Open pomodoren opened 3 years ago
"Endogenous" is a synonym for dependent variable/target/regressand (i.e. the variable you are predicting in a regression/classification model).
I surely did not think of that in this way ...
I thought that you were asking for understanding more the inter-dependability of the processes, and how the machine learning process might be corrupt because of error correlation with the independent. Thank you for clearing this.
Then maybe a structure like this would make more sense:
class Instance:
...
class PredictionModel:
# properties
target = db.Column(db.String, default='promotion')
bulk_id = Column(Integer, ForeignKey('bulk_check.id'))
...
# remove class method start_training
class BulkCheck:
# properties
...
pms = relationship("PredictionModel", back_populates="parent")
...
# methods
def start_training(self, target):
...
class Task:
....
I guess the next question would be: Do you want to connect the Tasks to the BulkCheck or to the PredictionModel? I guess I would connect them both - to ease the interaction later.
class Task:
...
pm_id = db.Column(db.Integer, db.ForeignKey("prediction_model.id"))
bulk_id = db.Column(db.Integer, db.ForeignKey("bulk_check.id"))
Working with weird words
I guess here its described the inter-dependency between promotion and network ability because having more network will eventually create more options for promotion, or the inter-dependency between promotion and competence.
Step by step