0
$\begingroup$

I'm trying to do prediction on capacity column, however each data point consist of more data.

Each data point represent a cycle data. Each cycle has a capacity. Each cycle runs for some time duration, and in that duration some data is collected over which capacity is dependant

I tried exploding the dataset and copying the capacity values to each row, but that shouldn't be the case because each row will get different capacity predicted. Is there a way to train such kind of dataset?

enter image description here

$\endgroup$
  • $\begingroup$ Question: are the lists in the columns of equal length? Why don‘t you simply make $x$ columns out of a (now) single column? $\endgroup$ – Peter Aug 1 '19 at 18:32
  • $\begingroup$ No. They are not equal columns $\endgroup$ – Bhaskar Dhariyal Aug 1 '19 at 18:35
  • $\begingroup$ is there a logical structure you can use? You need to have the same thing in one column anyway $\endgroup$ – Peter Aug 1 '19 at 18:36
0
$\begingroup$

If your list are of fixed size you can separate them in different columns.e.g. load1,load2... If thats not the case, you need to define some statistics for the cycles like average load, max_load, min_load..

| improve this answer | |
$\endgroup$
0
$\begingroup$

What's the difference between different rows. In addition to Nikita's answer, you may want to consider the temporal correlations.

| improve this answer | |
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.