def __init__(self, config: Configuration, tracing_enabled: bool = False):
"""Set up required properties to run training or prediction.
:param config: configuration provided via yaml or environment variables
"""
self.config = config
# self.feedback_strategy = feedback_strategy
self.pipeline = DetectorPipeline()
self.tracing_enabled = tracing_enabled
def __init__(self,
config: Configuration,
feedback_strategy: FeedbackStrategy = None,
tracing_enabled: bool = False):
"""Set up required properties to run training or prediction.
:param config: configuration provided via yaml or environment variables
:param feedback_strategy: a function that runs to improve the feedback of system
"""
self.config = config
self.feedback_strategy = feedback_strategy
self.pipeline = DetectorPipeline()
self.tracing_enabled = tracing_enabled
def test_valid_command(self):
"""Test case for validating that when we train a model and add it to task queue that it will run."""
mgr = DetectorPipeline()
class mock_func(AbstractCommand):
def execute(self):
print("Test")
mock = mock_func()
mgr.add_steps(mock)
self.assertEqual(len(mgr), TASKS_IN_QUEUE)
self.assertNotEqual(mgr.count, TASKS_IN_QUEUE)
mgr.execute_steps()
self.assertEqual(mgr.count, TASKS_IN_QUEUE)
mgr.clear()
def test_train_command(self):
"""Test case for validating that when we train a model and add it to task queue that it will run."""
mgr = DetectorPipeline()
config = Configuration()
config.STORAGE_DATASOURCE = "local"
config.STORAGE_DATASINK = "stdout"
config.LS_INPUT_PATH = "validation_data/Hadoop_2k.json"
storage_adapter = SomStorageAdapter(config=config, feedback_strategy=None)
model_adapter = SomModelAdapter(storage_adapter)
tc = SomTrainJob(node_map=2, model_adapter=model_adapter)
mgr.add_steps(tc)
self.assertEqual(len(mgr), TASKS_IN_QUEUE)
self.assertNotEqual(mgr.count, TASKS_IN_QUEUE)
mgr.execute_steps()
self.assertEqual(mgr.count, TASKS_IN_QUEUE)
mgr.clear()
def test_invalid_command(self):
"""Test case for validating that when we train a model and add it to task queue that it will run."""
mgr = DetectorPipeline()
class mock_func:
def execute(self):
print("Test")
mock = mock_func()
with self.assertRaises(TypeError) as context:
mgr.add_steps(mock)
mgr.clear()
def pipeline():
"""Providing pipeline that clears up history of task runs."""
pipeline = DetectorPipeline()
yield pipeline
pipeline.clear()
class Facade:
"""For external interface for integration different adapters for custom models and training logic."""
def __init__(self, config: Configuration, tracing_enabled: bool = False):
"""Set up required properties to run training or prediction.
:param config: configuration provided via yaml or environment variables
"""
self.config = config
# self.feedback_strategy = feedback_strategy
self.pipeline = DetectorPipeline()
self.tracing_enabled = tracing_enabled
@staticmethod
def create_tracer(service):
"""Initialize tracer for open tracing.
:param service: service name to use for tracing this application.
:return: Tracer(opentracing.Tracer)
"""
logging.getLogger('').handlers = []
logging.basicConfig(format='%(message)s', level=logging.DEBUG)
config = Config(
config={
'sampler': {
'type': 'const',
'param': 1,
},
'logging': True,
},
service_name=service,
)
return config.initialize_tracer()
def run(self, single_run: bool = False):
"""Run train and inference and main event loop.
:param single_run: if this is set to TRUE then we exit loop after first iteration.
:return: None
"""
exit = False
while exit is False:
try:
jobs = DetectorPipelineCatalog(config=self.config,
job="sompy.train.inference")
self.pipeline = jobs.get_pipeline()
self.start_job()
logging.info("Job ran succesfully")
except EmptyDataSetException as e:
logging.debug(e)
finally:
time.sleep(5)
exit = single_run
def start_job(self):
"""Start job to run all steps in workflow."""
job_id = uuid.uuid4()
logging.info("Executing job: {}".format(job_id))
tracer = self.create_tracer('log-anomaly-detection')
if self.tracing_enabled:
logging.info("Tracing enabled")
with tracer.start_span('anomaly_facade_run') as span:
span.set_tag('job_id', job_id)
with span_in_context(span):
self.pipeline.execute_steps(tracer=tracer)
else:
self.pipeline.execute_steps()
def train(self):
"""Run training of model and provides size of map.
:return: None
"""
jobs = DetectorPipelineCatalog(config=self.config, job="sompy.train")
self.pipeline = jobs.get_pipeline()
self.start_job()
def infer(self):
"""Run inference of model.
:return: None
"""
jobs = DetectorPipelineCatalog(config=self.config,
job="sompy.inference")
self.pipeline = jobs.get_pipeline()
self.start_job()