search

wang-fujin / SOHbenchmark

A code library and benchmark study on SOH estimation of lithium-ion battery
79 stars 16 forks source link

Why not compare with published models? #3

Open Leminhhuy2386 opened 5 months ago

Leminhhuy2386 commented 5 months ago

Dear Authors, I found this is an interesting paper with open sources code and datasets. I think our research community will get benefit from your works. I have one big question is that, why didn't you compare your results with previous published models from other authors (or at least get survey)? You provided detail source code with various deep learning models, and I think you are easily to compare them the others.

Thank you very much!

wang-fujin commented 5 months ago

There are two reasons why we did not compare with other models: First, our dataset is a newly public dataset, and no other published papers have used this dataset when we wrote our article. Second, the purpose of our article is to provide a benchmark, not to propose a method. In addition, the data preprocessing methods in other published articles are different from ours, making it difficult to compare them fairly. Therefore, we did not compare with other articles.