<!-- html title in the middle -->
<div align="center">
# Synthetic Time Series Data Generator
[](https://pypi.org/project/ts-data-generator) 
A Python library for generating synthetic time series data
<sup>Special thanks to: [Nike-Inc](https://github.com/Nike-Inc/timeseries-generator) repo
<img src="https://github.com/manojmanivannan/ts-data-generator/raw/main/notebooks/image.png" alt="MarineGEO circle logo" style="height: 1000px; width:800px;"/>
<!-- ![Tutorial][tutorial] -->
</div>
## Installation
### PyPi (recommended)
You can install with pip directly by
```bash
pip install ts-data-generator
```
### Repo
After cloning this repo and creating a virtual environment, run the following command:
```bash
pip install --editable .
```
## Usage
1. To check out constructing for time series data, check the sample notebook [here](https://github.com/manojmanivannan/ts-data-generator/blob/main/notebooks/sample.ipynb)
2. To extract the trends from an existing data, check this sample notebook [here](https://github.com/manojmanivannan/ts-data-generator/blob/main/notebooks/imputer.ipynb)
### CLI
You can also use the command line utility `tsdata` to generate the data.
```bash
(venv) ~/ts-data-generator cli $ tsdata generate --help
Usage: tsdata generate [OPTIONS]
Generate time series data and save it to a CSV file.
Options:
--start TEXT Start datetime 'YYYY-MM-DD' [required]
--end TEXT End datetime 'YYYY-MM-DD' [required]
--granularity [FIVE_MIN|HOURLY|DAILY]
Granularity of the time series data [required]
--dims TEXT + separated list of dimensions definition of format 'name:function:values' [required]
--mets TEXT + separated list of metrics definition trends of format 'name:trend(*params)' [required]
--output TEXT Output file name [required]
--help Show this message and exit.
```
For example you can call this cli tool like below to generate data
```bash
tsdata generate \
--start "2019-01-01" \
--end "2019-01-12" \
--granularity "FIVE_MIN" \
--dims "product:random_choice:A,B,C,D" \
--dims "product_id:random_float:1,4" \
--dims "const:constant:5" \
--mets "sales:LinearTrend(limit=500)+WeekendTrend(weekend_effect=50)" \
--mets "trend:LinearTrend(limit=10)" \
--output "data.csv"
```
#### Release method
1. `git tag <x.x.x>`
2. `git push origin <x.x.x>`
<!-- [tutorial]: /notebooks/test.gif -->
Raw data
{
"_id": null,
"home_page": null,
"name": "ts-data-generator",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.8",
"maintainer_email": null,
"keywords": "data generator, python, synthetic data, time series",
"author": null,
"author_email": "Manoj Manivannan <manojm18@live.in>",
"download_url": "https://files.pythonhosted.org/packages/51/37/5ff7030059ec2889c4b0f5f02d939097aab10d634996c4d3a725fd9aac29/ts_data_generator-0.1.3.tar.gz",
"platform": null,
"description": "<!-- html title in the middle -->\n<div align=\"center\">\n\n# Synthetic Time Series Data Generator\n\n[](https://pypi.org/project/ts-data-generator) \n\nA Python library for generating synthetic time series data\n\n<sup>Special thanks to: [Nike-Inc](https://github.com/Nike-Inc/timeseries-generator) repo\n\n<img src=\"https://github.com/manojmanivannan/ts-data-generator/raw/main/notebooks/image.png\" alt=\"MarineGEO circle logo\" style=\"height: 1000px; width:800px;\"/>\n\n<!-- ![Tutorial][tutorial] -->\n\n</div>\n\n## Installation\n### PyPi (recommended)\nYou can install with pip directly by\n```bash\npip install ts-data-generator\n```\n\n### Repo\nAfter cloning this repo and creating a virtual environment, run the following command:\n```bash\npip install --editable .\n```\n\n## Usage\n1. To check out constructing for time series data, check the sample notebook [here](https://github.com/manojmanivannan/ts-data-generator/blob/main/notebooks/sample.ipynb)\n2. To extract the trends from an existing data, check this sample notebook [here](https://github.com/manojmanivannan/ts-data-generator/blob/main/notebooks/imputer.ipynb)\n\n### CLI\n\nYou can also use the command line utility `tsdata` to generate the data.\n```bash\n(venv) ~/ts-data-generator \ue0a0 cli $ tsdata generate --help\nUsage: tsdata generate [OPTIONS]\n\n Generate time series data and save it to a CSV file.\n\nOptions:\n --start TEXT Start datetime 'YYYY-MM-DD' [required]\n --end TEXT End datetime 'YYYY-MM-DD' [required]\n --granularity [FIVE_MIN|HOURLY|DAILY]\n Granularity of the time series data [required]\n --dims TEXT + separated list of dimensions definition of format 'name:function:values' [required]\n --mets TEXT + separated list of metrics definition trends of format 'name:trend(*params)' [required]\n --output TEXT Output file name [required]\n --help Show this message and exit.\n ```\nFor example you can call this cli tool like below to generate data\n```bash\ntsdata generate \\\n --start \"2019-01-01\" \\\n --end \"2019-01-12\" \\\n --granularity \"FIVE_MIN\" \\\n --dims \"product:random_choice:A,B,C,D\" \\\n --dims \"product_id:random_float:1,4\" \\\n --dims \"const:constant:5\" \\\n --mets \"sales:LinearTrend(limit=500)+WeekendTrend(weekend_effect=50)\" \\\n --mets \"trend:LinearTrend(limit=10)\" \\\n --output \"data.csv\"\n```\n\n#### Release method\n1. `git tag <x.x.x>`\n2. `git push origin <x.x.x>`\n\n<!-- [tutorial]: /notebooks/test.gif -->",
"bugtrack_url": null,
"license": "MIT",
"summary": "A Python library for generating synthetic time series data",
"version": "0.1.3",
"project_urls": null,
"split_keywords": [
"data generator",
" python",
" synthetic data",
" time series"
],
"urls": [
{
"comment_text": null,
"digests": {
"blake2b_256": "603f395dc18692247499e2ca8a4ace76f894df4702a5a3ff1a376d633ea24af7",
"md5": "f298985b08aea48970bd3bdad8119cf4",
"sha256": "030639b23bff9aa411f8f4e46abeeeb9a397318b036660f48364ccf2e83de217"
},
"downloads": -1,
"filename": "ts_data_generator-0.1.3-py3-none-any.whl",
"has_sig": false,
"md5_digest": "f298985b08aea48970bd3bdad8119cf4",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.8",
"size": 17173,
"upload_time": "2025-02-19T20:05:29",
"upload_time_iso_8601": "2025-02-19T20:05:29.033175Z",
"url": "https://files.pythonhosted.org/packages/60/3f/395dc18692247499e2ca8a4ace76f894df4702a5a3ff1a376d633ea24af7/ts_data_generator-0.1.3-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": null,
"digests": {
"blake2b_256": "51375ff7030059ec2889c4b0f5f02d939097aab10d634996c4d3a725fd9aac29",
"md5": "0ab2fe3ee721c6b4ddc7f7d4c081b9e0",
"sha256": "1a6e4bb71a41baa86a1a66101a2f3881b9b332bb4c8d51d1ed0d263bc281110f"
},
"downloads": -1,
"filename": "ts_data_generator-0.1.3.tar.gz",
"has_sig": false,
"md5_digest": "0ab2fe3ee721c6b4ddc7f7d4c081b9e0",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.8",
"size": 844386,
"upload_time": "2025-02-19T20:05:33",
"upload_time_iso_8601": "2025-02-19T20:05:33.129759Z",
"url": "https://files.pythonhosted.org/packages/51/37/5ff7030059ec2889c4b0f5f02d939097aab10d634996c4d3a725fd9aac29/ts_data_generator-0.1.3.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2025-02-19 20:05:33",
"github": false,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"lcname": "ts-data-generator"
}
JSON