search

DatrikIntelligence / Stacked-DCNN-RUL-PHM21

Source code of the paper "A stacked DCNN to predict the RUL of a turbofan engine", third place ranked in the PHM21 data challenge.
GNU General Public License v3.0
72 stars 9 forks source link

May I get trained models? #2

Open run-si opened 1 year ago

run-si commented 1 year ago

Dear author, I would like to obtain the trained model of your project to be used as part of the undergraduate graduation design paper. If you have the trained model, can you send it to me by email? My email address is suxianningcm@Gmail.com. This will be of great help to me.

dasolma commented 1 year ago

Hello, I have uploaded the trained models to the folder models in this repository.

Thanks by the interest in this work.

HanealO commented 1 year ago

Thank you very much for your efforts! It helped me a lot

HanealO commented 1 year ago

If I were to directly predict the remaining life, which model file should be called？

dasolma commented 1 year ago

Each model file is trained on a particular set of data, so it's hard to say which one will perform the best on your specific problem without more information. However, you can try using any of them and compare their predictions with the actual remaining life of the object to evaluate their performance.

Alternatively, you could combine the predictions of multiple models by taking their average or using more sophisticated ensemble methods such as weighted averaging or stacking. This can often lead to better performance and more robust predictions, as the different models may capture different aspects of the underlying relationship between the input features and the remaining life.