'Why there are difference with GridSearchCV vs programming GridSearch by myself
While i tried to tunning my parameter by using Grid Search, i found the different outcome with GridSearchCV in sklearn and GridSearch programming by myself.
I used same model and same parameters,optimizer, loss function but they show me the different best parameters so i want to know is there a error in my code or there are difference in GridSearchCV!
GridSearchCV code enter image description here
GridSearchCV outcome enter image description here
GridSearch programming by myself code enter image description here enter image description here
GridSearch programming by myself outcome enter image description here
I used all same settings(model,optimizer,loss_func,epoch,parameters. etc) but GridSeachCV in sklearn 's best parameters are batchsize=5,learning_rate=0.02 but GridSearch programming by myself 's best parameters are batchsize=5,learning_rate=0.05 Why they have difference? Please Teach me!!
Sources
This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.
Source: Stack Overflow
| Solution | Source |
|---|
