def update_train_task_by_id(train_job_id, train_task_id, is_check_train_terms, model_type, args):
"""
1. 根据字段状态更新训练状态和结果
2. 直接设置训练状态和结果
3. 模型上线状态更新（分类和抽取还不一样）
"""
train_job = TrainJobModel().get_by_id(train_job_id)
train_task = TrainTaskModel().get_by_id(train_task_id)
if is_check_train_terms: # 是否需要检查train_term的状态
_, training_terms = TrainTermTaskModel().get_by_filter(limit=99999, train_task_id=train_task_id,
train_term_status=int(StatusEnum.training))
_, failed_terms = TrainTermTaskModel().get_by_filter(limit=99999, train_task_id=train_task_id,
train_term_status=int(StatusEnum.fail))
if not training_terms:
# 没有处于训练中
if not failed_terms:
# 没有处于失败的
args["train_status"] = int(StatusEnum.success)
else:
args["train_status"] = int(StatusEnum.fail)
else:
args["train_status"] = int(StatusEnum.training)
else:
# no limit to set model_train_state=success/failed
if args["train_status"] == int(StatusEnum.online):
# validation
if train_task.train_status == StatusEnum.online:
abort(400, message="该模型已经上线")
if train_task.train_status != StatusEnum.success:
abort(400, message="只能上线训练成功的模型")
# send model train http request
service_url = _get("CLASSIFY_MODEL_ONLINE") if model_type == "classify" else _get("EXTRACT_MODEL_ONLINE")
resp = requests.post(f"{service_url}?model_version={train_task.model_version}")
if resp.status_code < 200 or resp.status_code >= 300:
abort(500, message=f"上线服务 <{service_url}> 出现错误: {resp.text}")
# find all online model under this doc_type_id
online_models = TrainTaskModel().get_by_doc_type_id(doc_type_id=train_job.doc_type_id, train_status=int(StatusEnum.online))
# unload online models
TrainTaskModel().bulk_update([train.train_task_id for train in online_models], train_status=int(StatusEnum.success))
# update train task
train_task = TrainTaskModel().update(train_task_id, **args)
session.commit()
return train_task
def create_evaluate_task_by_train_job_id(train_job_id, evaluate_task_name, evaluate_task_desc, mark_job_ids, doc_term_ids, doc_relation_ids, use_rule=0):
"""
如果后面要加重新训练的逻辑，这部分要改，不能根据train_job_id去创建评估任务，而是根据train_task_id，
目前先保留，因为目前train_job和train_task是一一对应，不会有影响
"""
# get correspondent train_job, doc_type, train_task, nlp_task by train_job_id
train_job = TrainJobModel().get_by_id(train_job_id)
doc_type = DocTypeModel().get_by_id(train_job.doc_type_id)
doc_term_list = DocTermModel().get_by_filter(limit=99999, doc_type_id=doc_type.doc_type_id)
doc_type.doc_term_list = doc_term_list
nlp_task = NlpTaskEnum(doc_type.nlp_task_id)
_, train_task_list = TrainTaskModel().get_by_filter(train_job_id=train_job_id)
train_task = train_task_list[0]
# create evaluate_task
evaluate_task = EvaluateTaskModel().create(evaluate_task_name=evaluate_task_name,
evaluate_task_desc=evaluate_task_desc,
train_task_id=train_task.train_task_id,
evaluate_task_status=int(StatusEnum.processing))
# bulk create evaluate m2m mark
evaluate_m2m_mark_list = [{"evaluate_task_id": evaluate_task.evaluate_task_id, "mark_job_id": _id} for _id in mark_job_ids]
EvaluateM2mMarkModel().bulk_create(evaluate_m2m_mark_list)
# push to evaluate redis queue
doc_term_ids = [str(t.doc_term_id) for t in RelationM2mTermModel().get_by_filter(limit=99999, doc_relation_ids=[int(rl) for rl in doc_relation_ids])]
push_evaluate_task_to_redis(nlp_task, evaluate_task, train_task, doc_type, mark_job_ids, doc_term_ids, doc_relation_ids, use_rule)
session.commit()
return evaluate_task
def update_train_task_by_model_version(model_version, is_check_train_terms, args):
train_task = TrainTaskModel().get_by_filter(model_version=model_version)[1][0]
if is_check_train_terms:
_, training_terms = TrainTermTaskModel().get_by_filter(limit=99999, train_task_id=train_task.train_task_id,
train_term_status=int(StatusEnum.training))
_, failed_terms = TrainTermTaskModel().get_by_filter(limit=99999, train_task_id=train_task.train_task_id,
train_term_status=int(StatusEnum.fail))
if not training_terms:
# 没有处于训练中
if not failed_terms:
# 没有处于失败的
args["train_status"] = int(StatusEnum.success)
else:
args["train_status"] = int(StatusEnum.fail)
else:
args["train_status"] = int(StatusEnum.training)
train_task = TrainTaskModel().update(train_task.train_task_id, **args)
session.commit()
return train_task
def create_train_task():
from app.model import TrainTaskModel
if TrainTaskModel().is_empty_table():
TrainTaskModel().create(app_id=1,
created_by=1,
train_model_name="train1",
train_status=int(StatusEnum.online),
train_job_id=1)
TrainTaskModel().create(app_id=1,
created_by=1,
train_model_name="train2",
train_status=int(StatusEnum.success),
train_job_id=2)
TrainTaskModel().create(app_id=1,
created_by=1,
train_model_name="train3",
train_status=int(StatusEnum.training),
train_job_id=3)
TrainTaskModel().create(app_id=1,
created_by=1,
train_model_name="train4",
train_status=int(StatusEnum.training),
train_job_id=4)
TrainTaskModel().create(app_id=1,
created_by=1,
train_model_name="train5",
train_status=int(StatusEnum.training),
train_job_id=5)
TrainTaskModel().create(app_id=1,
created_by=1,
train_model_name="train6",
train_status=int(StatusEnum.training),
train_job_id=6)
TrainTaskModel().create(app_id=1,
created_by=1,
train_model_name="train7",
train_status=int(StatusEnum.training),
train_job_id=7)
TrainTaskModel().create(app_id=1,
created_by=1,
train_model_name="train8",
train_status=int(StatusEnum.online),
train_job_id=8)
TrainTaskModel().create(app_id=1,
created_by=1,
train_model_name="train9",
train_status=int(StatusEnum.success),
train_job_id=9)
TrainTaskModel().create(app_id=1,
created_by=1,
train_model_name="train10",
train_status=int(StatusEnum.training),
train_job_id=10)
TrainTaskModel().create(app_id=1,
created_by=1,
train_model_name="train11",
train_status=int(StatusEnum.training),
train_job_id=11)
TrainTaskModel().create(app_id=1,
created_by=1,
train_model_name="train12",
train_status=int(StatusEnum.training),
train_job_id=12)
TrainTaskModel().create(app_id=1,
created_by=1,
train_model_name="train13",
train_status=int(StatusEnum.training),
train_job_id=13)
TrainTaskModel().create(app_id=1,
created_by=1,
train_model_name="train14",
train_status=int(StatusEnum.training),
train_job_id=14)
TrainTaskModel().create(app_id=1,
created_by=1,
train_model_name="train15",
train_status=int(StatusEnum.online),
train_job_id=15)
TrainTaskModel().create(app_id=1,
created_by=1,
train_model_name="train16",
train_status=int(StatusEnum.success),
train_job_id=16)
TrainTaskModel().create(app_id=1,
created_by=1,
train_model_name="train17",
train_status=int(StatusEnum.training),
train_job_id=17)
TrainTaskModel().create(app_id=1,
created_by=1,
train_model_name="train18",
train_status=int(StatusEnum.training),
train_job_id=18)
TrainTaskModel().create(app_id=1,
created_by=1,
train_model_name="train19",
train_status=int(StatusEnum.training),
train_job_id=19)
TrainTaskModel().create(app_id=1,
created_by=1,
train_model_name="train20",
train_status=int(StatusEnum.training),
train_job_id=20)
TrainTaskModel().create(app_id=1,
created_by=1,
train_model_name="train21",
train_status=int(StatusEnum.training),
train_job_id=21)
TrainTaskModel().create(app_id=1,
created_by=1,
train_model_name="train22",
train_status=int(StatusEnum.training),
train_job_id=22)
TrainTaskModel().create(app_id=1,
created_by=1,
train_model_name="train23",
train_status=int(StatusEnum.online),
train_job_id=23)
TrainTaskModel().create(app_id=1,
created_by=1,
train_model_name="train24",
train_status=int(StatusEnum.success),
train_job_id=24)
TrainTaskModel().create(app_id=1,
created_by=1,
train_model_name="train25",
train_status=int(StatusEnum.training),
train_job_id=25)
TrainTaskModel().create(app_id=1,
created_by=1,
train_model_name="train26",
train_status=int(StatusEnum.training),
train_job_id=26)
TrainTaskModel().create(app_id=1,
created_by=1,
train_model_name="train27",
train_status=int(StatusEnum.training),
train_job_id=27)
TrainTaskModel().create(app_id=1,
created_by=1,
train_model_name="train28",
train_status=int(StatusEnum.training),
train_job_id=28)
TrainTaskModel().create(app_id=1,
created_by=1,
train_model_name="train29",
train_status=int(StatusEnum.training),
train_job_id=29)
session.commit()
def get_evaluate_task_by_id(evaluate_task_id):
evaluate_task = EvaluateTaskModel().get_by_id(evaluate_task_id)
evaluate_task.train_job_id = TrainTaskModel().get_by_id(evaluate_task.train_task_id).train_job_id
return evaluate_task
def get_train_task_by_id(train_task_id) -> TrainTask:
train_task = TrainTaskModel().get_by_id(train_task_id)
train_task.mark_job_ids = [m2m.mark_job_id for m2m in TrainM2mMarkbModel().get_by_filter(limit=99999, train_job_id=train_task.train_job_id)]
return train_task
def get_train_task_list_by_train_job_id(train_job_id, order_by, order_by_desc, offset, limit) -> (int, [TrainTask]):
count, result = TrainTaskModel().get_by_filter(order_by=order_by, order_by_desc=order_by_desc, offset=offset, limit=limit, train_job_id=train_job_id)
return count, result
def delete_train_task_by_id(train_task_id):
TrainTaskModel().delete(train_task_id)
session.commit()