Regression instead of a classification

I have a question. How can I write a regression instead of a categorical classification like:
.classification(finalSchema.getIndexOfColumn(“STOCK”), 2)
The Column STOCK has an input like this: 78941079, 78941015, 78941011, 78941025…?

You tell it to use regression instead of classification:
https://deeplearning4j.org/api/latest/org/deeplearning4j/datasets/datavec/RecordReaderDataSetIterator.Builder.html

And then you set up your model to use an appropriate activation function (most likely Identity) and loss function (most likely MSE).

labelIndexFrom = 0, labelIndexTo = 3
RecordReaderDataSetIterator testIterator = new RecordReaderDataSetIterator.Builder(testRecordReader, batchSize)
.regression(0, 3)
.build

???

What is your question here? I’m not a compiler.

sorry I dont know what you mean with the appropriate activation function and the loss function? Could you give me any example?

There are a few examples on how to set up the model for regression in the examples repository:

Given how you’ve stated your original question, I understood it to be about loading the data for regression. This is a separate step from setting up your model for it.

ok thank you very much!