Trjtdtk

Why does Surtur say that Thor is Asgard's doom?

How do researchers send unsolicited emails asking for feedback on their works?

Are hand made posters acceptable in Academia?

Determine voltage drop over 10G resistors with cheap multimeter

"Marked down as someone wanting to sell shares." What does that mean?

Do people actually use the word "kaputt" in conversation?

Parts of mini page are not placed properly

What is the difference between something being completely legal and being completely decriminalized?

Does fire aspect on a sword, destroy mob drops?

What should be the ideal length of sentences in a blog post for ease of reading?

I got the following comment from a reputed math journal. What does it mean?

Exposing a company lying about themselves in a tightly knit industry (videogames) : Is my career at risk on the long run?

What do the positive and negative (+/-) transmit and receive pins mean on Ethernet cables?

God... independent

Do native speakers use "ultima" and "proxima" frequently in spoken English?

SQL PRINT vs SQL EXEC

What kind of footwear is suitable for walking in micro gravity environment?

What are the consequences of changing the number of hours in a day?

Can other pieces capture a threatening piece and prevent a checkmate?

Exit shell with shortcut (not typing exit) that closes session properly

Why are there no stars visible in cislunar space?

How do you justify more code being written by following clean code practices?

Would this string work as string?

Jem'Hadar, something strange about their life expectancy

Problem about tuning hyper-parametres

sklearn - overfitting problemStrategies for automatically tuning the hyper-parameters of deep learning modelsAutomated tuning of HyperparameterWhich parameters are hyper parameters in a linear regression?Hyper parameters and ValidationSetOverfitting problem in modelModel Selection with Oversampling/ Cross-Validation leads to similar test results in 2 approachesHyperparameter tuning for stacked modelsDuring a regression task, I am getting low R^2 values, but elementwise difference between test set and prediction values is hugeHyper-parameter tuning when you don't have an access to the test data

0

$begingroup$

I have tried GridSearchCV and BayesSearchCV for tuning my lightGBM algorithm (for binary classification).

I have used 10 iterations. And I have indicated scoring ="roc_auc"

In the first iteration, I have got:

best score (e.g :0.71...) 
and best param (e.g: max-depth: 10 , learning-rate: 0.17..., num-leave:175, n-estimators: 176, ....)

In the 10th iteration, I have got :

best score (e.g :0.72...) 
and best param (e.g: max-depth: 9 , learning-rate: 0.19..., num-leave:168, n-estimators: 172, ....)

Then I tried to train my LightGBM classifier with the 10th param (which supposed that it get the best score!!). I have got :

AUC : (0.7541.., 0.6467..)
Accuracy: 0.7338..
RMSE: 0.5216..

Then because I had some curiosity I have tried to train my classifier with (best param) of the First iteration (which is considered as worst score)!. I had surprised by the result that I have got:

AUC : (0.7545., 0.6592..)
Accuracy: 0.7332..
RMSE: 0.5152..

Because I have fixed previously scoring by roc-auc, Generally, I should get AUC in the 10th iteration better than the first iteration but I have got the contrary.

I have supposed that it is considered AUC train of 10th iteration 0.7541 as better than 0.7545 of the 1st iteration because of the overfitting but when I tried to check the 3rd and the 5th iteration I get on the 3rd : 0.7532 and in the 5th : 0.7548.

So I don't know what the best score in this algorithm means. And what is its role exactly if it gets values as the described situation. I had tried then many times with other tuning parameters but I have got the same case. I don't know where is the problem exactly.

machine-learning data-mining xgboost hyperparameter

share|improve this question

edited 6 mins ago

amal amal

asked 11 mins ago

amal amalamal amal

225

$endgroup$

add a comment | 

0

$begingroup$

I have tried GridSearchCV and BayesSearchCV for tuning my lightGBM algorithm (for binary classification).

I have used 10 iterations. And I have indicated scoring ="roc_auc"

In the first iteration, I have got:

best score (e.g :0.71...) 
and best param (e.g: max-depth: 10 , learning-rate: 0.17..., num-leave:175, n-estimators: 176, ....)

In the 10th iteration, I have got :

best score (e.g :0.72...) 
and best param (e.g: max-depth: 9 , learning-rate: 0.19..., num-leave:168, n-estimators: 172, ....)

Then I tried to train my LightGBM classifier with the 10th param (which supposed that it get the best score!!). I have got :

AUC : (0.7541.., 0.6467..)
Accuracy: 0.7338..
RMSE: 0.5216..

Then because I had some curiosity I have tried to train my classifier with (best param) of the First iteration (which is considered as worst score)!. I had surprised by the result that I have got:

AUC : (0.7545., 0.6592..)
Accuracy: 0.7332..
RMSE: 0.5152..

Because I have fixed previously scoring by roc-auc, Generally, I should get AUC in the 10th iteration better than the first iteration but I have got the contrary.

I have supposed that it is considered AUC train of 10th iteration 0.7541 as better than 0.7545 of the 1st iteration because of the overfitting but when I tried to check the 3rd and the 5th iteration I get on the 3rd : 0.7532 and in the 5th : 0.7548.

So I don't know what the best score in this algorithm means. And what is its role exactly if it gets values as the described situation. I had tried then many times with other tuning parameters but I have got the same case. I don't know where is the problem exactly.

machine-learning data-mining xgboost hyperparameter

share|improve this question

edited 6 mins ago

amal amal

asked 11 mins ago

amal amalamal amal

225

$endgroup$

add a comment | 

$begingroup$

I have tried GridSearchCV and BayesSearchCV for tuning my lightGBM algorithm (for binary classification).

I have used 10 iterations. And I have indicated scoring ="roc_auc"

In the first iteration, I have got:

best score (e.g :0.71...) 
and best param (e.g: max-depth: 10 , learning-rate: 0.17..., num-leave:175, n-estimators: 176, ....)

In the 10th iteration, I have got :

best score (e.g :0.72...) 
and best param (e.g: max-depth: 9 , learning-rate: 0.19..., num-leave:168, n-estimators: 172, ....)

Then I tried to train my LightGBM classifier with the 10th param (which supposed that it get the best score!!). I have got :

AUC : (0.7541.., 0.6467..)
Accuracy: 0.7338..
RMSE: 0.5216..

Then because I had some curiosity I have tried to train my classifier with (best param) of the First iteration (which is considered as worst score)!. I had surprised by the result that I have got:

AUC : (0.7545., 0.6592..)
Accuracy: 0.7332..
RMSE: 0.5152..

Because I have fixed previously scoring by roc-auc, Generally, I should get AUC in the 10th iteration better than the first iteration but I have got the contrary.

I have supposed that it is considered AUC train of 10th iteration 0.7541 as better than 0.7545 of the 1st iteration because of the overfitting but when I tried to check the 3rd and the 5th iteration I get on the 3rd : 0.7532 and in the 5th : 0.7548.

So I don't know what the best score in this algorithm means. And what is its role exactly if it gets values as the described situation. I had tried then many times with other tuning parameters but I have got the same case. I don't know where is the problem exactly.

machine-learning data-mining xgboost hyperparameter

share|improve this question

edited 6 mins ago

amal amal

asked 11 mins ago

amal amalamal amal

225

$endgroup$

best score (e.g :0.71...) 
and best param (e.g: max-depth: 10 , learning-rate: 0.17..., num-leave:175, n-estimators: 176, ....)

best score (e.g :0.72...) 
and best param (e.g: max-depth: 9 , learning-rate: 0.19..., num-leave:168, n-estimators: 172, ....)

AUC : (0.7541.., 0.6467..)
Accuracy: 0.7338..
RMSE: 0.5216..

AUC : (0.7545., 0.6592..)
Accuracy: 0.7332..
RMSE: 0.5152..

share|improve this question

edited 6 mins ago

amal amal

asked 11 mins ago

amal amalamal amal

225

share|improve this question

edited 6 mins ago

amal amal

asked 11 mins ago

amal amalamal amal

225

asked 11 mins ago

amal amalamal amal

225

asked 11 mins ago

amal amalamal amal

225