@@ -87,3 +87,11 @@ ENV/

 # Rope project settings

 .ropeproject

+

+# MacOS files

+*.DS_Store

+

+# asv files

+benchmarks/html/

+benchmarks/results/

+.asv

\ No newline at end of file

@@ -1,4 +1,5 @@

 # Tools for adaptive parallel evaluation of functions

+[![asv](http://img.shields.io/badge/benchmarked%20by-asv-green.svg?style=flat)](https://gitlab.kwant-project.org/qt/adaptive/)

 ## Development

new file mode 100644

@@ -0,0 +1,44 @@

+# adaptive benchmarks

+

+Benchmarking adaptive with Airspeed Velocity.

+

+## Usage

+

+Airspeed Velocity manages building and Python conda environments by itself,

+unless told otherwise. To run the benchmarks, you do not need to install a

+development version of adaptive to your current Python environment.

+

+Run ASV commands (record results and generate HTML):

+

+```bash

+cd benchmarks

+asv run --skip-existing-commits --steps 10 ALL

+asv publish

+asv preview

+```

+

+More on how to use ``asv`` can be found in `ASV documentation`_

+Command-line help is available as usual via ``asv --help`` and

+``asv run --help``.

+

+

+## Writing benchmarks

+

+See [`ASV documentation`](https://asv.readthedocs.io/) for basics on how to write benchmarks.

+

+Some things to consider:

+

+- The benchmark suite should be importable with any adaptive version.

+

+- The benchmark parameters etc. should not depend on which adaptive version

+  is installed.

+

+- Try to keep the runtime of the benchmark reasonable.

+

+- Prefer ASV's ``time_`` methods for benchmarking times rather than cooking up

+  time measurements via ``time.clock``, even if it requires some juggling when

+  writing the benchmark.

+

+- Preparing arrays etc. should generally be put in the ``setup`` method rather

+  than the ``time_`` methods, to avoid counting preparation time together with

+  the time of the benchmarked operation.

new file mode 100644

@@ -0,0 +1,25 @@

+{

+    "version": 1,

+    "project": "adaptive",

+    "project_url": "https://gitlab.kwant-project.org/qt/adaptive",

+    "repo": "..",

+    "dvcs": "git",

+    "environment_type": "conda",

+    "install_timeout": 600,

+    "show_commit_url": "https://gitlab.kwant-project.org/qt/adaptive/commit/",

+    "pythons": ["3.6"],

+    "conda_channels": ["conda-forge"],

+    "matrix": {

+        "numpy": ["1.13"],

+        "holoviews": ["1.9.1"],

+        "scipy": ["0.19.1"],

+        "ipyparallel": ["6.0.2"],

+        "sortedcontainers": ["1.5.7"],

+    },

+    "benchmark_dir": "benchmarks",

+    "env_dir": "env",

+    "results_dir": "results",

+    "html_dir": "html",

+    "hash_length": 8,

+    "wheel_cache_size": 2

+}

new file mode 100644

@@ -0,0 +1 @@

+

new file mode 100644

@@ -0,0 +1,47 @@

+import adaptive

+

+import numpy as np

+import random

+

+

+offset = random.uniform(-0.5, 0.5)

+

+def f_1d(x, offset=offset):

+    a = 0.01

+    return x + a**2 / (a**2 + (x - offset)**2)

+

+

+def f_2d(xy):

+    x, y = xy

+    a = 0.2

+    return x + np.exp(-(x**2 + y**2 - 0.75**2)**2/a**4)

+

+

+class TimeLearner1D:

+    def setup(self):

+        self.learner = adaptive.Learner1D(f_1d, bounds=(-1, 1))

+

+    def time_run(self):

+        for _ in range(1000):

+            points, _ = self.learner.choose_points(1)

+            self.learner.add_data(points, map(f_1d, points))

+

+

+class TimeLearner2D:

+    def setup(self):

+        self.learner = adaptive.Learner2D(f_2d, bounds=[(-1, 1), (-1, 1)])

+        self.xs = np.random.rand(50**2, 2)

+        self.ys = np.random.rand(50**2)

+

+    def time_run(self):

+        for _ in range(50**2):

+            points, _ = self.learner.choose_points(1)

+            self.learner.add_data(points, map(f_2d, points))

+

+    def time_choose_points(self):

+        for _ in range(50**2):

+            self.learner.choose_points(1)

+

+    def time_add_point(self):

+        for x, y in zip(self.xs, self.ys):

+            self.learner.add_point(x, y)