/dports/math/py-hdbscan/hdbscan-0.8.27/hdbscan/tests/ |
H A D | test_hdbscan.py | 13 from hdbscan import (HDBSCAN, 145 labels = HDBSCAN(metric="precomputed", 156 labels = HDBSCAN().fit(X).labels_ 170 labels = HDBSCAN(algorithm='prims_kdtree', 241 labels = HDBSCAN(algorithm='generic', 589 labels = HDBSCAN().fit(sparse_X).labels_ 608 clusterer = HDBSCAN().fit(H) 619 clusterer = HDBSCAN().fit(X) 628 labels = HDBSCAN(min_cluster_size=5, 638 labels = HDBSCAN(min_cluster_size=5, [all …]
|
H A D | test_flat.py | 7 from hdbscan import HDBSCAN, approximate_predict 51 clusterer = HDBSCAN(cluster_selection_method='eom').fit(X) 64 clusterer = HDBSCAN(cluster_selection_method='leaf').fit(X) 91 clusterer = HDBSCAN(cluster_selection_method='eom', 107 clusterer = HDBSCAN(cluster_selection_method='leaf', 124 clusterer = HDBSCAN(cluster_selection_method='eom', 142 clusterer_leaf = HDBSCAN(cluster_selection_method='leaf', 155 clusterer = HDBSCAN(cluster_selection_method='eom',
|
/dports/math/py-hdbscan/hdbscan-0.8.27/ |
H A D | README.rst | 31 HDBSCAN chapter 34 HDBSCAN - Hierarchical Density-Based Spatial Clustering of Applications 37 This allows HDBSCAN to find clusters of varying densities (unlike DBSCAN), 40 In practice this means that HDBSCAN returns a good clustering straight 44 HDBSCAN is ideal for exploratory data analysis; it's a fast and robust 61 …HDBSCAN to other clustering algorithms <http://nbviewer.jupyter.org/github/scikit-learn-contrib/hd… 64 How to use HDBSCAN 80 clusterer = hdbscan.HDBSCAN(min_cluster_size=10) 109 `how HDBSCAN works <http://nbviewer.jupyter.org/github/scikit-learn-contrib/hdbscan/blob/master/not… 121 The HDBSCAN clusterer objects also support the GLOSH outlier detection algorithm. [all …]
|
H A D | PKG-INFO | 39 HDBSCAN 42 HDBSCAN - Hierarchical Density-Based Spatial Clustering of Applications 45 This allows HDBSCAN to find clusters of varying densities (unlike DBSCAN), 48 In practice this means that HDBSCAN returns a good clustering straight 52 HDBSCAN is ideal for exploratory data analysis; it's a fast and robust 69 …HDBSCAN to other clustering algorithms <http://nbviewer.jupyter.org/github/scikit-learn-contrib/hd… 72 How to use HDBSCAN 88 clusterer = hdbscan.HDBSCAN(min_cluster_size=10) 117 …`how HDBSCAN works <http://nbviewer.jupyter.org/github/scikit-learn-contrib/hdbscan/blob/master/no… 129 The HDBSCAN clusterer objects also support the GLOSH outlier detection algorithm. [all …]
|
/dports/math/py-hdbscan/hdbscan-0.8.27/hdbscan.egg-info/ |
H A D | PKG-INFO | 39 HDBSCAN 42 HDBSCAN - Hierarchical Density-Based Spatial Clustering of Applications 45 This allows HDBSCAN to find clusters of varying densities (unlike DBSCAN), 48 In practice this means that HDBSCAN returns a good clustering straight 52 HDBSCAN is ideal for exploratory data analysis; it's a fast and robust 69 …HDBSCAN to other clustering algorithms <http://nbviewer.jupyter.org/github/scikit-learn-contrib/hd… 72 How to use HDBSCAN 88 clusterer = hdbscan.HDBSCAN(min_cluster_size=10) 117 …`how HDBSCAN works <http://nbviewer.jupyter.org/github/scikit-learn-contrib/hdbscan/blob/master/no… 129 The HDBSCAN clusterer objects also support the GLOSH outlier detection algorithm. [all …]
|
/dports/math/py-hdbscan/hdbscan-0.8.27/notebooks/ |
H A D | Performance data generation .ipynb | 9 …"We need to generate data comparing performance of the reference implementation of HDBSCAN and var… 180 " hdbscan01.HDBSCAN().fit(data)\n", 204 " hdbscan02.HDBSCAN().fit(data)\n", 228 " hdbscan03.HDBSCAN().fit(data)\n", 252 " hdbscan04.HDBSCAN().fit(data)\n", 276 " hdbscan05.HDBSCAN().fit(data)\n", 307 " hdbscan.HDBSCAN().fit(data)\n",
|
H A D | How HDBSCAN Works.ipynb | 7 "# How HDBSCAN Works\n", 9 …HDBSCAN is a clustering algorithm developed by [Campello, Moulavi, and Sander](http://link.springe… 72 …"Now, the best way to explain HDBSCAN is actually just use it and then go through the steps that o… 92 "HDBSCAN(algorithm='best', allow_single_cluster=False, alpha=1.0,\n", 104 "clusterer = hdbscan.HDBSCAN(min_cluster_size=5, gen_min_span_tree=True)\n", 131 …) are good for. Let's formalise this and (following the DBSCAN, LOF, and HDBSCAN literature) call … 174 …s the current tree to a vertex not yet in the tree. You can see the tree HDBSCAN constructed below… 249 …different places to select our clusters. This is where the next steps of HDBSCAN begin and create … 258 …eed a notion of **minimum cluster size** which we take as a parameter to HDBSCAN. Once we have a v… 301 …And in fact that intuitive notion of what should be done is exactly what HDBSCAN does. Of course w… [all …]
|
H A D | Benchmarking scalability of clustering implementations 2D v0.7.ipynb | 24 …"* [HDBSCAN](https://github.com/scikit-learn-contrib/hdbscan) (A robust hierarchical version of DB… 156 "hdbscan_ = hdbscan.HDBSCAN()\n", 218 "sns.regplot(x='x', y='y', data=hdbscan_data, order=2, label='HDBSCAN', x_estimator=np.mean)\n", 260 "hdbscan_prims = hdbscan.HDBSCAN(algorithm='prims_kdtree')\n", 264 "hdbscan_boruvka = hdbscan.HDBSCAN(algorithm='boruvka_kdtree')\n", 658 "## Comparison of K-Means and DBSCAN and HDBSCAN implementations\n", 684 "hdbscan_boruvka = hdbscan.HDBSCAN(algorithm='boruvka_kdtree')\n", 724 …"sns.regplot(x='x', y='y', data=huge_hdbscan_data, order=2, label='HDBSCAN', x_estimator=np.mean)\… 853 " <th>HDBSCAN</th>\n", 885 "HDBSCAN 500000 2500000 10000000 100000000\n", [all …]
|
H A D | Benchmarking scalability of clustering implementations-v0.7.ipynb | 26 …"* [HDBSCAN](https://github.com/scikit-learn-contrib/hdbscan) (A robust hierarchical version of DB… 168 "hdbscan_ = hdbscan.HDBSCAN()\n", 230 " label='HDBSCAN', x_estimator=np.mean)\n", 280 "hdbscan_boruvka = hdbscan.HDBSCAN(algorithm='boruvka_kdtree')\n", 351 " label='HDBSCAN Boruvka', x_estimator=np.mean)\n", 552 " label='HDBSCAN Boruvka', x_estimator=np.mean)\n", 609 "hdbscan_boruvka = hdbscan.HDBSCAN(algorithm='boruvka_kdtree')\n", 655 " label='HDBSCAN', x_estimator=np.mean)\n", 785 " <th>HDBSCAN</th>\n", 817 "HDBSCAN 100000 500000 1000000 5000000\n", [all …]
|
H A D | Flat clustering.ipynb | 7 "# Flat Clustering with HDBSCAN\n", 13 …"A new module 'flat' is added to modify a few things here and there with the HDBSCAN class to prop… 14 "1. Flat clusterings can be easily extracted from the HDBSCAN hierarchy\n", 15 …"2. The HDBSCAN class can be made to point to a particular flat clustering, so that later queries … 47 "from hdbscan import HDBSCAN\n", 111 "# Train HDBSCAN aimed at some flat clustering\n", 113 …"The function used for this is hdbscan.flat.HDBSCAN_flat. Unlike hdbscan.HDBSCAN, this function di… 120 "## Let HDBSCAN choose n_clusters" 464 "# Train the base HDBSCAN class\n", 465 "clusterer = HDBSCAN(cluster_selection_method='eom', min_cluster_size=10).fit(X)\n", [all …]
|
H A D | Python vs Java.ipynb | 7 "# The Development of Python HDBSCAN Compared to the Reference Implementation in Java\n", 20 …HDBSCAN. In mid-2014 I was doing some general research on the current state of clustering, particu… 67 …"The very first implementation of HDBSCAN that we did was coded up in an afternoon, and that code … 69 …because it was the remarkable promise of those results that made us pick HDBSCAN as the ideal clus… 78 …ve and it was decided that we needed to just write and implementation of HDBSCAN, I stuck with pyt… 265 …HDBSCAN algorithm relied on a modified single linkage algorithm, which in turn relied on something… 267 …ely python made that easy. As in the case of the first naive versions of HDBSCAN, the notebook pro…
|
H A D | Looking at cluster consistency.ipynb | 9 …art of this. It is also helpful to look at some representative examples. HDBSCAN provides methods … 49 "clusterer = hdbscan.HDBSCAN(min_cluster_size=15)\n", 896 " HDBSCAN\n", 898 " clusterer : Instance of HDBSCAN that has been fit to data\n", 1000 …ed to instantiate the class. We pass in our sample data and the pretrained HDBSCAN class instance."
|
H A D | Comparing Clustering Algorithms.ipynb | 8 "#### (Why you should use HDBSCAN)\n", 442 "## HDBSCAN\n", 444 …"HDBSCAN is a recent algorithm developed by some of the same people who write the original DBSCAN … 448 …l the benefits of DBSCAN and removed the varying density clusters issue. HDBSCAN is easily the str… 449 …y based space transformation. Sadly `min_samples` is not that intuitive; HDBSCAN is not that sensi… 450 …"* **Stability**: HDBSCAN is stable over runs and subsampling (since the variable density clusteri… 451 …"* **Performance**: When implemented well HDBSCAN can be very efficient. The current implementatio… 453 …"How does HDBSCAN perform on our test dataset? Unfortunately HDBSCAN is not part of `sklearn`. For… 487 "plot_clusters(data, hdbscan.HDBSCAN, (), {'min_cluster_size':15})"
|
H A D | How Soft Clustering for HDBSCAN Works.ipynb | 7 "How Soft Clustering for HDBSCAN Works\n", 10 …"This is a general description of how the soft clustering algorithm for HDBSCAN Works. We will imp… 19 "Soft Clustering for HDBSCAN\n", 93 "clusterer = hdbscan.HDBSCAN(min_cluster_size=15).fit(data)" 142 … be the points that persist in the the cluster (and it's children in the HDBSCAN condensed tree) f…
|
/dports/math/py-hdbscan/hdbscan-0.8.27/hdbscan/ |
H A D | flat.py | 34 from .hdbscan_ import HDBSCAN, _tree_to_labels 107 if (not isinstance(clusterer, HDBSCAN)) or (not inplace): 110 new_clusterer = HDBSCAN(**kwargs) 124 if not isinstance(clusterer, HDBSCAN): 128 new_clusterer = HDBSCAN(**kwargs)
|
H A D | __init__.py | 1 from .hdbscan_ import HDBSCAN, hdbscan
|
H A D | hdbscan_.py | 643 class HDBSCAN(BaseEstimator, ClusterMixin): class
|
H A D | _hdbscan_tree.pyx | 348 from a single HDBSCAN run.
|
H A D | _hdbscan_boruvka.pyx | 260 The min_samples parameter of HDBSCAN used to 868 The min_samples parameter of HDBSCAN used to
|
H A D | dist_metrics.pyx | 8 # modified for HDBSCAN Dual Tree Boruvka algorithm
|
/dports/math/py-hdbscan/hdbscan-0.8.27/examples/ |
H A D | plot_hdbscan.py | 16 from hdbscan import HDBSCAN 44 hdb = HDBSCAN(min_cluster_size=10).fit(X)
|
H A D | plot_cluster_comparison.py | 102 hdbscanner = hdbscan.HDBSCAN()
|
/dports/science/py-scikit-learn/scikit-learn-1.0.2/doc/ |
H A D | related_projects.rst | 238 - `hdbscan <https://github.com/scikit-learn-contrib/hdbscan>`_ HDBSCAN and Robust Single
|
/dports/science/py-scikit-learn/scikit-learn-1.0.2/doc/modules/ |
H A D | clustering.rst | 937 `HDBSCAN <https://hdbscan.readthedocs.io>`_. The HDBSCAN implementation is 940 exhaust system memory using HDBSCAN, OPTICS will maintain :math:`n` (as opposed
|