handler
notification_handler=notification_handler,
)
await activities.shutdown()

237
tests/laborious/activities/test_gates.py
View File

@@ -1,6 +1,8 @@
from unittest.mock import MagicMock, ANY, patch
from unittest.mock import ANY, MagicMock, patch
from pytest import fixture, mark
from sientia_do.notifications.models import NotificationLevel
from model_manager.activities.gates import Gates
@@ -20,11 +22,11 @@ def gates_activity():
metadata = {
"metadata": {
"model_id": "test_model",
"model_name": "test_model",
"workflow_name": "test_workflow",
"schema_name": "test_schedule",
'metadata': {
'model_id': 'test_model',
'model_name': 'test_model',
'workflow_name': 'test_workflow',
'schema_name': 'test_schedule',
},
}
@@ -34,20 +36,18 @@ async def test_input_gate_invalid_filter(gates_activity):
# Arrange
input_data = {
**metadata,
'filters': {
'INVALID_FILTER': {'POLICY': 'STOP'}
},
'filters': {'INVALID_FILTER': {'POLICY': 'STOP'}},
'data': {'value': [1, 2, 3]},
'path_priority': ['STOP', 'CONTINUE', 'REPEAT']
'path_priority': ['STOP', 'CONTINUE', 'REPEAT'],
}
# Act
result = await gates_activity.input_gate(input_data)
# Assert
assert result == (None, 0, "")
assert result == (None, 0, '')
gates_activity.error.assert_called_once_with(
"Filter INVALID_FILTER not found", metadata['metadata']
'Filter INVALID_FILTER not found', metadata['metadata']
)
@@ -57,28 +57,27 @@ async def test_input_gate_filter_exception(mock_input_filter_functions, gates_ac
# Arrange
mock_input_filter_functions.__contains__.return_value = True
mock_input_filter_functions.__getitem__.return_value = MagicMock(
side_effect=Exception("Test error"))
side_effect=Exception('Test error')
)
input_data = {
**metadata,
'filters': {
'EMPTY_DATA': {'policy': 'STOP', 'config': {}}
},
'filters': {'EMPTY_DATA': {'policy': 'STOP', 'config': {}}},
'data': {'value': []},
'path_priority': ['STOP', 'CONTINUE', 'REPEAT']
'path_priority': ['STOP', 'CONTINUE', 'REPEAT'],
}
# Act
result = await gates_activity.input_gate(input_data)
# Assert
assert result == (None, 0, "")
assert result == (None, 0, '')
gates_activity.send_notification.assert_called_once_with(
metadata=metadata['metadata'],
notification_id="INTPUT_GATE_ERROR__EMPTY_DATA",
notification_id='INTPUT_GATE_ERROR__EMPTY_DATA',
message="Error in filter EMPTY_DATA:{'policy': 'STOP', 'config': {}}: \n Test error",
block="input_gate",
block='input_gate',
level=NotificationLevel.ERROR,
attachment_content=ANY
attachment_content=ANY,
)
@@ -89,14 +88,14 @@ async def test_input_gate_no_filters(gates_activity):
**metadata,
'filters': {},
'data': {'value': [1, 2, 3]},
'path_priority': ['CONTINUE', 'STOP', 'REPEAT']
'path_priority': ['CONTINUE', 'STOP', 'REPEAT'],
}
# Act
result = await gates_activity.input_gate(input_data)
# Assert
assert result == (None, 0, "")
assert result == (None, 0, '')
gates_activity.debug.assert_called()
@@ -105,18 +104,16 @@ async def test_input_gate_with_filter(gates_activity):
# Arrange
input_data = {
**metadata,
'filters': {
'EMPTY_DATA': {'policy': 'STOP', 'config': {}}
},
'filters': {'EMPTY_DATA': {'policy': 'STOP', 'config': {}}},
'data': {'value': []},
'path_priority': ['STOP', 'CONTINUE', 'REPEAT']
'path_priority': ['STOP', 'CONTINUE', 'REPEAT'],
}
# Act
result = await gates_activity.input_gate(input_data)
# Assert
assert result == ('STOP', -1, "Input data with bad quality")
assert result == ('STOP', -1, 'Input data with bad quality')
gates_activity.debug.assert_called()
@@ -125,51 +122,49 @@ async def test_mlflow_response_gate_invalid_filter(gates_activity):
# Arrange
input_data = {
**metadata,
'filters': {
'INVALID_FILTER': {'POLICY': 'STOP'}
},
'filters': {'INVALID_FILTER': {'POLICY': 'STOP'}},
'data': {'content': {'message': 'success'}},
'type': 'test',
'path_priority': ['STOP', 'CONTINUE', 'REPEAT']
'path_priority': ['STOP', 'CONTINUE', 'REPEAT'],
}
# Act
result = await gates_activity.mlflow_response_gate(input_data)
# Assert
assert result == (None, 0, "")
assert result == (None, 0, '')
@mark.asyncio
@patch('model_manager.activities.gates.mlflow_response_filter_functions')
async def test_mlflow_response_gate_filter_exception(mock_mlflow_response_filter_functions,
gates_activity):
async def test_mlflow_response_gate_filter_exception(
mock_mlflow_response_filter_functions, gates_activity
):
# Arrange
mock_mlflow_response_filter_functions.__contains__.return_value = True
mock_mlflow_response_filter_functions.__getitem__.return_value = MagicMock(
side_effect=Exception("Test error"))
side_effect=Exception('Test error')
)
input_data = {
**metadata,
'filters': {
'INVALID_FILTER': {'POLICY': 'STOP'}
},
'filters': {'INVALID_FILTER': {'POLICY': 'STOP'}},
'data': {'content': {'message': 'success'}},
'type': 'test',
'path_priority': ['STOP', 'CONTINUE', 'REPEAT']
'path_priority': ['STOP', 'CONTINUE', 'REPEAT'],
}
# Act
result = await gates_activity.mlflow_response_gate(input_data)
# Assert
assert result == (None, 0, "")
assert result == (None, 0, '')
gates_activity.send_notification.assert_called_once_with(
metadata=metadata['metadata'],
notification_id="MLFLOW_GATE_RESPONSE_FILTER__INVALID_FILTER",
notification_id='MLFLOW_GATE_RESPONSE_FILTER__INVALID_FILTER',
message="Error in filter INVALID_FILTER:{'POLICY': 'STOP'}: \n Test error",
block="mlflow_gate",
block='mlflow_gate',
level=NotificationLevel.ERROR,
attachment_content=ANY
attachment_content=ANY,
)
@@ -181,14 +176,14 @@ async def test_mlflow_response_gate_no_filters(gates_activity):
'filters': {},
'data': {'content': {'message': 'success'}},
'type': 'test',
'path_priority': ['CONTINUE', 'STOP', 'REPEAT']
'path_priority': ['CONTINUE', 'STOP', 'REPEAT'],
}
# Act
result = await gates_activity.mlflow_response_gate(input_data)
# Assert
assert result == (None, 0, "")
assert result == (None, 0, '')
gates_activity.debug.assert_called()
@@ -197,25 +192,20 @@ async def test_mlflow_response_gate_with_filter(gates_activity):
# Arrange
input_data = {
**metadata,
'filters': {
'API_ERROR': {'policy': 'STOP'}
},
'filters': {'API_ERROR': {'policy': 'STOP'}},
'data': {
'success': False,
'content': {
'message': 'API error occurred',
'traceback': 'error trace'
}
'content': {'message': 'API error occurred', 'traceback': 'error trace'},
},
'type': 'test',
'path_priority': ['STOP', 'CONTINUE', 'REPEAT']
'path_priority': ['STOP', 'CONTINUE', 'REPEAT'],
}
# Act
result = await gates_activity.mlflow_response_gate(input_data)
# Assert
assert result == ('STOP', -1, "API error occurred")
assert result == ('STOP', -1, 'API error occurred')
gates_activity.debug.assert_called()
gates_activity.send_notification.assert_called()
@@ -225,58 +215,53 @@ async def test_mlflow_content_gate_invalid_filter(gates_activity):
# Arrange
input_data = {
**metadata,
'filters': {
'INVALID_FILTER': {'POLICY': 'STOP'}
},
'filters': {'INVALID_FILTER': {'POLICY': 'STOP'}},
'data': {'value': [1, 2, 3]},
'type': 'test',
'path_priority': ['STOP', 'CONTINUE', 'REPEAT']
'path_priority': ['STOP', 'CONTINUE', 'REPEAT'],
}
# Act
result = await gates_activity.mlflow_content_gate(input_data)
# Assert
assert result == (None, 0, "")
assert result == (None, 0, '')
@mark.asyncio
@patch('model_manager.activities.gates.mlflow_content_filter_functions')
async def test_mlflow_content_gate_filter_exception(mock_mlflow_content_filter_functions,
gates_activity):
async def test_mlflow_content_gate_filter_exception(
mock_mlflow_content_filter_functions, gates_activity
):
# Arrange
mock_mlflow_content_filter_functions.__contains__.return_value = True
mock_mlflow_content_filter_functions.__getitem__.return_value = MagicMock(
side_effect=Exception("Test error"))
side_effect=Exception('Test error')
)
input_data = {
**metadata,
'filters': {
'API_ERROR': {'POLICY': 'STOP'}
},
'filters': {'API_ERROR': {'POLICY': 'STOP'}},
'data': {
'success': False,
'content': {
'message': 'API error occurred',
'traceback': 'error trace'
}
'content': {'message': 'API error occurred', 'traceback': 'error trace'},
},
'type': 'test',
'path_priority': ['STOP', 'CONTINUE', 'REPEAT']
'path_priority': ['STOP', 'CONTINUE', 'REPEAT'],
}
# Act
result = await gates_activity.mlflow_content_gate(input_data)
# Assert
assert result == (None, 0, "")
assert result == (None, 0, '')
gates_activity.debug.assert_called()
gates_activity.send_notification.assert_called_once_with(
metadata=metadata['metadata'],
notification_id="MLFLOW_GATE_CONTENT_FILTER__API_ERROR",
notification_id='MLFLOW_GATE_CONTENT_FILTER__API_ERROR',
message="Error in filter API_ERROR:{'POLICY': 'STOP'}: \n Test error",
block="mlflow_gate",
block='mlflow_gate',
level=NotificationLevel.ERROR,
attachment_content=ANY
attachment_content=ANY,
)
@@ -288,14 +273,14 @@ async def test_mlflow_content_gate_no_filters(gates_activity):
'filters': {},
'data': {'value': [1, 2, 3]},
'type': 'test',
'path_priority': ['CONTINUE', 'STOP', 'REPEAT']
'path_priority': ['CONTINUE', 'STOP', 'REPEAT'],
}
# Act
result = await gates_activity.mlflow_content_gate(input_data)
# Assert
assert result == (None, 0, "")
assert result == (None, 0, '')
gates_activity.debug.assert_called()
@@ -304,20 +289,17 @@ async def test_mlflow_content_gate_with_filter(gates_activity):
# Arrange
input_data = {
**metadata,
'filters': {
'NAN_VALUES': {'policy': 'STOP', 'config': {}}
},
'filters': {'NAN_VALUES': {'policy': 'STOP', 'config': {}}},
'data': {'value': [None, None, None]},
'type': 'test',
'path_priority': ['STOP', 'CONTINUE', 'REPEAT']
'path_priority': ['STOP', 'CONTINUE', 'REPEAT'],
}
# Act
result = await gates_activity.mlflow_content_gate(input_data)
# Assert
assert result == (
'STOP', -1, "Transformed data not passed the content filter")
assert result == ('STOP', -1, 'Transformed data not passed the content filter')
gates_activity.debug.assert_called()
gates_activity.send_notification.assert_called()
@@ -328,7 +310,8 @@ def test_get_prediction_store_policy_invalid_policy(gates_activity):
# Act
policy_type, policy_value = gates_activity.get_prediction_store_policy(
prediction_store_policy, metadata)
prediction_store_policy, metadata
)
# Assert
assert policy_type == 'lts'
@@ -341,7 +324,8 @@ def test_get_prediction_store_policy_invalid_policy_value(gates_activity):
# Act
policy_type, policy_value = gates_activity.get_prediction_store_policy(
prediction_store_policy, metadata)
prediction_store_policy, metadata
)
# Assert
assert policy_type == 'lts'
@@ -354,7 +338,8 @@ def test_get_prediction_store_policy_valid_policy_type(gates_activity):
# Act
policy_type, policy_value = gates_activity.get_prediction_store_policy(
prediction_store_policy, metadata)
prediction_store_policy, metadata
)
# Assert
assert policy_type == 'lts'
@@ -367,7 +352,8 @@ def test_get_prediction_store_policy_valid_policy(gates_activity):
# Act
policy_type, policy_value = gates_activity.get_prediction_store_policy(
prediction_store_policy, metadata)
prediction_store_policy, metadata
)
# Assert
assert policy_type == 'erl'
@@ -380,16 +366,12 @@ async def test_format_prediction_no_timestamp(gates_activity):
input_data = {
**metadata,
'data': {
'prediction': {
'2023-05-26 11:12:27': 1
},
'response_time': {
'2023-05-26 11:12:27': 0.1
}
'prediction': {'2023-05-26 11:12:27': 1},
'response_time': {'2023-05-26 11:12:27': 0.1},
},
'model_id': 'test_model',
'prediction_confidence': 0.9,
'prediction_store_policy': 'lts:1'
'prediction_store_policy': 'lts:1',
}
# Act
@@ -402,7 +384,7 @@ async def test_format_prediction_no_timestamp(gates_activity):
assert result['model_id'] == {0: 'test_model'}
assert result['prediction_confidence'] == {0: 0.9}
assert result['prediction_status'] == {0: 'Good'}
assert result['comments'] == {0: ""}
assert result['comments'] == {0: ''}
@mark.asyncio
@@ -420,11 +402,11 @@ async def test_format_prediction_with_timestamp_erl(gates_activity):
'2023-05-26 11:12:27': 0.1,
'2023-05-26 11:12:28': 0.2,
'2023-05-26 11:12:29': 0.3,
}
},
},
'model_id': 'test_model',
'prediction_confidence': 0.9,
'prediction_store_policy': 'erl:2'
'prediction_store_policy': 'erl:2',
}
# Act
@@ -433,12 +415,11 @@ async def test_format_prediction_with_timestamp_erl(gates_activity):
# Assert
assert result['prediction'] == {0: 2, 1: 1}
assert result['response_time'] == {0: 0.2, 1: 0.1}
assert result['timestamp'] == {
0: '2023-05-26 11:12:28', 1: '2023-05-26 11:12:27'}
assert result['timestamp'] == {0: '2023-05-26 11:12:28', 1: '2023-05-26 11:12:27'}
assert result['model_id'] == {0: 'test_model', 1: 'test_model'}
assert result['prediction_confidence'] == {0: 0.9, 1: 0.9}
assert result['prediction_status'] == {0: 'Good', 1: 'Good'}
assert result['comments'] == {0: "", 1: ""}
assert result['comments'] == {0: '', 1: ''}
@mark.asyncio
@@ -456,11 +437,11 @@ async def test_format_prediction_with_timestamp_lts(gates_activity):
'2023-05-26 11:12:27': 0.1,
'2023-05-26 11:12:28': 0.2,
'2023-05-26 11:12:29': 0.3,
}
},
},
'model_id': 'test_model',
'prediction_confidence': 0.9,
'prediction_store_policy': 'lts:2'
'prediction_store_policy': 'lts:2',
}
# Act
@@ -469,12 +450,11 @@ async def test_format_prediction_with_timestamp_lts(gates_activity):
# Assert
assert result['prediction'] == {0: 3, 1: 2}
assert result['response_time'] == {0: 0.3, 1: 0.2}
assert result['timestamp'] == {
0: '2023-05-26 11:12:29', 1: '2023-05-26 11:12:28'}
assert result['timestamp'] == {0: '2023-05-26 11:12:29', 1: '2023-05-26 11:12:28'}
assert result['model_id'] == {0: 'test_model', 1: 'test_model'}
assert result['prediction_confidence'] == {0: 0.9, 1: 0.9}
assert result['prediction_status'] == {0: 'Good', 1: 'Good'}
assert result['comments'] == {0: "", 1: ""}
assert result['comments'] == {0: '', 1: ''}
@mark.asyncio
@@ -482,22 +462,23 @@ async def test_format_prediction_with_timestamp_invalid_policy(gates_activity):
# Arrange
input_data = {
**metadata,
'data': {'prediction': [1, 2, 3],
'response_time': [0.1, 0.2, 0.3],
'timestamp': ['2023-05-26 11:12:27', '2023-05-26 11:12:28', '2023-05-26 11:12:29']},
'data': {
'prediction': [1, 2, 3],
'response_time': [0.1, 0.2, 0.3],
'timestamp': ['2023-05-26 11:12:27', '2023-05-26 11:12:28', '2023-05-26 11:12:29'],
},
'model_id': 'test_model',
'prediction_confidence': 0.9,
'prediction_store_policy': 'lts:2'
'prediction_store_policy': 'lts:2',
}
gates_activity.get_prediction_store_policy = MagicMock(
return_value=('invalid', 1))
gates_activity.get_prediction_store_policy = MagicMock(return_value=('invalid', 1))
try:
result = await gates_activity.format_prediction(input_data)
await gates_activity.format_prediction(input_data)
except ValueError as e:
assert str(e) == "Invalid policy type: invalid"
assert str(e) == 'Invalid policy type: invalid'
else:
assert False, "Expected ValueError"
raise AssertionError('Expected ValueError')
@mark.asyncio
@@ -508,7 +489,7 @@ async def test_format_default_prediction(gates_activity):
'timestamp': '2023-05-26 11:12:27',
'model_id': 'test_model',
'prediction_confidence': 0.1,
'comment': 'Test comment'
'comment': 'Test comment',
}
# Act
@@ -528,12 +509,7 @@ async def test_format_default_prediction(gates_activity):
@mark.asyncio
async def test_get_last_timestamp_with_data(gates_activity):
# Arrange
input_data = {
**metadata,
'data': {
'timestamp': ['2023-05-26 11:12:27', '2023-05-26 11:12:28']
}
}
input_data = {**metadata, 'data': {'timestamp': ['2023-05-26 11:12:27', '2023-05-26 11:12:28']}}
# Act
result = await gates_activity.get_last_timestamp(input_data)
@@ -545,10 +521,7 @@ async def test_get_last_timestamp_with_data(gates_activity):
@mark.asyncio
async def test_get_last_timestamp_no_data(gates_activity):
# Arrange
input_data = {
'data': {},
**metadata
}
input_data = {'data': {}, **metadata}
# Act
result = await gates_activity.get_last_timestamp(input_data)
@@ -567,30 +540,28 @@ async def test_write_metrics(mock_metrics, gates_activity):
'prediction': {
'prediction': [1, 2, 3],
'prediction_confidence': [0.9, 0.8, 0.7],
'response_time': [0.1, 0.2, 0.3]
}
'response_time': [0.1, 0.2, 0.3],
},
}
await gates_activity.write_metrics(input_data)
mock_metrics.PREDICTIONS_WRITTEN_COUNT.labels.assert_called_once_with(
pod_id=gates_activity.pod_id,
model_name=metadata['metadata']['model_name'],
pipeline_name=metadata['metadata']['workflow_name']
pipeline_name=metadata['metadata']['workflow_name'],
)
mock_metrics.PREDICTIONS_WRITTEN_COUNT.labels.return_value.inc.assert_called_once_with()
mock_metrics.PREDICTION_CONFIDENCE_MONITOR.labels.assert_called_once_with(
pod_id=gates_activity.pod_id,
model_name=metadata['metadata']['model_name'],
pipeline_name=metadata['metadata']['workflow_name']
)
mock_metrics.PREDICTION_CONFIDENCE_MONITOR.labels.return_value.set.assert_called_once_with(
0.9
pipeline_name=metadata['metadata']['workflow_name'],
)
mock_metrics.PREDICTION_CONFIDENCE_MONITOR.labels.return_value.set.assert_called_once_with(0.9)
mock_metrics.PREDICTION_RESPONSE_TIME_MONITOR.labels.assert_called_once_with(
pod_id=gates_activity.pod_id,
model_name=metadata['metadata']['model_name'],
pipeline_name=metadata['metadata']['workflow_name']
pipeline_name=metadata['metadata']['workflow_name'],
)
mock_metrics.PREDICTION_RESPONSE_TIME_MONITOR.labels.return_value.observe.assert_called_once_with(
0.1

212
tests/laborious/activities/test_mlflow.py
View File

@@ -1,45 +1,42 @@
from datetime import datetime
from unittest.mock import ANY, MagicMock, patch
import numpy as np
from pandas import DataFrame, Timestamp
from pytest import fixture, mark, raises
from sientia_do.temporal.constants import DATETIME_FORMAT, DATETIME_FORMAT_WITH_TZ
from model_manager.activities.mlflow import MLFlow
from pytest import fixture, mark
from sientia_do.notifications.models import NotificationLevel
from sientia_do.temporal.constants import DATETIME_FORMAT, DATETIME_FORMAT_WITH_TZ
from model_manager.activities.mlflow import MLFlow
@patch("model_manager.activities.mlflow.MLFlowRepository")
@patch('model_manager.activities.mlflow.MLFlowRepository')
def test___init__(mock_mlflow_repository):
mlflow = MLFlow(
mlflow_host="http://localhost",
mlflow_host='http://localhost',
mlflow_port=5000,
mlflow_username="admin",
mlflow_password="admin",
mlflow_username='admin',
mlflow_password='admin',
logger=MagicMock(),
notification_handler=MagicMock()
notification_handler=MagicMock(),
)
assert mlflow.mlflow_host == "http://localhost"
assert mlflow.mlflow_host == 'http://localhost'
assert mlflow.mlflow_port == 5000
assert mlflow.mlflow_username == "admin"
assert mlflow.mlflow_password == "admin"
assert mlflow.mlflow_username == 'admin'
assert mlflow.mlflow_password == 'admin'
mock_mlflow_repository.assert_called_once_with(
"http://localhost:5000", "admin", "admin", ANY
)
mock_mlflow_repository.assert_called_once_with('http://localhost:5000', 'admin', 'admin', ANY)
@fixture
@patch("model_manager.activities.mlflow.MLFlowRepository")
@patch('model_manager.activities.mlflow.MLFlowRepository')
def mlflow(mock_mlflow_repository):
mlflow = MLFlow(
mlflow_host="http://localhost:5000",
mlflow_host='http://localhost:5000',
mlflow_port=5000,
mlflow_username="admin",
mlflow_password="admin",
mlflow_username='admin',
mlflow_password='admin',
logger=MagicMock(),
notification_handler=MagicMock()
notification_handler=MagicMock(),
)
mlflow.send_notification = MagicMock()
@@ -48,44 +45,67 @@ def mlflow(mock_mlflow_repository):
metadata = {
"metadata": {
"model_id": "test_model",
"model_name": "test_model",
"workflow_name": "test_workflow",
"schema_name": "test_schedule",
'metadata': {
'model_id': 'test_model',
'model_name': 'test_model',
'workflow_name': 'test_workflow',
'schema_name': 'test_schedule',
},
}
@mark.asyncio
@patch("model_manager.activities.mlflow.DataFrame")
@patch("model_manager.activities.mlflow.max")
@patch('model_manager.activities.mlflow.DataFrame')
@patch('model_manager.activities.mlflow.max')
async def test_request_transform_success(mock_max, mock_dataframe, mlflow):
mock_max.return_value = '2024-01-02'
# Mock input data
input_data = {
**metadata,
'data': [
{'timestamp': '2024-01-01', 'variable': 'var1',
'value': 1.0, 'created_at': '2024-01-01 12:00:00'},
{'timestamp': '2024-01-01', 'variable': 'var2',
'value': 2.0, 'created_at': '2024-01-01 12:00:00'},
{'timestamp': '2024-01-02', 'variable': 'var1',
'value': 3.0, 'created_at': '2024-01-02 12:00:00'},
{'timestamp': '2024-01-02', 'variable': 'var2',
'value': 4.0, 'created_at': '2024-01-02 12:00:00'},
{'timestamp': '2024-01-02', 'variable': 'var1',
'value': 1.0, 'created_at': '2024-01-01 12:00:00'},
{'timestamp': '2024-01-02', 'variable': 'var2',
'value': 1.0, 'created_at': '2024-01-01 12:00:00'}
{
'timestamp': '2024-01-01',
'variable': 'var1',
'value': 1.0,
'created_at': '2024-01-01 12:00:00',
},
{
'timestamp': '2024-01-01',
'variable': 'var2',
'value': 2.0,
'created_at': '2024-01-01 12:00:00',
},
{
'timestamp': '2024-01-02',
'variable': 'var1',
'value': 3.0,
'created_at': '2024-01-02 12:00:00',
},
{
'timestamp': '2024-01-02',
'variable': 'var2',
'value': 4.0,
'created_at': '2024-01-02 12:00:00',
},
{
'timestamp': '2024-01-02',
'variable': 'var1',
'value': 1.0,
'created_at': '2024-01-01 12:00:00',
},
{
'timestamp': '2024-01-02',
'variable': 'var2',
'value': 1.0,
'created_at': '2024-01-01 12:00:00',
},
],
'model_name': 'test_model',
'model_config': {}
'model_config': {},
}
# Mock the transform response
expected_response = {'prediction': [0.5, 0.6], 'timestamp': [
'2024-01-01', '2024-01-02']}
expected_response = {'prediction': [0.5, 0.6], 'timestamp': ['2024-01-01', '2024-01-02']}
mlflow.model_monitoring_repository.transform.return_value = expected_response
mock_dataframe.return_value.sort_values.return_value = mock_dataframe.return_value
@@ -114,30 +134,25 @@ async def test_request_transform_success(mock_max, mock_dataframe, mlflow):
@mark.asyncio
@patch("model_manager.activities.mlflow.DataFrame")
@patch("model_manager.activities.mlflow.to_datetime")
@patch("model_manager.activities.mlflow.max")
@patch('model_manager.activities.mlflow.DataFrame')
@patch('model_manager.activities.mlflow.to_datetime')
@patch('model_manager.activities.mlflow.max')
async def test_request_predict(mock_max, mock_to_datetime, mock_dataframe, mlflow):
mock_max.return_value = '2024-01-02'
# Mock input data
input_data = {
**metadata,
'data': {
"variable": {
"2024-01-01": "var1",
"2024-01-02": "var2",
"2024-01-03": "var1",
"2024-01-04": "var2"
'variable': {
'2024-01-01': 'var1',
'2024-01-02': 'var2',
'2024-01-03': 'var1',
'2024-01-04': 'var2',
},
"value": {
"2024-01-01": 1.0,
"2024-01-02": 2.0,
"2024-01-03": 3.0,
"2024-01-04": 4.0
}
'value': {'2024-01-01': 1.0, '2024-01-02': 2.0, '2024-01-03': 3.0, '2024-01-04': 4.0},
},
'model_name': 'test_model',
'model_config': {}
'model_config': {},
}
# Mock the predict response
@@ -148,9 +163,7 @@ async def test_request_predict(mock_max, mock_to_datetime, mock_dataframe, mlflo
response_data = await mlflow.request_predict(input_data)
mock_dataframe.assert_called_once_with(input_data['data'])
mock_dataframe.return_value.replace.assert_called_once_with(
np.nan, None, inplace=True
)
mock_dataframe.return_value.replace.assert_called_once_with(np.nan, None, inplace=True)
mock_dataframe.return_value.__setitem__.assert_any_call(
'timestamp', mock_to_datetime.return_value.dt.strftime.return_value
)
@@ -158,9 +171,7 @@ async def test_request_predict(mock_max, mock_to_datetime, mock_dataframe, mlflo
mock_to_datetime.assert_called_once_with(
mock_dataframe.return_value.__getitem__.return_value, format=DATETIME_FORMAT_WITH_TZ
)
mock_to_datetime.return_value.dt.strftime.assert_called_once_with(
DATETIME_FORMAT
)
mock_to_datetime.return_value.dt.strftime.assert_called_once_with(DATETIME_FORMAT)
# Verify the response
assert response_data == expected_response
@@ -174,28 +185,26 @@ async def test_request_predict(mock_max, mock_to_datetime, mock_dataframe, mlflo
@mark.asyncio
async def test_retrain_model(mlflow):
data = {
"model_id": [4, 5, 6, 7],
"created_at": [1, 2, 3, 4],
"timestamp": [1, 1, 2, 2],
"variable": ["var1", "var2", "var1", "var2"],
"value": [1, 2, 3, 4]
'model_id': [4, 5, 6, 7],
'created_at': [1, 2, 3, 4],
'timestamp': [1, 1, 2, 2],
'variable': ['var1', 'var2', 'var1', 'var2'],
'value': [1, 2, 3, 4],
}
mlflow.model_monitoring_repository.retrain_model.return_value = (
'Model retrained successfully', 'test')
'Model retrained successfully',
'test',
)
response = await mlflow.retrain_model({
**metadata,
'data': data,
'model_name': 'test_model'
})
response = await mlflow.retrain_model({**metadata, 'data': data, 'model_name': 'test_model'})
mlflow.model_monitoring_repository.retrain_model.assert_called_once()
assert response == {
"status": 'Model retrained successfully',
"timestamp": 2,
"experiment": 'test'
'status': 'Model retrained successfully',
'timestamp': 2,
'experiment': 'test',
}
@@ -206,20 +215,16 @@ async def test_retrain_model_error(mlflow):
)
data = {
"model_id": [4, 5, 6, 7],
"created_at": [1, 2, 3, 4],
"timestamp": [1, 1, 2, 2],
"variable": ["var1", "var2", "var1", "var2"],
"value": [1, 2, 3, 4]
'model_id': [4, 5, 6, 7],
'created_at': [1, 2, 3, 4],
'timestamp': [1, 1, 2, 2],
'variable': ['var1', 'var2', 'var1', 'var2'],
'value': [1, 2, 3, 4],
}
try:
await mlflow.retrain_model({
**metadata,
'data': data,
'model_name': 'test_model'
})
except Exception as e:
await mlflow.retrain_model({**metadata, 'data': data, 'model_name': 'test_model'})
except Exception as e: # noqa: BLE001
assert str(e) == 'Error retraining model'
mlflow.send_notification.assert_called_once_with(
metadata=metadata['metadata'],
@@ -227,20 +232,18 @@ async def test_retrain_model_error(mlflow):
message='Error retraining model test_model: Error retraining model',
block='retrain_model',
level=NotificationLevel.ERROR,
attachment_content=ANY
attachment_content=ANY,
)
else:
assert False, "No exception raised"
raise AssertionError('No exception raised')
@mark.asyncio
async def test_update_production_model(mlflow):
mlflow.model_monitoring_repository.update_production_model.return_value = (
{
"data1": 1,
"data2": 2
}
)
mlflow.model_monitoring_repository.update_production_model.return_value = {
'data1': 1,
'data2': 2,
}
input_data = {
**metadata,
@@ -248,13 +251,14 @@ async def test_update_production_model(mlflow):
'model_id': 1,
'experiment': 'test',
'timestamp': 2,
'status': 'success'
'status': 'success',
}
response = await mlflow.update_production_model(input_data)
mlflow.model_monitoring_repository.update_production_model.assert_called_once_with(
experiment='test', model_name='test_model')
experiment='test', model_name='test_model'
)
assert response == {
'data1': {0: 1},
@@ -262,7 +266,7 @@ async def test_update_production_model(mlflow):
'model_id': {0: 1},
'model_name': {0: 'test_model'},
'timestamp': {0: 2},
'status': {0: 'success'}
'status': {0: 'success'},
}
@@ -278,12 +282,12 @@ async def test_update_production_model_error(mlflow):
'model_id': 1,
'experiment': 'test',
'timestamp': 2,
'status': 'success'
'status': 'success',
}
try:
await mlflow.update_production_model(input_data)
except Exception as e:
except Exception as e: # noqa: BLE001
assert str(e) == 'Error updating production model'
mlflow.send_notification.assert_called_once_with(
metadata=metadata['metadata'],
@@ -291,7 +295,7 @@ async def test_update_production_model_error(mlflow):
message='Error updating production model test_model: Error updating production model',
block='update_production_model',
level=NotificationLevel.ERROR,
attachment_content=ANY
attachment_content=ANY,
)
else:
assert False, "No exception raised"
raise AssertionError('No exception raised')