# Online evolutionary Spiking Neural Network unsupervised anomaly detector
## Description
Python implementation of OeSNN-UAD model. Model finds anomalies in one dimensional data streams. Theoretical basics about this model could be find here: https://arxiv.org/pdf/1912.08785.pdf.
## Dependencies
* numpy
* PyEMD (pip install EMD-signal)
## Package instalation
To install package for python you should type in terminal:
```bash
pip install OeSNN-AD
```
## Usage
Our model require from data stream to be numpy array. Additional model parameters are passed as arguments in object constructor.
The following code snippet shows package basic usage.
```python
from oesnn_ad import OeSNNAD
import numpy as np
data_stream = np.array([1, 2, 3, 4, 5])
model = oesnn_ad(data_stream)
results = model.predict()
```
## Parameters
The following table shows model parameters and their values range.
<center>
| Parameter | Default value | Minimal value | Maximum value |
| --------------- | :-----------: | :-----------: | :-----------: |
| window_size | 100 | 1 | - |
| num_in_neurons | 10 | 1 | - |
| num_out_neurons | 50 | 1 | - |
| ts_factor | 1000 | 0 | - |
| mod | 0.6 | 0 | 1 |
| c_factor | 0.6 | 0 | 1 |
| epsilon | 2 | 2 | - |
| ksi | 0.9 | 0 | 1 |
| sim | 0.15 | 0 | - |
| beta | 1.6 | 0 | - |
</center>
Raw data
{
"_id": null,
"home_page": "",
"name": "OeSNN-AD",
"maintainer": "",
"docs_url": null,
"requires_python": "",
"maintainer_email": "",
"keywords": "python,data stream,spiking neural network,SNN,anomalies,anomaly detection,CEEMDAN",
"author": "Mariusz Paluch",
"author_email": "mariuszpaluch001@gmail.com",
"download_url": "https://files.pythonhosted.org/packages/a4/42/f614a49315e1b4e5a1f5ea590e8897f5cd6ea3e546399c80bbcc2d4d51a6/OeSNN-AD-1.0.1.tar.gz",
"platform": null,
"description": "\n# Online evolutionary Spiking Neural Network unsupervised anomaly detector\n\n## Description\n\n Python implementation of OeSNN-UAD model. Model finds anomalies in one dimensional data streams. Theoretical basics about this model could be find here: https://arxiv.org/pdf/1912.08785.pdf.\n\n## Dependencies\n\n* numpy\n* PyEMD (pip install EMD-signal) \n\n## Package instalation\n\nTo install package for python you should type in terminal:\n\n```bash\n pip install OeSNN-AD\n```\n\n## Usage\n\nOur model require from data stream to be numpy array. Additional model parameters are passed as arguments in object constructor.\n\nThe following code snippet shows package basic usage.\n\n```python\n from oesnn_ad import OeSNNAD\n import numpy as np\n\n data_stream = np.array([1, 2, 3, 4, 5])\n model = oesnn_ad(data_stream)\n\n results = model.predict()\n```\n\n## Parameters\n\nThe following table shows model parameters and their values range.\n\n<center>\n\n| Parameter | Default value | Minimal value | Maximum value |\n| --------------- | :-----------: | :-----------: | :-----------: |\n| window_size | 100 | 1 | - |\n| num_in_neurons | 10 | 1 | - |\n| num_out_neurons | 50 | 1 | - |\n| ts_factor | 1000 | 0 | - |\n| mod | 0.6 | 0 | 1 |\n| c_factor | 0.6 | 0 | 1 |\n| epsilon | 2 | 2 | - |\n| ksi | 0.9 | 0 | 1 |\n| sim | 0.15 | 0 | - |\n| beta | 1.6 | 0 | - |\n\n</center>\n",
"bugtrack_url": null,
"license": "",
"summary": "OeSNN-UAD anomaly detector implementation for Python.",
"version": "1.0.1",
"project_urls": null,
"split_keywords": [
"python",
"data stream",
"spiking neural network",
"snn",
"anomalies",
"anomaly detection",
"ceemdan"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "711e281b55c78cc3df76c8fd3205ce6a71db687443cfa54144986ce384832647",
"md5": "21e3196cd4910badeb37a98c741c574d",
"sha256": "f74279bf9f91810305bda83851be353e3eac86ef73bc27b69c67d4739096ee48"
},
"downloads": -1,
"filename": "OeSNN_AD-1.0.1-py3-none-any.whl",
"has_sig": false,
"md5_digest": "21e3196cd4910badeb37a98c741c574d",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": null,
"size": 18408,
"upload_time": "2023-12-30T10:45:38",
"upload_time_iso_8601": "2023-12-30T10:45:38.491655Z",
"url": "https://files.pythonhosted.org/packages/71/1e/281b55c78cc3df76c8fd3205ce6a71db687443cfa54144986ce384832647/OeSNN_AD-1.0.1-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "a442f614a49315e1b4e5a1f5ea590e8897f5cd6ea3e546399c80bbcc2d4d51a6",
"md5": "124b6a67755a06f25edba4592659df3b",
"sha256": "841a4d09286fa0f98812ea8445444e06f0fcc4d90ccbb3aca92b9b323f523789"
},
"downloads": -1,
"filename": "OeSNN-AD-1.0.1.tar.gz",
"has_sig": false,
"md5_digest": "124b6a67755a06f25edba4592659df3b",
"packagetype": "sdist",
"python_version": "source",
"requires_python": null,
"size": 13208,
"upload_time": "2023-12-30T10:45:41",
"upload_time_iso_8601": "2023-12-30T10:45:41.250078Z",
"url": "https://files.pythonhosted.org/packages/a4/42/f614a49315e1b4e5a1f5ea590e8897f5cd6ea3e546399c80bbcc2d4d51a6/OeSNN-AD-1.0.1.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2023-12-30 10:45:41",
"github": false,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"lcname": "oesnn-ad"
}
JSON