coco-experiment 2.7.4__tar.gz → 2.8.0__tar.gz

{coco_experiment-2.7.4 → coco_experiment-2.8.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: coco-experiment
-Version: 2.7.4
+Version: 2.8.0
 Summary: Benchmarking framework for all types of black-box optimization algorithms.
 Project-URL: Homepage, https://github.com/numbbo/coco-experiment
 Project-URL: Documentation, https://numbbo.github.io/coco-doc/apidocs/cocoex/
@@ -22,8 +22,15 @@ Requires-Dist: numpy>=1.24.0
 Description-Content-Type: text/markdown
 
 <h1 align="center">
-<img src="https://raw.githubusercontent.com/numbbo/coco-experiment/main/logo/coco-ex-300.webp" width="300">
-</h1><br>
+    <table border="0">
+  <td>
+      <img src="https://raw.githubusercontent.com/numbbo/coco-experiment/main/logo/coco-ex-300.webp">          
+  </td>
+  <td>
+      COmparing Continuous Optimisers COCO Experiment
+</td>
+</table>
+</h1>
 
 [![version](https://img.shields.io/pypi/v/coco-experiment?color=blue)](https://pypi.org/project/coco-experiment)
 [![license](https://img.shields.io/pypi/l/coco-experiment)](https://pypi.org/project/coco-experiment)
@@ -31,11 +38,20 @@ Description-Content-Type: text/markdown
 [![download](https://img.shields.io/pypi/dm/coco-experiment.svg?label=PyPI%20downloads)][pypi]
 [![paper](https://img.shields.io/badge/DOI-10.1080%2F10556788.2020.1808977-blue)][paper]
 
-*coco-experiment* is a Python module implementing the experimental part of [COCO: A Platform for Comparing Continuous Optimizers in a Black-Box Setting][paper].
+The `coco-experiment` Python module implements the experimental part of [COCO: A Platform for Comparing Continuous Optimizers in a Black-Box Setting][paper].
+It provides several benchmark function suites and the data logging.
 
-- **Website:** https://numbbo.github.io/coco/
-- **Source code:** https://github.com/numbbo/coco-experiment
-- **Bug reports:** https://github.com/numbbo/coco-experiment/issues
+## Documentation
+
+The main documentation pages for the `coco-experiment` package `cocoex` can be found at
+
+- [getting-started](https://numbbo.it/getting-started#experiment)
+- [API documentation](https://numbbo.github.io/coco-doc/apidocs/cocoex)
+- [issue tracker and bug reports](https://github.com/numbbo/coco-experiment/issues)
+- [source code page](https://github.com/numbbo/coco-experiment) on GitHub
+- [COCO home page](https://coco-platform.org)
+
+Below are installation instruction and some usage hints.
 
 ## Installation
 
@@ -45,12 +61,22 @@ The easiest way to install coco-experiment is using `pip` (preferably in a virtu
 pip install -U coco-experiment
 ```
 
+For the difficult way [look at this page][codepython] and start with `git clone https://github.com/numbbo/coco-experiment.git`.
+
 ## Usage
 
-Similar to scikit-learn, the name of the coco-experiment module is _not_ `coco-experiment` but rather `cocoex`.
-To get started, take a look at the [simple example experiment][expy].
+Similar to `scikit-learn`, _the module name_ (`coco-experiment`) **is not the import name**, hence usage starts with
+
+```python
+import cocoex
+```
+
+To get started, take a look at the [getting-started] page and at this [simple example experiment][expy]. To run an experiment in practice, take a look at the [complete example experiment][expy2].
 
 [build]: https://github.com/numbbo/coco-experiment/actions?query=branch%3Amain
 [pypi]: https://pypi.org/project/coco-experiment/
 [paper]: https://doi.org/10.1080/10556788.2020.1808977
-[expy]: https://github.com/numbbo/coco-experiment/blob/main/build/python/example/example_experiment_complete.py
+[getting-started]: https://coco-platform.org/getting-started/#experiment
+[expy]: https://github.com/numbbo/coco-experiment/blob/main/build/python/example/example_experiment_simple.py
+[expy2]: https://github.com/numbbo/coco-experiment/blob/main/build/python/example/example_experiment_complete.py
+[codepython]: https://github.com/numbbo/coco-experiment/tree/main/build

coco_experiment-2.8.0/README.md

@@ -0,0 +1,59 @@
+<h1 align="center">
+    <table border="0">
+  <td>
+      <img src="https://raw.githubusercontent.com/numbbo/coco-experiment/main/logo/coco-ex-300.webp">          
+  </td>
+  <td>
+      COmparing Continuous Optimisers COCO Experiment
+</td>
+</table>
+</h1>
+
+[![version](https://img.shields.io/pypi/v/coco-experiment?color=blue)](https://pypi.org/project/coco-experiment)
+[![license](https://img.shields.io/pypi/l/coco-experiment)](https://pypi.org/project/coco-experiment)
+[![build](https://img.shields.io/github/actions/workflow/status/numbbo/coco-experiment/build.yml?branch=main)][build]
+[![download](https://img.shields.io/pypi/dm/coco-experiment.svg?label=PyPI%20downloads)][pypi]
+[![paper](https://img.shields.io/badge/DOI-10.1080%2F10556788.2020.1808977-blue)][paper]
+
+The `coco-experiment` Python module implements the experimental part of [COCO: A Platform for Comparing Continuous Optimizers in a Black-Box Setting][paper].
+It provides several benchmark function suites and the data logging.
+
+## Documentation
+
+The main documentation pages for the `coco-experiment` package `cocoex` can be found at
+
+- [getting-started](https://numbbo.it/getting-started#experiment)
+- [API documentation](https://numbbo.github.io/coco-doc/apidocs/cocoex)
+- [issue tracker and bug reports](https://github.com/numbbo/coco-experiment/issues)
+- [source code page](https://github.com/numbbo/coco-experiment) on GitHub
+- [COCO home page](https://coco-platform.org)
+
+Below are installation instruction and some usage hints.
+
+## Installation
+
+The easiest way to install coco-experiment is using `pip` (preferably in a virtual environment):
+
+```python
+pip install -U coco-experiment
+```
+
+For the difficult way [look at this page][codepython] and start with `git clone https://github.com/numbbo/coco-experiment.git`.
+
+## Usage
+
+Similar to `scikit-learn`, _the module name_ (`coco-experiment`) **is not the import name**, hence usage starts with
+
+```python
+import cocoex
+```
+
+To get started, take a look at the [getting-started] page and at this [simple example experiment][expy]. To run an experiment in practice, take a look at the [complete example experiment][expy2].
+
+[build]: https://github.com/numbbo/coco-experiment/actions?query=branch%3Amain
+[pypi]: https://pypi.org/project/coco-experiment/
+[paper]: https://doi.org/10.1080/10556788.2020.1808977
+[getting-started]: https://coco-platform.org/getting-started/#experiment
+[expy]: https://github.com/numbbo/coco-experiment/blob/main/build/python/example/example_experiment_simple.py
+[expy2]: https://github.com/numbbo/coco-experiment/blob/main/build/python/example/example_experiment_complete.py
+[codepython]: https://github.com/numbbo/coco-experiment/tree/main/build

{coco_experiment-2.7.4 → coco_experiment-2.8.0}/src/cocoex/_version.py

@@ -1,7 +1,7 @@
 # file generated by fabricate
 # Do not change, do not track in version control!
 
-__version__ = "2.7.4"
+__version__ = "2.8.0"
 version = __version__
 
 if __name__ == "__main__":

{coco_experiment-2.7.4 → coco_experiment-2.8.0}/src/cocoex/coco.c

@@ -31526,7 +31526,7 @@ void coco_archive_free(coco_archive_t *archive) {
  * don't change, don't track in version control!
  */
 
-const char *coco_version = "2.7.4";
+const char *coco_version = "2.8.0";
 #line 1 "coco_runtime_c.c"
 /**
  * @file coco_runtime_c.c

{coco_experiment-2.7.4 → coco_experiment-2.8.0}/src/cocoex/noiser.py

@@ -74,12 +74,14 @@ class Noisifier:
     https://github.com/numbbo/coco-experiment/blob/main/build/python/example/example_experiment_complete.py
 
     """
-    def __init__(self, p_add=0.2, p_subtract=0.0, p_epsilon=0, epsilon=1e-4,
+    p_add_single_default = 0.2
+    def __init__(self, p_add=None, p_subtract=0.0, p_epsilon=0.0, epsilon=1e-4,
                  rands=(rand, randn, randc)):
         """constructor with 4 optional parameters for the noise model,
 
         `p_add`: probability for adding a positive heavy tail random value
-        making the solution look worse than it is.
+        making the solution look worse than it actually is. The default is
+        0.2 iff ``p_subtract == p_epsilon == 0`` and 0.0 otherwise.
 
         `p_subtract`: probability for subtracting a positive heavy tail
         random value making the solution look better than it is.
@@ -88,6 +90,11 @@ class Noisifier:
 
         `epsilon`: standard deviation of the Gaussian random value
         """
+        if p_add is None:
+            if p_subtract == p_epsilon == 0:
+                p_add = Noisifier.p_add_single_default  # == 0.2
+            else:
+                p_add = 0.0
         self._params = {k: v for k, v in locals().items() if k != 'self'}
         if sum(self._params[p] for p in ('p_subtract', 'p_add')) > 1:
             warnings.warn("p_subtract={0} + p_add={1} > 1, hence p_add is interpreted"

{coco_experiment-2.7.4 → coco_experiment-2.8.0}/src/cocoex/utilities.py

@@ -722,6 +722,10 @@ class ExperimentRepeater:
             return [ii for ii in d if not d[ii]]  # list of instance numbers with no data
         return []
     @property
+    def sweep(self):
+        """current sweep/repetition number (starts with 1)"""
+        return self._sweeps
+    @property
     def n_problems(self):
         """current number of tracked problems (different ``(id_function, dimension)``)"""
         return sum(len(self.trials(k)) > 0 for k in self._data)

coco_experiment-2.7.4/README.md

@@ -1,33 +0,0 @@
-<h1 align="center">
-<img src="https://raw.githubusercontent.com/numbbo/coco-experiment/main/logo/coco-ex-300.webp" width="300">
-</h1><br>
-
-[![version](https://img.shields.io/pypi/v/coco-experiment?color=blue)](https://pypi.org/project/coco-experiment)
-[![license](https://img.shields.io/pypi/l/coco-experiment)](https://pypi.org/project/coco-experiment)
-[![build](https://img.shields.io/github/actions/workflow/status/numbbo/coco-experiment/build.yml?branch=main)][build]
-[![download](https://img.shields.io/pypi/dm/coco-experiment.svg?label=PyPI%20downloads)][pypi]
-[![paper](https://img.shields.io/badge/DOI-10.1080%2F10556788.2020.1808977-blue)][paper]
-
-*coco-experiment* is a Python module implementing the experimental part of [COCO: A Platform for Comparing Continuous Optimizers in a Black-Box Setting][paper].
-
-- **Website:** https://numbbo.github.io/coco/
-- **Source code:** https://github.com/numbbo/coco-experiment
-- **Bug reports:** https://github.com/numbbo/coco-experiment/issues
-
-## Installation
-
-The easiest way to install coco-experiment is using `pip` (preferably in a virtual environment):
-
-```python
-pip install -U coco-experiment
-```
-
-## Usage
-
-Similar to scikit-learn, the name of the coco-experiment module is _not_ `coco-experiment` but rather `cocoex`.
-To get started, take a look at the [simple example experiment][expy].
-
-[build]: https://github.com/numbbo/coco-experiment/actions?query=branch%3Amain
-[pypi]: https://pypi.org/project/coco-experiment/
-[paper]: https://doi.org/10.1080/10556788.2020.1808977
-[expy]: https://github.com/numbbo/coco-experiment/blob/main/build/python/example/example_experiment_complete.py