TSUtilities

Name	TSUtilities JSON
Version	0.0.2 JSON
	download
home_page	https://github.com/tblume1992/TSUtilities
Summary	Various utilities for time series forecasting.
upload_time	2023-01-24 15:59:42
maintainer
docs_url	None
author	Tyler Blume
requires_python
license
keywords	forecasting time series seasonality trend
VCS
bugtrack_url
requirements	No requirements were recorded.
Travis-CI	No Travis.
coveralls test coverage	No coveralls.

            # TSUtilities v0.0.2

## Recent Changes

pip install TSUtilities:
```
pip install TSUtilities
```

Example of trend dampening:

```
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
import numpy as np

sns.set_style('darkgrid')
y = np.linspace(0, 100, 100)
plt.plot(y)
plt.show()

y_train = y[:80]
future_y = y[80:]
future_trend = future_y


from TSUtilities.TSTrend.trend_dampen import TrendDampen

dampener = TrendDampen(damp_factor=.7,
                       damp_style='smooth')
dampened_trend = dampener.dampen(future_trend)
```

Example of Prophet Trend Dampening helper function where ts is your input to prophet:

```
from TSUtilities.functions import dampen_prophet

prophet = Prophet()
prophet.fit(ts)
fitted = prophet.predict()

# create a future data frame
future = prophet.make_future_dataframe(periods=len(y_test))
forecast = prophet.predict(future)

#get predictions and required data inputs for auto-damping
predictions = forecast.tail(len(y_test))
predicted_trend = predictions['trend'].values
trend_component = fitted['trend'].values
seasonality_component = fitted['additive_terms'].values
forecasts_no_dampen = predictions['yhat'].values
forecasts_damped = dampen_prophet(y=y.values,
                                  fit_df=fitted,
                                  forecast_df=forecast)
```

            

Raw data

            {
    "_id": null,
    "home_page": "https://github.com/tblume1992/TSUtilities",
    "name": "TSUtilities",
    "maintainer": "",
    "docs_url": null,
    "requires_python": "",
    "maintainer_email": "",
    "keywords": "forecasting,time series,seasonality,trend",
    "author": "Tyler Blume",
    "author_email": "tblume@mail.USF.edu",
    "download_url": "",
    "platform": null,
    "description": "# TSUtilities v0.0.2\n\n## Recent Changes\n\npip install TSUtilities:\n```\npip install TSUtilities\n```\n\nExample of trend dampening:\n\n```\nimport pandas as pd\nimport matplotlib.pyplot as plt\nimport seaborn as sns\nimport numpy as np\n\nsns.set_style('darkgrid')\ny = np.linspace(0, 100, 100)\nplt.plot(y)\nplt.show()\n\ny_train = y[:80]\nfuture_y = y[80:]\nfuture_trend = future_y\n\n\nfrom TSUtilities.TSTrend.trend_dampen import TrendDampen\n\ndampener = TrendDampen(damp_factor=.7,\n                       damp_style='smooth')\ndampened_trend = dampener.dampen(future_trend)\n```\n\nExample of Prophet Trend Dampening helper function where ts is your input to prophet:\n\n```\nfrom TSUtilities.functions import dampen_prophet\n\nprophet = Prophet()\nprophet.fit(ts)\nfitted = prophet.predict()\n\n# create a future data frame\nfuture = prophet.make_future_dataframe(periods=len(y_test))\nforecast = prophet.predict(future)\n\n#get predictions and required data inputs for auto-damping\npredictions = forecast.tail(len(y_test))\npredicted_trend = predictions['trend'].values\ntrend_component = fitted['trend'].values\nseasonality_component = fitted['additive_terms'].values\nforecasts_no_dampen = predictions['yhat'].values\nforecasts_damped = dampen_prophet(y=y.values,\n                                  fit_df=fitted,\n                                  forecast_df=forecast)\n```\n",
    "bugtrack_url": null,
    "license": "",
    "summary": "Various utilities for time series forecasting.",
    "version": "0.0.2",
    "split_keywords": [
        "forecasting",
        "time series",
        "seasonality",
        "trend"
    ],
    "urls": [
        {
            "comment_text": "",
            "digests": {
                "blake2b_256": "595a5bbda093a004a9b455f54b40f8b8054cc698c12b1809fdda92ff13030496",
                "md5": "279db5aca849478581128569f7e4480d",
                "sha256": "00158e456079212048b931ca2182a642bf9bd4ac6064b13d03f7f5a5289fb06b"
            },
            "downloads": -1,
            "filename": "TSUtilities-0.0.2-py3-none-any.whl",
            "has_sig": false,
            "md5_digest": "279db5aca849478581128569f7e4480d",
            "packagetype": "bdist_wheel",
            "python_version": "py3",
            "requires_python": null,
            "size": 6847,
            "upload_time": "2023-01-24T15:59:42",
            "upload_time_iso_8601": "2023-01-24T15:59:42.918959Z",
            "url": "https://files.pythonhosted.org/packages/59/5a/5bbda093a004a9b455f54b40f8b8054cc698c12b1809fdda92ff13030496/TSUtilities-0.0.2-py3-none-any.whl",
            "yanked": false,
            "yanked_reason": null
        }
    ],
    "upload_time": "2023-01-24 15:59:42",
    "github": true,
    "gitlab": false,
    "bitbucket": false,
    "github_user": "tblume1992",
    "github_project": "TSUtilities",
    "travis_ci": false,
    "coveralls": false,
    "github_actions": false,
    "lcname": "tsutilities"
}