Merge pull request #36 from gmrukwa/develop

Release v2.3.14

- fix seeding issues in GAP
- add  CLI for Dunn's index version of DiviK
- more permissive tests - 1 cluster of misidentification allowed
- compatibility layer for gin-config

.bettercodehub.yml

@@ -1,3 +1,3 @@
-component_depth: 3
+component_depth: 2
 languages:
 - python

.github/workflows/deploy.yml

@@ -12,7 +12,7 @@ on:
 env:
   MAJOR: ${{ 2 }}
   MINOR: ${{ 3 }}
-  FIXUP: ${{ 13 }}
+  FIXUP: ${{ 14 }}
   PACKAGE_INIT_FILE: ${{ 'divik/__init__.py' }}
   DOCKER_REPO: ${{ 'gmrukwa/divik' }}
   IS_ALPHA: ${{ github.event_name == 'pull_request' }}

README.md

@@ -13,8 +13,9 @@ Python implementation of Divisive iK-means (DiviK) algorithm.
 
 > This section will be further developed soon.
 
-1) [`divik`](divik/_cli/divik.md) - runs DiviK in one of many scenarios
-2) [`kmeans`](divik/_cli/auto_kmeans.md) - runs K-means
+1) [`divik`](divik/_cli/divik.md) - runs DiviK in GAP-only scenario
+2) [`dunn-divik`](dunn-divik/_cli/dunn-divik.md) - runs DiviK in GAP & Dunn scenario
+2) [`kmeans`](divik/_cli/auto_kmeans.md) - runs K-means with GAP statistic
 3) `linkage` - runs agglomerative clustering
 4) [`inspect`](divik/_cli/inspect.md) - visualizes DiviK result
 5) `visualize` - generates `.png` file with visualization of clusters for 2D
@@ -39,7 +40,7 @@ docker pull gmrukwa/divik
 To install specific version, you can specify it in the command, e.g.:
 
 ```bash
-docker pull gmrukwa/divik:2.3.13
+docker pull gmrukwa/divik:2.3.14
 ```
 
 ## Python package
@@ -59,9 +60,19 @@ pip install divik
 or any stable tagged version, e.g.:
 
 ```bash
-pip install divik==2.3.13
+pip install divik==2.3.14
 ```
 
+If you want to have compatibility with
+[`gin-config`](https://github.com/google/gin-config), you can install
+necessary extras with:
+
+```bash
+pip install divik[gin]
+```
+
+**Note:** Remember about `\` before `[` and `]` in `zsh` shell.
+
 # References
 
 This software is part of contribution made by [Data Mining Group of Silesian

divik/__init__.py

@@ -1,4 +1,4 @@
-__version__ = '2.3.13'
+__version__ = '2.3.14'
 
 from ._seeding import seeded
 from ._utils import DivikResult
@@ -7,6 +7,30 @@
 from divik import cluster
 from divik import sampler
 from ._summary import plot, reject_split
+from ._gin_compat import (
+    configurable,
+    parse_gin_args
+)
+
+for __estimator in [
+    feature_extraction.KneePCA,
+    feature_extraction.LocallyAdjustedRbfSpectralEmbedding,
+    feature_selection.GMMSelector,
+    feature_selection.OutlierSelector,
+    feature_selection.PercentageSelector,
+    feature_selection.HighAbundanceAndVarianceSelector,
+    feature_selection.NoSelector,
+    feature_selection.OutlierAbundanceAndVarianceSelector,
+    cluster.KMeans,
+    cluster.GAPSearch,
+    cluster.DunnSearch,
+    cluster.DiviK,
+    cluster.DunnDiviK,
+    sampler.UniformSampler,
+    sampler.UniformPCASampler,
+    sampler.StratifiedSampler,
+]:
+    configurable(__estimator)
 
 __all__ = [
     "__version__",
@@ -15,6 +39,7 @@
     "feature_extraction",
     "sampler",
     "seeded",
+    "configurable", "parse_gin_args",
     'DivikResult',
     "plot", "reject_split",
 ]

divik/_cli/_utils.py

@@ -49,12 +49,6 @@ def prepare_destination(destination: str, omit_datetime: bool = False) -> str:
 
 
 def setup_logger(destination: str, verbose: bool = False):
-    try:
-        import divik._matlab_legacy
-        logger = logging.getLogger(divik._matlab_legacy.__name__)
-        logger.setLevel(logging.CRITICAL)
-    except ImportError:
-        pass  # In environments without MATLAB this should work as well
     log_destination = os.path.join(destination, 'logs.txt')
     if verbose:
         log_format = '%(asctime)s [%(levelname)s] %(filename)40s:%(lineno)3s' \

divik/_cli/dunn_divik.json

@@ -0,0 +1,21 @@
+{
+    "gap_trials": 10,
+    "distance_percentile": 99.0,
+    "max_iter": 100,
+    "distance": "correlation",
+    "minimal_size": 16,
+    "rejection_size": 2,
+    "rejection_percentage": null,
+    "minimal_features_percentage": 0.01,
+    "features_percentage": 0.05,
+    "fast_kmeans_iter": 10,
+    "k_max": 10,
+    "sample_size": 1000,
+    "normalize_rows": true,
+    "use_logfilters": true,
+    "filter_type": "gmm",
+    "n_jobs": -1,
+    "random_seed": 0,
+    "verbose": true
+  }
+