Roman Yurchak

Conditionally skip unnecessary CI steps

3

This is an attempt to save CI resources by not not running build steps when they are not needed. Current rules are, 1. do not build core (and packages) if...

Refactor common tests in tslearn/tslearn/tests/test_estimators.py

1

Currently `tslearn/tests/test_estimators.py` redefine a custom `check_estimator` and then monkeypatch some of the tests in `sklearn.utils.estimator_checks` to work with time series data. In the latest version scikit-learn introduced `parametrize_with_checks` which should...

Joblib pickling performance for pandas DataFrame

11

Related to https://github.com/joblib/joblib/issues/467 Below are a few fairly limited benchmarks, that illustrate the serialization / de-serialization performance of `pandas.DataFrame` with joblib, ```py import pandas as pd from sklearn.externals import joblib...

Sample selection bias and up/down-sampling

5

It's a bit of an open-ended question. In my understanding up/down-sampling the input data depending on the target class is equivalent to having a dataset with sample selection bias. The...

Scalability of feature extraction

8

Thank you for this package! Somewhat related to #356 , as far as I understand currently the input format is a pandas.DataFrame, then data is aggregated by the `date` and...

Python 3 support

5

Thanks for this library! Just a reminder that most major scientific Python packages have dropped Python 2 support now, so only supporting Python 2 is quite problematic. Related to https://github.com/weld-project/weld/issues/110...

[RFC] Data processing pipelines

5

In line with code reorganization in issue #33 , I was wondering what's your opinion on data pipelines. I could be wrong but, I looks like what the current OridnaryKrigging...

Scaling to large datasets

1

This issues aims to discuss scaling of PyKridge to large datsets (which could impact, for instance, the optimization approaches in issue #35). Here are approximate (and possibly inaccurate) time complexity...

Performance optimization with LU decomposition

2

Making this @bsmurphy comment a separate issue, > BTW, @rth (and anyone else who thinks about matrices, I don't think about them that much admittedly), I've been wondering if we...

Benchmark L-BFGS

12

Thanks for this very nice package! On the general topic of benchmarks, I would be quite interested to know how the L-BFGS implementation in argmin would compare to, - [liblbfgs](http://chokkan.org/software/liblbfgs/)...