What's production model?
In result of Auto Model, there are "Model" and "Production Model". Those are slightly different. What's production model? In other words, what's the difference between Model and Production Model?
Thank a lot in advance.
Koichi
Thank a lot in advance.
Koichi
Find more posts tagged with
Sort by:
1 - 4 of
41
Dear Ingo,
Thank you for your answer.
>BTW, this type of informations is also explained in the in-product docs which are shown when you click on the (i)-icon in >the top right corner.
>Hope this helps,
I have checked already (i) icon. But, this does not help me. There is no information about this. And also no information in
Thank you for your answer.
>BTW, this type of informations is also explained in the in-product docs which are shown when you click on the (i)-icon in >the top right corner.
>Hope this helps,
I have checked already (i) icon. But, this does not help me. There is no information about this. And also no information in
And also no information in
Documentation should follow the updates, I believe.
Any way, thank you for your quick answer.
Regards,
Koichi
https://docs.rapidminer.com/latest/studio/guided/auto-model/Documentation should follow the updates, I believe.
Any way, thank you for your quick answer.
Regards,
Koichi
Another question regarding the production model,
If the production model uses all labeled data, how could we know if it overfits or not ?Sort by:
1 - 1 of
11
Hi,
The ‚model‘ is what is used to calculate the performance values on the labeled testing data sets and we show it so that the model is consistent with the ‘predictions’ table and the confusion matrix. However, this model is not using all the labeled data since it needs some of the labeled data for model validation. It’s main purpose is to estimate how well the model will work in production.
The ‘production model’ is using exactly the same preprocessing, feature sets, optimized parameters etc. - but is uses ALL labeled data for training. This is the model you should use in production and it makes use of all available information. It is also the model which would be deployed when you click on Deploy.
BTW, this type of informations is also explained in the in-product docs which are shown when you click on the (i)-icon in the top right corner.
Hope this helps,
Ingo
The ‚model‘ is what is used to calculate the performance values on the labeled testing data sets and we show it so that the model is consistent with the ‘predictions’ table and the confusion matrix. However, this model is not using all the labeled data since it needs some of the labeled data for model validation. It’s main purpose is to estimate how well the model will work in production.
The ‘production model’ is using exactly the same preprocessing, feature sets, optimized parameters etc. - but is uses ALL labeled data for training. This is the model you should use in production and it makes use of all available information. It is also the model which would be deployed when you click on Deploy.
BTW, this type of informations is also explained in the in-product docs which are shown when you click on the (i)-icon in the top right corner.
Hope this helps,
Ingo
The ‚model‘ is what is used to calculate the performance values on the labeled testing data sets and we show it so that the model is consistent with the ‘predictions’ table and the confusion matrix. However, this model is not using all the labeled data since it needs some of the labeled data for model validation. It’s main purpose is to estimate how well the model will work in production.
The ‘production model’ is using exactly the same preprocessing, feature sets, optimized parameters etc. - but is uses ALL labeled data for training. This is the model you should use in production and it makes use of all available information. It is also the model which would be deployed when you click on Deploy.
BTW, this type of informations is also explained in the in-product docs which are shown when you click on the (i)-icon in the top right corner.
Hope this helps,
Ingo