question about labels of inference (predict) data and thread

[liujing0055]

Hi，Auto-sklearm Team:
After scanning the code， I found that the data were divided into two parts,training(fit) and inference(predict) when calling the fit function as shown below.
automl.fit(data, label,feat_type=feature_types)
But I could not find the variable of label of inference(predict),Could you please tell me where I can find the variable?
Another question is about thread, is there any possible way to run the program with only one thread?
Thanks.

[mfeurer]

But I could not find the variable of label of inference(predict),Could you please tell me where I can find the variable?

The data is split internally and there is no easy way accessing the split. What do you need it for?

Another question is about thread, is there any possible way to run the program with only one thread?

Yes, please check the sequential usage example.

[liujing0055]

I need to cheek the difference between my own designed algorithm and sklearn's algorithm.Could you please tell me the reason why not easy?

[mfeurer]

Because there is no interface to access the data.

[mfeurer]

Good news, this is actually possible using scikit-learn's internal splitter and we finally have an example for doing so: https://automl.github.io/auto-sklearn/development/examples/40_advanced/example_resampling.html#scikit-learn-splitter-objects

The example is currently only available on the documentation of the development branch, but the example also works on the master branch.