=========================
Comparing implementations
=========================
The following benchmarks compare different implementations
of the same algorithm.
.. contents::
:local:
Benchmarks around scikit-learn
==============================
.. toctree::
:maxdepth: 1
scikit-learn/gridsearch_cache
Some benchmarks are available on :epkg:`PolynomialFeatures`
at `Benchmark of PolynomialFeatures + partialfit of SGDClassifier
`_.
Benchmarks of toy implementations in C++, Python
================================================
The following benchmarks were implemented in other
repositories. The first one measures differents way to write
the dot product in C++ using a couple of processors
optimization such as branching or
`AVX `_
instructions.
* `Measures branching in C++ from python
`_
* `Measures a vector sum with different accumulator type
`_
The second one looks into the implementation
of a logistic regression with python, C++ or C++ optimization
provided by other libraries.
* `Optimisation de code avec cffi, numba, cython
`_
The next benchmark compares the gain obtained by
playing a criterion for decision tree regressor.
* `Custom Criterion for DecisionTreeRegressor `_