0
$\begingroup$

The available tutorials are most focused on time series prediction. I am wondering how shall we prepare the input data when it is an imbalanced data? Here is how data looks like.

|---------|--------------|-----------|-----------|
|    id   |  timestamp   |  feature1 |  feature2 |
|---------|--------------|-----------|-----------|
|  1      |      1       |    0.54   |     2     |
|---------|--------------|-----------|-----------|
|  1      |      2       |    0.67   |     3     |
|---------|--------------|-----------|-----------|
|  1      |      3       |    0.32   |     3     |
|------------------------|-----------|-----------|
|  1      |      4       |    0.43   |     4     |
|------------------------|-----------|-----------|
|  2      |      1       |    0.25   |     1     |
|------------------------|-----------|-----------|
|  2      |      2       |    0.93   |     2     |
|------------------------|-----------|-----------|

I am aware of this post that discusses LSTM for balanced data. But how about an imbalanced dataset?

$\endgroup$

Your Answer

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

Browse other questions tagged or ask your own question.