{
"cells": [
{
"cell_type": "markdown",
"id": "41296cc6-9d84-47c5-8a92-2d292f6f3c4a",
"metadata": {
"papermill": {
"duration": 0.01523,
"end_time": "2024-01-31T17:49:38.190595",
"exception": false,
"start_time": "2024-01-31T17:49:38.175365",
"status": "completed"
},
"tags": []
},
"source": [
"# Plots"
]
},
{
"cell_type": "code",
"execution_count": 1,
"id": "9286e0b8-3c78-4b0f-943c-d219e9840dfe",
"metadata": {
"ExecuteTime": {
"end_time": "2023-11-22T14:30:12.399978600Z",
"start_time": "2023-11-22T14:30:12.220250Z"
},
"papermill": {
"duration": 0.015281,
"end_time": "2024-01-31T17:49:38.214428",
"exception": false,
"start_time": "2024-01-31T17:49:38.199147",
"status": "completed"
},
"tags": []
},
"outputs": [],
"source": [
"# Import to be able to import python package from src\n",
"import sys\n",
"sys.path.insert(0, '../src')"
]
},
{
"cell_type": "code",
"execution_count": 2,
"id": "2028eed7-b1c3-4c9e-b6a0-00433caa7d0f",
"metadata": {
"ExecuteTime": {
"end_time": "2023-11-22T14:30:12.438915100Z",
"start_time": "2023-11-22T14:30:12.223262200Z"
},
"papermill": {
"duration": 2.494197,
"end_time": "2024-01-31T17:49:40.714680",
"exception": false,
"start_time": "2024-01-31T17:49:38.220483",
"status": "completed"
},
"tags": []
},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"The `LightGBM` module could not be imported. To enable LightGBM support in Darts, follow the detailed instructions in the installation guide: https://github.com/unit8co/darts/blob/master/INSTALL.md\n",
"The `Prophet` module could not be imported. To enable Prophet support in Darts, follow the detailed instructions in the installation guide: https://github.com/unit8co/darts/blob/master/INSTALL.md\n"
]
}
],
"source": [
"import pandas as pd\n",
"import ontime as on\n",
"from darts.datasets import EnergyDataset"
]
},
{
"cell_type": "markdown",
"id": "e24da8ab-6a83-4c2f-9ff0-c633d4693a91",
"metadata": {
"papermill": {
"duration": 0.004156,
"end_time": "2024-01-31T17:49:40.723198",
"exception": false,
"start_time": "2024-01-31T17:49:40.719042",
"status": "completed"
},
"tags": []
},
"source": [
"## Load data"
]
},
{
"cell_type": "code",
"execution_count": 3,
"id": "db08372d-8ab2-4290-9196-76eb0c275629",
"metadata": {
"ExecuteTime": {
"end_time": "2023-11-22T14:30:12.886891300Z",
"start_time": "2023-11-22T14:30:12.235870500Z"
},
"papermill": {
"duration": 0.070649,
"end_time": "2024-01-31T17:49:40.797936",
"exception": false,
"start_time": "2024-01-31T17:49:40.727287",
"status": "completed"
},
"tags": []
},
"outputs": [],
"source": [
"ts = EnergyDataset().load()"
]
},
{
"cell_type": "markdown",
"id": "789865e9-e840-4267-994a-f10743e46279",
"metadata": {
"papermill": {
"duration": 0.004193,
"end_time": "2024-01-31T17:49:40.806517",
"exception": false,
"start_time": "2024-01-31T17:49:40.802324",
"status": "completed"
},
"tags": []
},
"source": [
"Complete TimeSeries"
]
},
{
"cell_type": "code",
"execution_count": 4,
"id": "009bcbe3-b73c-4355-b280-1bcb3d98e113",
"metadata": {
"ExecuteTime": {
"end_time": "2023-11-22T14:30:12.888403100Z",
"start_time": "2023-11-22T14:30:12.382164600Z"
},
"papermill": {
"duration": 0.018375,
"end_time": "2024-01-31T17:49:40.829148",
"exception": false,
"start_time": "2024-01-31T17:49:40.810773",
"status": "completed"
},
"tags": []
},
"outputs": [],
"source": [
"df = ts.pd_dataframe()\n",
"df = df.interpolate()\n",
"cols = ['generation biomass', 'generation solar', 'generation nuclear']\n",
"df = df[cols]"
]
},
{
"cell_type": "code",
"execution_count": 5,
"id": "7baaa1c5-c460-44bb-a375-f2eff4509cee",
"metadata": {
"ExecuteTime": {
"end_time": "2023-11-22T14:30:12.888403100Z",
"start_time": "2023-11-22T14:30:12.406885100Z"
},
"papermill": {
"duration": 0.008708,
"end_time": "2024-01-31T17:49:40.842048",
"exception": false,
"start_time": "2024-01-31T17:49:40.833340",
"status": "completed"
},
"tags": []
},
"outputs": [],
"source": [
"ts = on.TimeSeries.from_dataframe(df)"
]
},
{
"cell_type": "markdown",
"id": "07386885-6139-43ed-a695-c2fa3cf23904",
"metadata": {
"papermill": {
"duration": 0.004101,
"end_time": "2024-01-31T17:49:40.850253",
"exception": false,
"start_time": "2024-01-31T17:49:40.846152",
"status": "completed"
},
"tags": []
},
"source": [
"Prepare data"
]
},
{
"cell_type": "code",
"execution_count": 6,
"id": "07b54c1f-4215-43c0-8c81-bb337b9f50fb",
"metadata": {
"ExecuteTime": {
"end_time": "2023-11-22T14:30:12.888403100Z",
"start_time": "2023-11-22T14:30:12.414642100Z"
},
"papermill": {
"duration": 0.010221,
"end_time": "2024-01-31T17:49:40.864515",
"exception": false,
"start_time": "2024-01-31T17:49:40.854294",
"status": "completed"
},
"tags": []
},
"outputs": [],
"source": [
"ts_uni = ts['generation solar'].slice(pd.Timestamp('2015'), pd.Timestamp('2016'))\n",
"ts_multi = ts.slice(pd.Timestamp('2015'), pd.Timestamp('2016'))"
]
},
{
"cell_type": "markdown",
"id": "a1a7ecfe-894b-42a8-8a94-5163f83b26a4",
"metadata": {},
"source": [
"## Primitive Plots"
]
},
{
"cell_type": "markdown",
"id": "a438a78b-3ed5-4501-a316-2414e881dbb3",
"metadata": {},
"source": [
"### Line(s)"
]
},
{
"cell_type": "markdown",
"id": "8c60d9ea-e027-4ba0-8fc6-87bfd8059084",
"metadata": {
"papermill": {
"duration": 0.00435,
"end_time": "2024-01-31T17:49:40.963591",
"exception": false,
"start_time": "2024-01-31T17:49:40.959241",
"status": "completed"
},
"tags": []
},
"source": [
"With univariate TimeSeries"
]
},
{
"cell_type": "code",
"execution_count": 7,
"id": "ad694d28-07ff-4563-81a6-84ffe6591b70",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"\n",
""
],
"text/plain": [
"alt.LayerChart(...)"
]
},
"execution_count": 7,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"on.Plot(ts_uni.head(400))\\\n",
" .add(on.marks.line)\\\n",
" .properties(width=600)\\\n",
" .show()"
]
},
{
"cell_type": "markdown",
"id": "4de943c7-b0fd-4943-8cb1-8ce496599864",
"metadata": {
"papermill": {
"duration": 0.006692,
"end_time": "2024-01-31T17:49:41.118951",
"exception": false,
"start_time": "2024-01-31T17:49:41.112259",
"status": "completed"
},
"tags": []
},
"source": [
"with multivariate TimeSeries"
]
},
{
"cell_type": "code",
"execution_count": 8,
"id": "2a4a0582-209b-461a-815b-7d4d4698c0f7",
"metadata": {
"ExecuteTime": {
"end_time": "2023-11-22T14:30:12.940457Z",
"start_time": "2023-11-22T14:30:12.570624900Z"
},
"papermill": {
"duration": 0.105309,
"end_time": "2024-01-31T17:49:41.230778",
"exception": false,
"start_time": "2024-01-31T17:49:41.125469",
"status": "completed"
},
"tags": []
},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"\n",
""
],
"text/plain": [
"alt.LayerChart(...)"
]
},
"execution_count": 8,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"on.Plot(ts_multi.head(400))\\\n",
" .add(on.marks.line)\\\n",
" .properties(width=600)\\\n",
" .show()"
]
},
{
"cell_type": "markdown",
"id": "2fd66619-22c4-43b5-b827-39b01d16e73b",
"metadata": {},
"source": [
"### Dots"
]
},
{
"cell_type": "markdown",
"id": "e1e97a4e-352d-446f-a70f-66c83e6754e1",
"metadata": {},
"source": [
"With univariate TimeSeries"
]
},
{
"cell_type": "code",
"execution_count": 9,
"id": "f25a59e9-98d3-4519-bc03-feb9f6cd4d2a",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"\n",
""
],
"text/plain": [
"alt.LayerChart(...)"
]
},
"execution_count": 9,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"on.Plot(ts_uni.head(400))\\\n",
" .add(on.marks.dots)\\\n",
" .properties(width=600)\\\n",
" .show()"
]
},
{
"cell_type": "markdown",
"id": "8b7e5730-9341-404d-93a9-f6811d2775d9",
"metadata": {},
"source": [
"with multivariate TimeSeries"
]
},
{
"cell_type": "code",
"execution_count": 10,
"id": "e02804dd-0c24-40e4-9acc-8ed817f03a12",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"\n",
""
],
"text/plain": [
"alt.LayerChart(...)"
]
},
"execution_count": 10,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"on.Plot(ts_multi.head(400))\\\n",
" .add(on.marks.dots)\\\n",
" .properties(width=600)\\\n",
" .show()"
]
},
{
"cell_type": "markdown",
"id": "f9b6835a-2bbf-48d6-935d-fa6226cf45c1",
"metadata": {},
"source": [
"### Areas"
]
},
{
"cell_type": "markdown",
"id": "8c8f8fc6-ccfd-4c3f-b513-e14842e7849c",
"metadata": {},
"source": [
"With a single time series"
]
},
{
"cell_type": "code",
"execution_count": 11,
"id": "7db16759-fbf2-46ef-a0da-4eea94f8ca60",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"\n",
""
],
"text/plain": [
"alt.LayerChart(...)"
]
},
"execution_count": 11,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"on.Plot(ts_uni.head(400))\\\n",
" .add(on.marks.area)\\\n",
" .properties(width=600)\\\n",
" .show()"
]
},
{
"cell_type": "markdown",
"id": "54c8348a-fafe-49c4-939f-77f1a7a2c9d6",
"metadata": {},
"source": [
"With a multivariate time series it works with exactly two"
]
},
{
"cell_type": "code",
"execution_count": 12,
"id": "d6b18ef8-7954-4780-8e9a-dd6e8afd6d15",
"metadata": {},
"outputs": [],
"source": [
"from darts import concatenate"
]
},
{
"cell_type": "code",
"execution_count": 13,
"id": "70178386-e035-4c7a-91c2-ee206131635d",
"metadata": {},
"outputs": [],
"source": [
"# First we create the series with two components\n",
"ts_ci = concatenate([\n",
" ts_multi.univariate_component(0), \n",
" ts_multi.univariate_component(1)\n",
"], axis=1)\n",
"ts_ci = on.TimeSeries.from_darts(ts_ci)"
]
},
{
"cell_type": "code",
"execution_count": 14,
"id": "98ca8c64-bcde-4e93-a0ec-df1410242f36",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"\n",
""
],
"text/plain": [
"alt.LayerChart(...)"
]
},
"execution_count": 14,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Then we plot it\n",
"on.Plot(ts_ci.head(200))\\\n",
" .add(on.marks.area, title='Diff. between solar and biomass generation')\\\n",
" .properties(width=600, height=200)\\\n",
" .show()"
]
},
{
"cell_type": "markdown",
"id": "f436ba6f-aa4f-47e8-9673-754c9c02e1b4",
"metadata": {
"papermill": {
"duration": 0.013993,
"end_time": "2024-01-31T17:49:41.259259",
"exception": false,
"start_time": "2024-01-31T17:49:41.245266",
"status": "completed"
},
"tags": []
},
"source": [
"### Heatmaps"
]
},
{
"attachments": {},
"cell_type": "markdown",
"id": "94b74653-0b50-442f-a666-9bd0984f9781",
"metadata": {
"papermill": {
"duration": 0.014072,
"end_time": "2024-01-31T17:49:41.320059",
"exception": false,
"start_time": "2024-01-31T17:49:41.305987",
"status": "completed"
},
"tags": []
},
"source": [
"with univariate TimeSeries"
]
},
{
"cell_type": "code",
"execution_count": 15,
"id": "8a70db18-2736-4b52-8a49-a09908048535",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"\n",
""
],
"text/plain": [
"alt.LayerChart(...)"
]
},
"execution_count": 15,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"on.Plot(ts_uni.head(1000))\\\n",
" .add(on.marks.heatmap)\\\n",
" .properties(width=600, height=50)\\\n",
" .show()"
]
},
{
"cell_type": "markdown",
"id": "bb43360b-c5ff-4e95-9af8-7ced3fb66d7d",
"metadata": {
"papermill": {
"duration": 0.01433,
"end_time": "2024-01-31T17:49:41.415862",
"exception": false,
"start_time": "2024-01-31T17:49:41.401532",
"status": "completed"
},
"tags": []
},
"source": [
"with multivariate Heatmap"
]
},
{
"cell_type": "code",
"execution_count": 23,
"id": "95e245b4-d594-45f7-ae7a-04db01fc860d",
"metadata": {
"ExecuteTime": {
"end_time": "2023-11-22T14:30:12.943458900Z",
"start_time": "2023-11-22T14:30:12.778200200Z"
},
"papermill": {
"duration": 0.052047,
"end_time": "2024-01-31T17:49:41.386466",
"exception": false,
"start_time": "2024-01-31T17:49:41.334419",
"status": "completed"
},
"tags": []
},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"\n",
""
],
"text/plain": [
"alt.LayerChart(...)"
]
},
"execution_count": 23,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"on.Plot(ts_multi.head(1000))\\\n",
" .add(on.marks.heatmap)\\\n",
" .properties(width=600, height=150)\\\n",
" .show()"
]
},
{
"cell_type": "markdown",
"id": "3b07d001-bef6-4b9b-86b6-105ad318adab",
"metadata": {},
"source": [
"---"
]
},
{
"cell_type": "markdown",
"id": "8a52c3c7-3adf-454e-9b6d-b0959adcd8f6",
"metadata": {},
"source": [
"## Combined Plots"
]
},
{
"cell_type": "markdown",
"id": "c29b1243-bc85-4a8d-899a-710cbfc2d77b",
"metadata": {},
"source": [
"Most of the plots in onTime can be combined as they are based on Altair layered charts. For instance, you can do the following to have a dots on a line."
]
},
{
"cell_type": "code",
"execution_count": 24,
"id": "e18041c8-4c74-4f63-be92-56ad1d971bf7",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"\n",
""
],
"text/plain": [
"alt.LayerChart(...)"
]
},
"execution_count": 24,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"on.Plot()\\\n",
" .add(on.marks.dots, ts_multi.univariate_component(1).head(400))\\\n",
" .add(on.marks.line, ts_multi.univariate_component(0).head(400))\\\n",
" .properties(width=600, height=200)\\\n",
" .show()"
]
},
{
"cell_type": "markdown",
"id": "8c1f4d05-d5b2-4d86-8cf0-6165175354eb",
"metadata": {},
"source": [
"---"
]
},
{
"cell_type": "markdown",
"id": "e87e999c-4ffa-4f3a-92a8-db894982a2ae",
"metadata": {},
"source": [
"## Thematic Plots"
]
},
{
"cell_type": "markdown",
"id": "b65f9e0b-381a-4085-b919-3074b9b18b0a",
"metadata": {},
"source": [
"### Forecasts"
]
},
{
"cell_type": "code",
"execution_count": 25,
"id": "79328857-d576-43d4-a54a-29196312448b",
"metadata": {},
"outputs": [],
"source": [
"ts_train, ts_test = ts_uni.split_before(0.9)"
]
},
{
"cell_type": "code",
"execution_count": 26,
"id": "53bdff9e-f198-4a24-9968-95c3efaf5ebb",
"metadata": {},
"outputs": [],
"source": [
"from ontime.context import common"
]
},
{
"cell_type": "code",
"execution_count": 27,
"id": "c592bccc-95bf-4b2d-9952-f1b4075288da",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
""
]
},
"execution_count": 27,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"model = common.GenericPredictor()\n",
"model.fit(ts_train)"
]
},
{
"cell_type": "code",
"execution_count": 28,
"id": "7a92c059-8854-4d26-a51e-5c44cb7fa1e1",
"metadata": {},
"outputs": [],
"source": [
"ts_pred = model.predict(24 * 3)"
]
},
{
"cell_type": "code",
"execution_count": 29,
"id": "609375fd-e994-4336-b1e8-07ea11b38d49",
"metadata": {},
"outputs": [],
"source": [
"ts_train = ts_train.rename({'generation solar':'Training set'})\n",
"ts_test = ts_test.rename({'generation solar':'Test set'})\n",
"ts_pred = ts_pred.rename({'generation solar':'Forecast'})"
]
},
{
"cell_type": "markdown",
"id": "2f6c72b8-2124-452f-9577-f4d4ece10803",
"metadata": {},
"source": [
"Plot a prediction"
]
},
{
"cell_type": "code",
"execution_count": 30,
"id": "99165fdc-d0a5-49d5-a882-adb8b8534e54",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"\n",
""
],
"text/plain": [
"alt.LayerChart(...)"
]
},
"execution_count": 30,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"(\n",
"on.Plot()\n",
" .add(on.marks.line, ts_test.head(24 * 3), type='dashed')\n",
" .add(on.marks.line, ts_train.tail(24 * 4))\n",
" .add(on.marks.line, ts_pred)\n",
" .properties(width=600, height=200)\n",
" .show()\n",
")"
]
},
{
"cell_type": "markdown",
"id": "8b2d1199-7a74-4586-a834-67fc367229e4",
"metadata": {},
"source": [
"### Anomalies"
]
},
{
"cell_type": "markdown",
"id": "9424f948-96fe-4a51-987a-9c323c1af142",
"metadata": {},
"source": [
"Create the mock data"
]
},
{
"cell_type": "code",
"execution_count": 56,
"id": "0bea5b4f-e11b-4836-a039-d61117a4684d",
"metadata": {},
"outputs": [],
"source": [
"td_point = on.detectors.quantile(high_quantile=0.99)\n",
"td_collective = on.detectors.threshold(low_threshold=-30)\n",
"td_contextual = on.detectors.quantile(high_quantile=0.98)"
]
},
{
"cell_type": "markdown",
"id": "06a7f815-c162-4243-97d9-c8e7e84a6b02",
"metadata": {},
"source": [
"Add anomalies"
]
},
{
"cell_type": "code",
"execution_count": 57,
"id": "e5c7dba4-1ff8-47c9-907b-4bde47ca09dc",
"metadata": {},
"outputs": [],
"source": [
"import numpy as np\n",
"import random\n",
"\n",
"def add_point_anomalies(ts, n, value):\n",
" df = ts.pd_dataframe()\n",
" random_indices = np.random.choice(df.index, size=n, replace=False)\n",
" df.loc[random_indices] = value\n",
" return on.TimeSeries.from_dataframe(df)\n",
"\n",
"def add_collective_anomalies(ts, n, min_duration=10, max_duration=20):\n",
" df = ts.pd_dataframe()\n",
" for i in range(n+1):\n",
" block_duration = random.randint(min_duration, max_duration)\n",
" start_index = np.random.choice(df.index[:-block_duration])\n",
" end_index = start_index + pd.Timedelta(days=block_duration - 1)\n",
" df.loc[start_index:end_index] = -40\n",
" return on.TimeSeries.from_dataframe(df)"
]
},
{
"cell_type": "markdown",
"id": "610cfd2a-c6eb-45bb-a51d-f384dc5a29e9",
"metadata": {},
"source": [
"Select univariate component"
]
},
{
"cell_type": "code",
"execution_count": 58,
"id": "9e89c5ca-3456-446c-8b5e-507574f43df5",
"metadata": {},
"outputs": [],
"source": [
"ts = ts.univariate_component(0)"
]
},
{
"cell_type": "code",
"execution_count": 59,
"id": "03364cbd-61aa-4660-9789-ab7f4662ab79",
"metadata": {},
"outputs": [],
"source": [
"ts = add_point_anomalies(ts, 10, 30)\n",
"ts = add_collective_anomalies(ts, 4)"
]
},
{
"cell_type": "markdown",
"id": "370bdde9-662f-43e7-ab93-cd79b8e2f624",
"metadata": {},
"source": [
"Create binary time series"
]
},
{
"cell_type": "code",
"execution_count": 60,
"id": "4f810223-cfd2-429b-8d7d-6182c26d7246",
"metadata": {},
"outputs": [],
"source": [
"td_point.fit(ts)\n",
"td_contextual.fit(ts)\n",
"\n",
"ts_ano_point = td_point.detect(ts)\n",
"ts_ano_collective = td_collective.detect(ts)\n",
"ts_ano_contextual = td_contextual.detect(ts)"
]
},
{
"cell_type": "code",
"execution_count": 61,
"id": "d0d58fb4-2104-444c-9385-6597beb81a6d",
"metadata": {},
"outputs": [],
"source": [
"ts_ano_point = ts_ano_point.rename({'0': 'Ponctual anomalies'})\n",
"ts_ano_collective = ts_ano_collective.rename({'0': 'Collective anomalies'})\n",
"ts_ano_contextual = ts_ano_contextual.rename({'0': 'Contextual anomalies'})"
]
},
{
"cell_type": "markdown",
"id": "873160d1-4b22-41c5-b374-67b9f9124754",
"metadata": {},
"source": [
"Plot the time series with marked anomalies"
]
},
{
"cell_type": "code",
"execution_count": 64,
"id": "2e944b87-b85d-4c5c-b963-857f79e6b2a4",
"metadata": {},
"outputs": [],
"source": [
"# Define windows for plotting\n",
"start = 24 * 7 * 4 * 9\n",
"duration = 24 * 7 * 10\n",
"end = start + duration"
]
},
{
"cell_type": "code",
"execution_count": 67,
"id": "d4e93bfd-5e3e-40d4-9c45-577eff06579a",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"\n",
""
],
"text/plain": [
"alt.LayerChart(...)"
]
},
"execution_count": 67,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Actually plot\n",
"(\n",
"on.Plot(ts[start:end])\n",
" .add(on.marks.mark, data=ts_ano_contextual[start:end], type='highlight')\n",
" .add(on.marks.mark, data=ts_ano_collective[start:end], type='background')\n",
" .add(on.marks.line)\n",
" .add(on.marks.mark, data=ts_ano_point[start:end], type='dot')\n",
" .properties(width=800, height=200)\n",
" .show()\n",
")"
]
},
{
"cell_type": "markdown",
"id": "ba882e58-d1ba-4bd9-9b8b-11c5257986b2",
"metadata": {},
"source": [
"### Confidence Intervals"
]
},
{
"cell_type": "code",
"execution_count": 91,
"id": "af517b10-9441-4d17-9664-6eb35e4afef3",
"metadata": {},
"outputs": [],
"source": [
"# Generate two time series\n",
"ts1 = on.generators.random_walk().generate(start=pd.Timestamp('2022-01-01'), end=pd.Timestamp('2022-12-31'))\n",
"ts2 = on.generators.random_walk().generate(start=pd.Timestamp('2022-01-01'), end=pd.Timestamp('2022-12-31'))"
]
},
{
"cell_type": "code",
"execution_count": 92,
"id": "0a73d11f-4b21-45a8-8101-79fa3105d293",
"metadata": {},
"outputs": [],
"source": [
"# First we create the series with two components\n",
"ts1_abs = ts1.map(np.abs)\n",
"ts2_abs = ts2.map(np.abs)\n",
"ts_ci = concatenate([ts1_abs, ts2_abs], axis=1)\n",
"ts_ci = on.TimeSeries.from_darts(ts_ci)\n",
"ts_ci = ts_ci.rename({'random_walk': 'CI Upper bound', 'random_walk_1': 'CI Lower bound'})\n",
"\n",
"# Then the hypothetical measurement\n",
"ts_mid = (ts1_abs + ts2_abs) / 2\n",
"ts_mid = ts_mid.rename({'random_walk': 'Measurement'})"
]
},
{
"cell_type": "code",
"execution_count": 93,
"id": "3a12d720-4522-4aac-ae54-5975b52848aa",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"\n",
""
],
"text/plain": [
"alt.LayerChart(...)"
]
},
"execution_count": 93,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Then we plot it\n",
"(\n",
"on.Plot() # main line\n",
" .add(on.marks.area, ts_ci.head(200), title='Confidence interval')\n",
" .add(on.marks.line, ts_mid.head(200))\n",
" .properties(width=600, height=200)\n",
" .show()\n",
")"
]
},
{
"cell_type": "markdown",
"id": "cd9ce3e4-e4a8-40ca-b128-91b80cf1b933",
"metadata": {},
"source": [
"---"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "2da36027-d7dc-4274-9046-a8054c615c0e",
"metadata": {},
"outputs": [],
"source": []
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3 (ipykernel)",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.10.10"
},
"papermill": {
"default_parameters": {},
"duration": 8.089169,
"end_time": "2024-01-31T17:49:45.591711",
"environment_variables": {},
"exception": null,
"input_path": "docs/user_guide/0_core/0.5-plots.ipynb",
"output_path": "docs/user_guide/0_core/0.5-plots.ipynb",
"parameters": {},
"start_time": "2024-01-31T17:49:37.502542",
"version": "2.5.0"
}
},
"nbformat": 4,
"nbformat_minor": 5
}