utilsforecast
================
<!-- WARNING: THIS FILE WAS AUTOGENERATED! DO NOT EDIT! -->
## Install
### PyPI
``` sh
pip install utilsforecast
```
### Conda
``` sh
conda install -c conda-forge utilsforecast
```
## How to use
### Generate synthetic data
``` python
from utilsforecast.data import generate_series
```
``` python
series = generate_series(3, with_trend=True, static_as_categorical=False)
series
```
<div>
<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th>unique_id</th>
<th>ds</th>
<th>y</th>
</tr>
</thead>
<tbody>
<tr>
<th>0</th>
<td>0</td>
<td>2000-01-01</td>
<td>0.422133</td>
</tr>
<tr>
<th>1</th>
<td>0</td>
<td>2000-01-02</td>
<td>1.501407</td>
</tr>
<tr>
<th>2</th>
<td>0</td>
<td>2000-01-03</td>
<td>2.568495</td>
</tr>
<tr>
<th>3</th>
<td>0</td>
<td>2000-01-04</td>
<td>3.529085</td>
</tr>
<tr>
<th>4</th>
<td>0</td>
<td>2000-01-05</td>
<td>4.481929</td>
</tr>
<tr>
<th>...</th>
<td>...</td>
<td>...</td>
<td>...</td>
</tr>
<tr>
<th>481</th>
<td>2</td>
<td>2000-06-11</td>
<td>163.914625</td>
</tr>
<tr>
<th>482</th>
<td>2</td>
<td>2000-06-12</td>
<td>166.018479</td>
</tr>
<tr>
<th>483</th>
<td>2</td>
<td>2000-06-13</td>
<td>160.839176</td>
</tr>
<tr>
<th>484</th>
<td>2</td>
<td>2000-06-14</td>
<td>162.679603</td>
</tr>
<tr>
<th>485</th>
<td>2</td>
<td>2000-06-15</td>
<td>165.089288</td>
</tr>
</tbody>
</table>
<p>486 rows × 3 columns</p>
</div>
### Plotting
``` python
from utilsforecast.plotting import plot_series
```
``` python
fig = plot_series(series, plot_random=False, max_insample_length=50, engine='matplotlib')
fig.savefig('imgs/index.png', bbox_inches='tight')
```
### Preprocessing
``` python
from utilsforecast.preprocessing import fill_gaps
```
``` python
serie = series[series['unique_id'].eq(0)].tail(10)
# drop some points
with_gaps = serie.sample(frac=0.5, random_state=0).sort_values('ds')
with_gaps
```
<div>
<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th>unique_id</th>
<th>ds</th>
<th>y</th>
</tr>
</thead>
<tbody>
<tr>
<th>213</th>
<td>0</td>
<td>2000-08-01</td>
<td>18.543147</td>
</tr>
<tr>
<th>214</th>
<td>0</td>
<td>2000-08-02</td>
<td>19.941764</td>
</tr>
<tr>
<th>216</th>
<td>0</td>
<td>2000-08-04</td>
<td>21.968733</td>
</tr>
<tr>
<th>220</th>
<td>0</td>
<td>2000-08-08</td>
<td>19.091509</td>
</tr>
<tr>
<th>221</th>
<td>0</td>
<td>2000-08-09</td>
<td>20.220739</td>
</tr>
</tbody>
</table>
</div>
``` python
fill_gaps(with_gaps, freq='D')
```
<div>
<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th>unique_id</th>
<th>ds</th>
<th>y</th>
</tr>
</thead>
<tbody>
<tr>
<th>0</th>
<td>0</td>
<td>2000-08-01</td>
<td>18.543147</td>
</tr>
<tr>
<th>1</th>
<td>0</td>
<td>2000-08-02</td>
<td>19.941764</td>
</tr>
<tr>
<th>2</th>
<td>0</td>
<td>2000-08-03</td>
<td>NaN</td>
</tr>
<tr>
<th>3</th>
<td>0</td>
<td>2000-08-04</td>
<td>21.968733</td>
</tr>
<tr>
<th>4</th>
<td>0</td>
<td>2000-08-05</td>
<td>NaN</td>
</tr>
<tr>
<th>5</th>
<td>0</td>
<td>2000-08-06</td>
<td>NaN</td>
</tr>
<tr>
<th>6</th>
<td>0</td>
<td>2000-08-07</td>
<td>NaN</td>
</tr>
<tr>
<th>7</th>
<td>0</td>
<td>2000-08-08</td>
<td>19.091509</td>
</tr>
<tr>
<th>8</th>
<td>0</td>
<td>2000-08-09</td>
<td>20.220739</td>
</tr>
</tbody>
</table>
</div>
### Evaluating
``` python
from functools import partial
import numpy as np
from utilsforecast.evaluation import evaluate
from utilsforecast.losses import mape, mase
```
``` python
valid = series.groupby('unique_id').tail(7).copy()
train = series.drop(valid.index)
rng = np.random.RandomState(0)
valid['seas_naive'] = train.groupby('unique_id')['y'].tail(7).values
valid['rand_model'] = valid['y'] * rng.rand(valid['y'].shape[0])
daily_mase = partial(mase, seasonality=7)
evaluate(valid, metrics=[mape, daily_mase], train_df=train)
```
<div>
<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th>unique_id</th>
<th>metric</th>
<th>seas_naive</th>
<th>rand_model</th>
</tr>
</thead>
<tbody>
<tr>
<th>0</th>
<td>0</td>
<td>mape</td>
<td>0.024139</td>
<td>0.440173</td>
</tr>
<tr>
<th>1</th>
<td>1</td>
<td>mape</td>
<td>0.054259</td>
<td>0.278123</td>
</tr>
<tr>
<th>2</th>
<td>2</td>
<td>mape</td>
<td>0.042642</td>
<td>0.480316</td>
</tr>
<tr>
<th>3</th>
<td>0</td>
<td>mase</td>
<td>0.907149</td>
<td>16.418014</td>
</tr>
<tr>
<th>4</th>
<td>1</td>
<td>mase</td>
<td>0.991635</td>
<td>6.404254</td>
</tr>
<tr>
<th>5</th>
<td>2</td>
<td>mase</td>
<td>1.013596</td>
<td>11.365040</td>
</tr>
</tbody>
</table>
</div>
Raw data
{
"_id": null,
"home_page": "https://github.com/Nixtla/utilsforecast",
"name": "utilsforecast",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.8",
"maintainer_email": null,
"keywords": "time-series analysis forecasting",
"author": "Nixtla",
"author_email": "business@nixtla.io",
"download_url": "https://files.pythonhosted.org/packages/18/b0/c5a3ef13b537515e57c865549f1c4d19837c6d5531109efec05becd1890a/utilsforecast-0.2.9.tar.gz",
"platform": null,
"description": "utilsforecast\n================\n\n<!-- WARNING: THIS FILE WAS AUTOGENERATED! DO NOT EDIT! -->\n\n## Install\n\n### PyPI\n\n``` sh\npip install utilsforecast\n```\n\n### Conda\n\n``` sh\nconda install -c conda-forge utilsforecast\n```\n\n## How to use\n\n### Generate synthetic data\n\n``` python\nfrom utilsforecast.data import generate_series\n```\n\n``` python\nseries = generate_series(3, with_trend=True, static_as_categorical=False)\nseries\n```\n\n<div>\n<table border=\"1\" class=\"dataframe\">\n <thead>\n <tr style=\"text-align: right;\">\n <th></th>\n <th>unique_id</th>\n <th>ds</th>\n <th>y</th>\n </tr>\n </thead>\n <tbody>\n <tr>\n <th>0</th>\n <td>0</td>\n <td>2000-01-01</td>\n <td>0.422133</td>\n </tr>\n <tr>\n <th>1</th>\n <td>0</td>\n <td>2000-01-02</td>\n <td>1.501407</td>\n </tr>\n <tr>\n <th>2</th>\n <td>0</td>\n <td>2000-01-03</td>\n <td>2.568495</td>\n </tr>\n <tr>\n <th>3</th>\n <td>0</td>\n <td>2000-01-04</td>\n <td>3.529085</td>\n </tr>\n <tr>\n <th>4</th>\n <td>0</td>\n <td>2000-01-05</td>\n <td>4.481929</td>\n </tr>\n <tr>\n <th>...</th>\n <td>...</td>\n <td>...</td>\n <td>...</td>\n </tr>\n <tr>\n <th>481</th>\n <td>2</td>\n <td>2000-06-11</td>\n <td>163.914625</td>\n </tr>\n <tr>\n <th>482</th>\n <td>2</td>\n <td>2000-06-12</td>\n <td>166.018479</td>\n </tr>\n <tr>\n <th>483</th>\n <td>2</td>\n <td>2000-06-13</td>\n <td>160.839176</td>\n </tr>\n <tr>\n <th>484</th>\n <td>2</td>\n <td>2000-06-14</td>\n <td>162.679603</td>\n </tr>\n <tr>\n <th>485</th>\n <td>2</td>\n <td>2000-06-15</td>\n <td>165.089288</td>\n </tr>\n </tbody>\n</table>\n<p>486 rows \u00d7 3 columns</p>\n</div>\n\n### Plotting\n\n``` python\nfrom utilsforecast.plotting import plot_series\n```\n\n``` python\nfig = plot_series(series, plot_random=False, max_insample_length=50, engine='matplotlib')\nfig.savefig('imgs/index.png', bbox_inches='tight')\n```\n\n\n\n### Preprocessing\n\n``` python\nfrom utilsforecast.preprocessing import fill_gaps\n```\n\n``` python\nserie = series[series['unique_id'].eq(0)].tail(10)\n# drop some points\nwith_gaps = serie.sample(frac=0.5, random_state=0).sort_values('ds')\nwith_gaps\n```\n\n<div>\n<table border=\"1\" class=\"dataframe\">\n <thead>\n <tr style=\"text-align: right;\">\n <th></th>\n <th>unique_id</th>\n <th>ds</th>\n <th>y</th>\n </tr>\n </thead>\n <tbody>\n <tr>\n <th>213</th>\n <td>0</td>\n <td>2000-08-01</td>\n <td>18.543147</td>\n </tr>\n <tr>\n <th>214</th>\n <td>0</td>\n <td>2000-08-02</td>\n <td>19.941764</td>\n </tr>\n <tr>\n <th>216</th>\n <td>0</td>\n <td>2000-08-04</td>\n <td>21.968733</td>\n </tr>\n <tr>\n <th>220</th>\n <td>0</td>\n <td>2000-08-08</td>\n <td>19.091509</td>\n </tr>\n <tr>\n <th>221</th>\n <td>0</td>\n <td>2000-08-09</td>\n <td>20.220739</td>\n </tr>\n </tbody>\n</table>\n</div>\n\n``` python\nfill_gaps(with_gaps, freq='D')\n```\n\n<div>\n<table border=\"1\" class=\"dataframe\">\n <thead>\n <tr style=\"text-align: right;\">\n <th></th>\n <th>unique_id</th>\n <th>ds</th>\n <th>y</th>\n </tr>\n </thead>\n <tbody>\n <tr>\n <th>0</th>\n <td>0</td>\n <td>2000-08-01</td>\n <td>18.543147</td>\n </tr>\n <tr>\n <th>1</th>\n <td>0</td>\n <td>2000-08-02</td>\n <td>19.941764</td>\n </tr>\n <tr>\n <th>2</th>\n <td>0</td>\n <td>2000-08-03</td>\n <td>NaN</td>\n </tr>\n <tr>\n <th>3</th>\n <td>0</td>\n <td>2000-08-04</td>\n <td>21.968733</td>\n </tr>\n <tr>\n <th>4</th>\n <td>0</td>\n <td>2000-08-05</td>\n <td>NaN</td>\n </tr>\n <tr>\n <th>5</th>\n <td>0</td>\n <td>2000-08-06</td>\n <td>NaN</td>\n </tr>\n <tr>\n <th>6</th>\n <td>0</td>\n <td>2000-08-07</td>\n <td>NaN</td>\n </tr>\n <tr>\n <th>7</th>\n <td>0</td>\n <td>2000-08-08</td>\n <td>19.091509</td>\n </tr>\n <tr>\n <th>8</th>\n <td>0</td>\n <td>2000-08-09</td>\n <td>20.220739</td>\n </tr>\n </tbody>\n</table>\n</div>\n\n### Evaluating\n\n``` python\nfrom functools import partial\n\nimport numpy as np\n\nfrom utilsforecast.evaluation import evaluate\nfrom utilsforecast.losses import mape, mase\n```\n\n``` python\nvalid = series.groupby('unique_id').tail(7).copy()\ntrain = series.drop(valid.index)\nrng = np.random.RandomState(0)\nvalid['seas_naive'] = train.groupby('unique_id')['y'].tail(7).values\nvalid['rand_model'] = valid['y'] * rng.rand(valid['y'].shape[0])\ndaily_mase = partial(mase, seasonality=7)\nevaluate(valid, metrics=[mape, daily_mase], train_df=train)\n```\n\n<div>\n<table border=\"1\" class=\"dataframe\">\n <thead>\n <tr style=\"text-align: right;\">\n <th></th>\n <th>unique_id</th>\n <th>metric</th>\n <th>seas_naive</th>\n <th>rand_model</th>\n </tr>\n </thead>\n <tbody>\n <tr>\n <th>0</th>\n <td>0</td>\n <td>mape</td>\n <td>0.024139</td>\n <td>0.440173</td>\n </tr>\n <tr>\n <th>1</th>\n <td>1</td>\n <td>mape</td>\n <td>0.054259</td>\n <td>0.278123</td>\n </tr>\n <tr>\n <th>2</th>\n <td>2</td>\n <td>mape</td>\n <td>0.042642</td>\n <td>0.480316</td>\n </tr>\n <tr>\n <th>3</th>\n <td>0</td>\n <td>mase</td>\n <td>0.907149</td>\n <td>16.418014</td>\n </tr>\n <tr>\n <th>4</th>\n <td>1</td>\n <td>mase</td>\n <td>0.991635</td>\n <td>6.404254</td>\n </tr>\n <tr>\n <th>5</th>\n <td>2</td>\n <td>mase</td>\n <td>1.013596</td>\n <td>11.365040</td>\n </tr>\n </tbody>\n</table>\n</div>\n",
"bugtrack_url": null,
"license": "Apache Software License 2.0",
"summary": "Forecasting utilities",
"version": "0.2.9",
"project_urls": {
"Homepage": "https://github.com/Nixtla/utilsforecast"
},
"split_keywords": [
"time-series",
"analysis",
"forecasting"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "c0dc75080416c546291c378b2651633a14900e950267183a05d9522f94ae9aec",
"md5": "a2049a31e322df41ea62e5956680f655",
"sha256": "684398bb8d6e460d53aa2a7c8b9f4256573d61fcd387fc2ecc193f8b5328cab9"
},
"downloads": -1,
"filename": "utilsforecast-0.2.9-py3-none-any.whl",
"has_sig": false,
"md5_digest": "a2049a31e322df41ea62e5956680f655",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.8",
"size": 41682,
"upload_time": "2024-11-22T17:28:31",
"upload_time_iso_8601": "2024-11-22T17:28:31.479042Z",
"url": "https://files.pythonhosted.org/packages/c0/dc/75080416c546291c378b2651633a14900e950267183a05d9522f94ae9aec/utilsforecast-0.2.9-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "18b0c5a3ef13b537515e57c865549f1c4d19837c6d5531109efec05becd1890a",
"md5": "f8148029517533efc68b5fd7654d925a",
"sha256": "f6e0515637d5f6556327e3b1784a3dc190c7a193b9419aca029e1395d099215c"
},
"downloads": -1,
"filename": "utilsforecast-0.2.9.tar.gz",
"has_sig": false,
"md5_digest": "f8148029517533efc68b5fd7654d925a",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.8",
"size": 40973,
"upload_time": "2024-11-22T17:28:33",
"upload_time_iso_8601": "2024-11-22T17:28:33.557352Z",
"url": "https://files.pythonhosted.org/packages/18/b0/c5a3ef13b537515e57c865549f1c4d19837c6d5531109efec05becd1890a/utilsforecast-0.2.9.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-11-22 17:28:33",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "Nixtla",
"github_project": "utilsforecast",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"lcname": "utilsforecast"
}
JSON
