1 Time-Based Cross-Validation Using TimeSeriesCV and TimeSeriesCVSplitter
+
In this tutorial, youāll learn how to use the TimeSeriesCV and TimeSeriesCVSplitter classes from pytimetk for time series cross-validation, using the walmart_sales_df dataset as an example, which contains 7 time series groups.
+
+
In Part 1, weāll start with exploring the data and move on to creating and visualizing time-based cross-validation splits. This will prepare you for the next section with Scikit Learn.
+
In Part 2, weāll implement time series cross-validation with Scikit-Learn, engineer features, train a random forest model, and visualize the results in Python. By following this process, you can ensure a robust evaluation of your time series models and gain insights into their predictive performance.
+
+
+
+
2 Part 1: Getting Started with TimeSeriesCV
+
TimeSeriesCV is used to generate many time series splits (or folds) for use in modeling and resampling with one or more time series groups contained in the data.
+
+
+
+
+
+
+Using with Scikit Learn
+
+
+
+
+
+
If you are wanting a drop-in replacement for Scikit Learnās TimeSeriesSplit, please use TimeSeriesCVSplitter() discussed next. The splitter uses TimeSeriesCV under the hood.
+
+
+
+
+
2.1 Step 1: Load and Explore the Data
+
First, letās load the Walmart sales dataset and explore its structure:
+
+
# libraries
+import pytimetk as tk
+import pandas as pd
+import numpy as np
+
+# Import Data
+walmart_sales_df = tk.load_dataset('walmart_sales_weekly')
+
+walmart_sales_df['Date'] = pd.to_datetime(walmart_sales_df['Date'])
+
+walmart_sales_df = walmart_sales_df[['id', 'Date', 'Weekly_Sales']]
+
+walmart_sales_df.glimpse()
This will generate an interactive time series plot, allowing you to explore sales data for different stores using a dropdown.
+
+
+
2.3 Step 3: Set Up TimeSeriesCV for Cross-Validation
+
Now, letās set up a time-based cross-validation scheme using TimeSeriesCV:
+
+
from pytimetk.crossvalidation import TimeSeriesCV
+
+# Define parameters for TimeSeriesCV
+tscv = TimeSeriesCV(
+ frequency="weeks",
+ train_size=52, # Use 52 weeks for training
+ forecast_horizon=12, # Forecast 12 weeks ahead
+ gap=0, # No gap between training and forecast sets
+ stride=4, # Move forward by 4 weeks after each split
+ window="rolling", # Use a rolling window
+ mode="backward"# Generate splits from end to start
+)
+
+# Glimpse the cross-validation splits
+tscv.glimpse(
+ walmart_sales_df['Weekly_Sales'],
+ time_series=walmart_sales_df['Date']
+)
This plot will show each fold, illustrating which weeks are used for training and which weeks are used for forecasting.
+
+
+
+
3 Part 2: Using TimeSeriesCVSplitter for Model Evaluation with Scikit Learn
+
When evaluating a modelās predictive performance on time series data, we need to split the data in a way that respects the order of time within the Scikit Learn framework. We use a custom splitter, TimeSeriesCVSplitter, from the pytimetk library to handle this.
+
+
3.1 Step 1: Setting Up the TimeSeriesCVSplitter
+
The TimeSeriesCVSplitter helps us divide our dataset into training and forecast sets in a rolling window fashion. Hereās how we configure it:
The TimeSeriesCVSplitter creates multiple splits of the time series data, allowing us to validate the model across different periods. By visualizing the cross-validation strategy, we can see how the training and forecast sets are structured.
+
+
+
3.2 Step 2: Feature Engineering for Time Series Data
+
Effective feature engineering can significantly impact the performance of a time series model. Using pytimetk, we extract a variety of features from the Date column.
+
+
Generating Time Series Features
+
We use get_timeseries_signature to generate useful features, such as year, quarter, month, and day-of-week indicators.
+
+
# Prepare data for modeling
+
+# Extract time series features from the 'Date' column
+X_time_features = tk.get_timeseries_signature(walmart_sales_df['Date'])
+
+# Select features to dummy encode
+features_to_dummy = ['Date_quarteryear', 'Date_month_lbl', 'Date_wday_lbl', 'Date_am_pm']
+
+# Dummy encode the selected features
+X_time_dummies = pd.get_dummies(X_time_features[features_to_dummy], drop_first=True)
+
+# Dummy encode the 'id' column
+X_id_dummies = pd.get_dummies(walmart_sales_df['id'], prefix='store')
+
+# Combine the time series features, dummy-encoded features, and the 'id' dummies
+X = pd.concat([X_time_features, X_time_dummies, X_id_dummies], axis=1)
+
+# Drop the original categorical columns that were dummy encoded
+X = X.drop(columns=features_to_dummy).drop('Date', axis=1)
+
+# Set the target variable
+y = walmart_sales_df['Weekly_Sales'].values
+
+
+
+
+
3.3 Step 3: Model Training and Evaluation with Random Forest
+
For this example, we use RandomForestRegressor from scikit-learn to model the time series data. A random forest is a robust, ensemble-based model that can handle a wide range of regression tasks.
+
+
# Initialize the RandomForestRegressor model
+model = RandomForestRegressor(
+ n_estimators=100, # Number of trees in the forest
+ max_depth=None, # Maximum depth of the trees (None means nodes are expanded until all leaves are pure)
+ random_state=42# Set a random state for reproducibility
+)
+
+# Evaluate the model using cross-validation scores
+scores = cross_val_score(model, X, y, cv=cv_splitter, scoring='neg_mean_squared_error')
+
+# Print cross-validation scores
+print("Cross-Validation Scores (Negative MSE):", scores)
Visualization is crucial to understand how well the model predicts future values. We collect the actual and predicted values for each fold and combine them for easy plotting.
+
+
# Lists to store the combined data
+combined_data = []
+
+# Iterate through each fold and collect the data
+for i, (train_index, test_index) inenumerate(cv_splitter.split(X, y), start=1):
+# Get the training and forecast data from the original DataFrame
+ train_df = walmart_sales_df.iloc[train_index].copy()
+ test_df = walmart_sales_df.iloc[test_index].copy()
+
+# Fit the model on the training data
+ model.fit(X.iloc[train_index], y[train_index])
+
+# Predict on the test set
+ y_pred = model.predict(X.iloc[test_index])
+
+# Add the actual and predicted values
+ train_df['Actual'] = y[train_index]
+ train_df['Predicted'] =None# No predictions for training data
+ train_df['Fold'] = i # Indicate the current fold
+
+ test_df['Actual'] = y[test_index]
+ test_df['Predicted'] = y_pred # Predictions for the test data
+ test_df['Fold'] = i # Indicate the current fold
+
+# Append both the training and forecast DataFrames to the combined data list
+ combined_data.extend([train_df, test_df])
+
+# Combine all the data into a single DataFrame
+full_forecast_df = pd.concat(combined_data, ignore_index=True)
+
+full_forecast_df = full_forecast_df[['id', 'Date', 'Actual', 'Predicted', 'Fold']]
+
+full_forecast_df.glimpse()
To make the data easier to plot, we use pd.melt() to transform the Actual and Predicted columns into a long format.
+
+
# Melt the Actual and Predicted columns
+melted_df = pd.melt(
+ full_forecast_df,
+ id_vars=['id', 'Date', 'Fold'], # Columns to keep
+ value_vars=['Actual', 'Predicted'], # Columns to melt
+ var_name='Type', # Name for the new column indicating 'Actual' or 'Predicted'
+ value_name='Value'# Name for the new column with the values
+)
+
+melted_df["unique_id"] ="ID_"+ melted_df['id'] +"-Fold_"+ melted_df["Fold"].astype(str)
+
+melted_df.glimpse()
This guide demonstrated how to implement time series cross-validation, engineer features, train a random forest model, and visualize the results in Python. By following this process, you can ensure a robust evaluation of your time series models and gain insights into their predictive performance. Happy modeling!
+
+
+
5 More Coming Soonā¦
+
We are in the early stages of development. But itās obvious the potential for pytimetk now in Python. š
diff --git a/docs/_site/search.json b/docs/_site/search.json
index 83e50c12..7949ee54 100644
--- a/docs/_site/search.json
+++ b/docs/_site/search.json
@@ -1400,102 +1400,67 @@
"text": "3.2 Pad by Time with Grouped Time Series\npad_by_time() can also be used with grouped time series. Letās use the stocks_daily dataset to showcase an example:\n\n\nCode\n# load dataset\nstocks_df = tk.load_dataset('stocks_daily', parse_dates = ['date'])\n\n# pad by time\nstocks_df \\\n .groupby('symbol') \\\n .pad_by_time(\n date_column = 'date',\n freq = 'D'\n ) \\\n .assign(id = lambda x: x['symbol'].ffill())\n\n\n\n\n\n\n\n\n\nsymbol\ndate\nopen\nhigh\nlow\nclose\nvolume\nadjusted\nid\n\n\n\n\n0\nAAPL\n2013-01-02\n19.779285\n19.821428\n19.343929\n19.608213\n560518000.0\n16.791180\nAAPL\n\n\n1\nAAPL\n2013-01-03\n19.567142\n19.631071\n19.321428\n19.360714\n352965200.0\n16.579241\nAAPL\n\n\n2\nAAPL\n2013-01-04\n19.177500\n19.236786\n18.779642\n18.821428\n594333600.0\n16.117437\nAAPL\n\n\n3\nAAPL\n2013-01-05\nNaN\nNaN\nNaN\nNaN\nNaN\nNaN\nAAPL\n\n\n4\nAAPL\n2013-01-06\nNaN\nNaN\nNaN\nNaN\nNaN\nNaN\nAAPL\n\n\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n\n\n23485\nNVDA\n2023-09-17\nNaN\nNaN\nNaN\nNaN\nNaN\nNaN\nNVDA\n\n\n23486\nNVDA\n2023-09-18\n427.480011\n442.420013\n420.000000\n439.660004\n50027100.0\n439.660004\nNVDA\n\n\n23487\nNVDA\n2023-09-19\n438.329987\n439.660004\n430.019989\n435.200012\n37306400.0\n435.200012\nNVDA\n\n\n23488\nNVDA\n2023-09-20\n436.000000\n439.029999\n422.230011\n422.390015\n36710800.0\n422.390015\nNVDA\n\n\n23489\nNVDA\n2023-09-21\n415.829987\n421.000000\n409.799988\n410.170013\n44893000.0\n410.170013\nNVDA\n\n\n\n\n23490 rows Ć 9 columns\n\n\n\nTo replace NaN with 0 in a dataframe with multiple columns:\n\n\nCode\nfrom functools import partial\n\n# columns to replace NaN with 0\ncols_to_fill = ['open', 'high', 'low', 'close', 'volume', 'adjusted']\n\n# define a function to fillna\ndef fill_na_col(df, col):\n return df[col].fillna(0)\n\n# pad by time and replace NaN with 0\nstocks_df \\\n .groupby('symbol') \\\n .pad_by_time(\n date_column = 'date',\n freq = 'D'\n ) \\\n .assign(id = lambda x: x['symbol'].ffill()) \\\n .assign(**{col: partial(fill_na_col, col=col) for col in cols_to_fill})\n\n\n\n\n\n\n\n\n\nsymbol\ndate\nopen\nhigh\nlow\nclose\nvolume\nadjusted\nid\n\n\n\n\n0\nAAPL\n2013-01-02\n19.779285\n19.821428\n19.343929\n19.608213\n560518000.0\n16.791180\nAAPL\n\n\n1\nAAPL\n2013-01-03\n19.567142\n19.631071\n19.321428\n19.360714\n352965200.0\n16.579241\nAAPL\n\n\n2\nAAPL\n2013-01-04\n19.177500\n19.236786\n18.779642\n18.821428\n594333600.0\n16.117437\nAAPL\n\n\n3\nAAPL\n2013-01-05\n0.000000\n0.000000\n0.000000\n0.000000\n0.0\n0.000000\nAAPL\n\n\n4\nAAPL\n2013-01-06\n0.000000\n0.000000\n0.000000\n0.000000\n0.0\n0.000000\nAAPL\n\n\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n\n\n23485\nNVDA\n2023-09-17\n0.000000\n0.000000\n0.000000\n0.000000\n0.0\n0.000000\nNVDA\n\n\n23486\nNVDA\n2023-09-18\n427.480011\n442.420013\n420.000000\n439.660004\n50027100.0\n439.660004\nNVDA\n\n\n23487\nNVDA\n2023-09-19\n438.329987\n439.660004\n430.019989\n435.200012\n37306400.0\n435.200012\nNVDA\n\n\n23488\nNVDA\n2023-09-20\n436.000000\n439.029999\n422.230011\n422.390015\n36710800.0\n422.390015\nNVDA\n\n\n23489\nNVDA\n2023-09-21\n415.829987\n421.000000\n409.799988\n410.170013\n44893000.0\n410.170013\nNVDA\n\n\n\n\n23490 rows Ć 9 columns"
},
{
- "objectID": "guides/01_visualization.html",
- "href": "guides/01_visualization.html",
- "title": "Data Visualization",
+ "objectID": "guides/07_timeseries_crossvalidation.html",
+ "href": "guides/07_timeseries_crossvalidation.html",
+ "title": "Time Series Cross Validation",
"section": "",
- "text": "How this guide benefits you\n\n\n\n\n\nThis guide covers how to use the plot_timeseries() for data visualization. Once you understand how it works, you can apply explore time series data easier than ever.\nThis tutorial focuses on, plot_timeseries(), a workhorse time-series plotting function that:"
+ "text": "In this tutorial, youāll learn how to use the TimeSeriesCV and TimeSeriesCVSplitter classes from pytimetk for time series cross-validation, using the walmart_sales_df dataset as an example, which contains 7 time series groups.\n\nIn Part 1, weāll start with exploring the data and move on to creating and visualizing time-based cross-validation splits. This will prepare you for the next section with Scikit Learn.\nIn Part 2, weāll implement time series cross-validation with Scikit-Learn, engineer features, train a random forest model, and visualize the results in Python. By following this process, you can ensure a robust evaluation of your time series models and gain insights into their predictive performance."
},
{
- "objectID": "guides/01_visualization.html#plotting-a-single-time-series",
- "href": "guides/01_visualization.html#plotting-a-single-time-series",
- "title": "Data Visualization",
- "section": "2.1 Plotting a Single Time Series",
- "text": "2.1 Plotting a Single Time Series\nLetās start with a popular time series, taylor_30_min, which includes energy demand in megawatts at a sampling interval of 30-minutes. This is a single time series.\n\n\nCode\n# Import a Time Series Data Set\ntaylor_30_min = tk.load_dataset(\"taylor_30_min\", parse_dates = ['date'])\ntaylor_30_min\n\n\n\n\n\n\n\n\n\ndate\nvalue\n\n\n\n\n0\n2000-06-05 00:00:00+00:00\n22262\n\n\n1\n2000-06-05 00:30:00+00:00\n21756\n\n\n2\n2000-06-05 01:00:00+00:00\n22247\n\n\n3\n2000-06-05 01:30:00+00:00\n22759\n\n\n4\n2000-06-05 02:00:00+00:00\n22549\n\n\n...\n...\n...\n\n\n4027\n2000-08-27 21:30:00+00:00\n27946\n\n\n4028\n2000-08-27 22:00:00+00:00\n27133\n\n\n4029\n2000-08-27 22:30:00+00:00\n25996\n\n\n4030\n2000-08-27 23:00:00+00:00\n24610\n\n\n4031\n2000-08-27 23:30:00+00:00\n23132\n\n\n\n\n4032 rows Ć 2 columns\n\n\n\nThe plot_timeseries() function generates an interactive plotly chart by default.\n\nSimply provide the date variable (time-based column, date_column) and the numeric variable (value_column) that changes over time as the first 2 arguments.\nBy default, the plotting engine is plotly, which is interactive and excellent for data exploration and apps. However, if you require static plots for reports, you can set the engine to engine = āplotnineā or engine = āmatplotlibā.\n\nInteractive plot\n\n\nCode\ntaylor_30_min.plot_timeseries('date', 'value')\n\n\n\n \n\n\nStatic plot\n\n\nCode\ntaylor_30_min.plot_timeseries(\n 'date', 'value',\n engine = 'plotnine'\n)\n\n\n\n\n\n<Figure Size: (700 x 500)>"
+ "objectID": "guides/07_timeseries_crossvalidation.html#step-1-load-and-explore-the-data",
+ "href": "guides/07_timeseries_crossvalidation.html#step-1-load-and-explore-the-data",
+ "title": "Time Series Cross Validation",
+ "section": "2.1 Step 1: Load and Explore the Data",
+ "text": "2.1 Step 1: Load and Explore the Data\nFirst, letās load the Walmart sales dataset and explore its structure:\n\n# libraries\nimport pytimetk as tk\nimport pandas as pd\nimport numpy as np\n\n# Import Data\nwalmart_sales_df = tk.load_dataset('walmart_sales_weekly')\n\nwalmart_sales_df['Date'] = pd.to_datetime(walmart_sales_df['Date'])\n\nwalmart_sales_df = walmart_sales_df[['id', 'Date', 'Weekly_Sales']]\n\nwalmart_sales_df.glimpse()\n\n<class 'pandas.core.frame.DataFrame'>: 1001 rows of 3 columns\nid: object ['1_1', '1_1', '1_1', '1_1', '1_1', '1_ ...\nDate: datetime64[ns] [Timestamp('2010-02-05 00:00:00'), Time ...\nWeekly_Sales: float64 [24924.5, 46039.49, 41595.55, 19403.54, ..."
},
{
- "objectID": "guides/01_visualization.html#plotting-groups",
- "href": "guides/01_visualization.html#plotting-groups",
- "title": "Data Visualization",
- "section": "2.2 Plotting Groups",
- "text": "2.2 Plotting Groups\nNext, letās move on to a dataset with time series groups, m4_monthly, which is a sample of 4 time series from the M4 competition that are sampled at a monthly frequency.\n\n\nCode\n# Import a Time Series Data Set\nm4_monthly = tk.load_dataset(\"m4_monthly\", parse_dates = ['date'])\nm4_monthly\n\n\n\n\n\n\n\n\n\nid\ndate\nvalue\n\n\n\n\n0\nM1\n1976-06-01\n8000\n\n\n1\nM1\n1976-07-01\n8350\n\n\n2\nM1\n1976-08-01\n8570\n\n\n3\nM1\n1976-09-01\n7700\n\n\n4\nM1\n1976-10-01\n7080\n\n\n...\n...\n...\n...\n\n\n1569\nM1000\n2015-02-01\n880\n\n\n1570\nM1000\n2015-03-01\n800\n\n\n1571\nM1000\n2015-04-01\n1140\n\n\n1572\nM1000\n2015-05-01\n970\n\n\n1573\nM1000\n2015-06-01\n1430\n\n\n\n\n1574 rows Ć 3 columns\n\n\n\nVisualizing grouped data is as simple as grouping the data set with groupby() before run it into the plot_timeseries() function. There are 2 methods:\n\nFacets\nPlotly Dropdown\n\n\nFacets (Subgroups on one plot)\nThis is great to see all time series in one plot. Here are the key points:\n\nGroups can be added using the pandas groupby().\nThese groups are then converted into facets.\nUsing facet_ncol = 2 returns a 2-column faceted plot.\nSetting facet_scales = \"free\" allows the x and y-axes of each plot to scale independently of the other plots.\n\n\n\nCode\nm4_monthly.groupby('id').plot_timeseries(\n 'date', 'value', \n facet_ncol = 2, \n facet_scales = \"free\"\n)\n\n\n\n \n\n\n\n\nPlotly Dropdown\nSometimes you have many groups and would prefer to see one plot per group. This can be accomplished with plotly_dropdown. You can adjust the x and y position as follows:\n\n\nCode\nm4_monthly.groupby('id').plot_timeseries(\n 'date', 'value', \n plotly_dropdown=True,\n plotly_dropdown_x=0,\n plotly_dropdown_y=1\n)\n\n\n\n \n\n\nThe groups can also be vizualized in the same plot using color_column paramenter. Letās come back to taylor_30_min dataframe.\n\n\nCode\n# load data\ntaylor_30_min = tk.load_dataset(\"taylor_30_min\", parse_dates = ['date'])\n\n# extract the month using pandas\ntaylor_30_min['month'] = pd.to_datetime(taylor_30_min['date']).dt.month\n\n# plot groups\ntaylor_30_min.plot_timeseries(\n 'date', 'value', \n color_column = 'month'\n)"
+ "objectID": "guides/07_timeseries_crossvalidation.html#step-2-visualize-the-time-series-data",
+ "href": "guides/07_timeseries_crossvalidation.html#step-2-visualize-the-time-series-data",
+ "title": "Time Series Cross Validation",
+ "section": "2.2 Step 2: Visualize the Time Series Data",
+ "text": "2.2 Step 2: Visualize the Time Series Data\nWe can visualize the weekly sales data for different store IDs using the plot_timeseries method from pytimetk:\n\nwalmart_sales_df \\\n .groupby('id') \\\n .plot_timeseries(\n \"Date\", \"Weekly_Sales\",\n plotly_dropdown = True,\n )\n\n\n \n\n\nThis will generate an interactive time series plot, allowing you to explore sales data for different stores using a dropdown."
},
{
- "objectID": "guides/02_timetk_concepts.html",
- "href": "guides/02_timetk_concepts.html",
- "title": "PyTimeTK Basics",
- "section": "",
- "text": "PyTimeTK has one mission: To make time series analysis simpler, easier, and faster in Python. This goal requires some opinionated ways of treating time series in Python. We will conceptually lay out how pytimetk can help.\nLetās first start with how to think about time series data conceptually. Time series data has 3 core properties."
+ "objectID": "guides/07_timeseries_crossvalidation.html#step-3-set-up-timeseriescv-for-cross-validation",
+ "href": "guides/07_timeseries_crossvalidation.html#step-3-set-up-timeseriescv-for-cross-validation",
+ "title": "Time Series Cross Validation",
+ "section": "2.3 Step 3: Set Up TimeSeriesCV for Cross-Validation",
+ "text": "2.3 Step 3: Set Up TimeSeriesCV for Cross-Validation\nNow, letās set up a time-based cross-validation scheme using TimeSeriesCV:\n\nfrom pytimetk.crossvalidation import TimeSeriesCV\n\n# Define parameters for TimeSeriesCV\ntscv = TimeSeriesCV(\n frequency=\"weeks\",\n train_size=52, # Use 52 weeks for training\n forecast_horizon=12, # Forecast 12 weeks ahead\n gap=0, # No gap between training and forecast sets\n stride=4, # Move forward by 4 weeks after each split\n window=\"rolling\", # Use a rolling window\n mode=\"backward\" # Generate splits from end to start\n)\n\n# Glimpse the cross-validation splits\ntscv.glimpse(\n walmart_sales_df['Weekly_Sales'], \n time_series=walmart_sales_df['Date']\n)\n\nSplit Number: 1\nTrain Shape: (364,), Forecast Shape: (84,)\nTrain Period: 2011-08-05 00:00:00 to 2012-07-27 00:00:00\nForecast Period: 2012-08-03 00:00:00 to 2012-10-19 00:00:00\n\nSplit Number: 2\nTrain Shape: (364,), Forecast Shape: (84,)\nTrain Period: 2011-07-08 00:00:00 to 2012-06-29 00:00:00\nForecast Period: 2012-07-06 00:00:00 to 2012-09-21 00:00:00\n\nSplit Number: 3\nTrain Shape: (364,), Forecast Shape: (84,)\nTrain Period: 2011-06-10 00:00:00 to 2012-06-01 00:00:00\nForecast Period: 2012-06-08 00:00:00 to 2012-08-24 00:00:00\n\nSplit Number: 4\nTrain Shape: (364,), Forecast Shape: (84,)\nTrain Period: 2011-05-13 00:00:00 to 2012-05-04 00:00:00\nForecast Period: 2012-05-11 00:00:00 to 2012-07-27 00:00:00\n\nSplit Number: 5\nTrain Shape: (364,), Forecast Shape: (84,)\nTrain Period: 2011-04-15 00:00:00 to 2012-04-06 00:00:00\nForecast Period: 2012-04-13 00:00:00 to 2012-06-29 00:00:00\n\nSplit Number: 6\nTrain Shape: (364,), Forecast Shape: (84,)\nTrain Period: 2011-03-18 00:00:00 to 2012-03-09 00:00:00\nForecast Period: 2012-03-16 00:00:00 to 2012-06-01 00:00:00\n\nSplit Number: 7\nTrain Shape: (364,), Forecast Shape: (84,)\nTrain Period: 2011-02-18 00:00:00 to 2012-02-10 00:00:00\nForecast Period: 2012-02-17 00:00:00 to 2012-05-04 00:00:00\n\nSplit Number: 8\nTrain Shape: (364,), Forecast Shape: (84,)\nTrain Period: 2011-01-21 00:00:00 to 2012-01-13 00:00:00\nForecast Period: 2012-01-20 00:00:00 to 2012-04-06 00:00:00\n\nSplit Number: 9\nTrain Shape: (364,), Forecast Shape: (84,)\nTrain Period: 2010-12-24 00:00:00 to 2011-12-16 00:00:00\nForecast Period: 2011-12-23 00:00:00 to 2012-03-09 00:00:00\n\nSplit Number: 10\nTrain Shape: (364,), Forecast Shape: (84,)\nTrain Period: 2010-11-26 00:00:00 to 2011-11-18 00:00:00\nForecast Period: 2011-11-25 00:00:00 to 2012-02-10 00:00:00\n\nSplit Number: 11\nTrain Shape: (364,), Forecast Shape: (84,)\nTrain Period: 2010-10-29 00:00:00 to 2011-10-21 00:00:00\nForecast Period: 2011-10-28 00:00:00 to 2012-01-13 00:00:00\n\nSplit Number: 12\nTrain Shape: (364,), Forecast Shape: (84,)\nTrain Period: 2010-10-01 00:00:00 to 2011-09-23 00:00:00\nForecast Period: 2011-09-30 00:00:00 to 2011-12-16 00:00:00\n\nSplit Number: 13\nTrain Shape: (364,), Forecast Shape: (84,)\nTrain Period: 2010-09-03 00:00:00 to 2011-08-26 00:00:00\nForecast Period: 2011-09-02 00:00:00 to 2011-11-18 00:00:00\n\nSplit Number: 14\nTrain Shape: (364,), Forecast Shape: (84,)\nTrain Period: 2010-08-06 00:00:00 to 2011-07-29 00:00:00\nForecast Period: 2011-08-05 00:00:00 to 2011-10-21 00:00:00\n\nSplit Number: 15\nTrain Shape: (364,), Forecast Shape: (84,)\nTrain Period: 2010-07-09 00:00:00 to 2011-07-01 00:00:00\nForecast Period: 2011-07-08 00:00:00 to 2011-09-23 00:00:00\n\nSplit Number: 16\nTrain Shape: (364,), Forecast Shape: (84,)\nTrain Period: 2010-06-11 00:00:00 to 2011-06-03 00:00:00\nForecast Period: 2011-06-10 00:00:00 to 2011-08-26 00:00:00\n\nSplit Number: 17\nTrain Shape: (364,), Forecast Shape: (84,)\nTrain Period: 2010-05-14 00:00:00 to 2011-05-06 00:00:00\nForecast Period: 2011-05-13 00:00:00 to 2011-07-29 00:00:00\n\nSplit Number: 18\nTrain Shape: (364,), Forecast Shape: (84,)\nTrain Period: 2010-04-16 00:00:00 to 2011-04-08 00:00:00\nForecast Period: 2011-04-15 00:00:00 to 2011-07-01 00:00:00\n\nSplit Number: 19\nTrain Shape: (364,), Forecast Shape: (84,)\nTrain Period: 2010-03-19 00:00:00 to 2011-03-11 00:00:00\nForecast Period: 2011-03-18 00:00:00 to 2011-06-03 00:00:00\n\nSplit Number: 20\nTrain Shape: (364,), Forecast Shape: (84,)\nTrain Period: 2010-02-19 00:00:00 to 2011-02-11 00:00:00\nForecast Period: 2011-02-18 00:00:00 to 2011-05-06 00:00:00\n\n\n\nThe glimpse method provides a summary of each cross-validation fold, including the start and end dates of the training and forecast periods."
},
{
- "objectID": "guides/02_timetk_concepts.html#type-1-pandas-dataframe-operations",
- "href": "guides/02_timetk_concepts.html#type-1-pandas-dataframe-operations",
- "title": "PyTimeTK Basics",
- "section": "2.1 Type 1: Pandas DataFrame Operations",
- "text": "2.1 Type 1: Pandas DataFrame Operations\nBefore we start using pytimetk, letās make sure our data is set up properly.\n\nTimetk Data Format Compliance\n\n\n\n\n\n\n3 Core Properties Must Be Upheald\n\n\n\n\n\nA pytimetk-Compliant Pandas DataFrame must have:\n\nTime Series Index: A Time Stamp column containing datetime64 values\nValue Column(s): The value column(s) containing float or int values\nGroup Column(s): Optionally for grouped time series analysis, one or more columns containg str or categorical values (shown as an object)\n\nIf these are NOT upheld, this will impact your ability to use pytimetk DataFrame operations.\n\n\n\n\n\n\n\n\n\nInspect the DataFrame\n\n\n\n\n\nUse the tk.glimpse() method to check compliance.\n\n\n\nUsing pytimetk glimpse() method, we can see that we have a compliant data frame with a date column containing datetime64 and a value column containing float64. For grouped analysis we have the id column containing object dtype.\n\n\nCode\n# Tip: Inspect for compliance with glimpse()\nm4_daily_df.glimpse()\n\n\n<class 'pandas.core.frame.DataFrame'>: 9743 rows of 3 columns\nid: object ['D10', 'D10', 'D10', 'D10', 'D10', 'D10', 'D1 ...\ndate: datetime64[ns] [Timestamp('2014-07-03 00:00:00'), Timestamp(' ...\nvalue: float64 [2076.2, 2073.4, 2048.7, 2048.9, 2006.4, 2017. ...\n\n\n\n\nGrouped Time Series Analysis with Summarize By Time\nFirst, inspect how the summarize_by_time function works by calling help().\n\n\nCode\n# Review the summarize_by_time documentation (output not shown)\nhelp(tk.summarize_by_time)\n\n\n\n\n\n\n\n\nHelp Doc Info: summarize_by_time()\n\n\n\n\n\n\nThe first parameter is data, indicating this is a DataFrame operation.\nThe Examples show different use cases for how to apply the function on a DataFrame\n\n\n\n\nLetās test the summarize_by_time() DataFrame operation out using the grouped approach with method chaining. DataFrame operations can be used as Pandas methods with method-chaining, which allows us to more succinctly apply time series operations.\n\n\nCode\n# Grouped Summarize By Time with Method Chaining\ndf_summarized = (\n m4_daily_df\n .groupby('id')\n .summarize_by_time(\n date_column = 'date',\n value_column = 'value',\n freq = 'QS', # QS = Quarter Start\n agg_func = [\n 'mean', \n 'median', \n 'min',\n ('q25', lambda x: np.quantile(x, 0.25)),\n ('q75', lambda x: np.quantile(x, 0.75)),\n 'max',\n ('range',lambda x: x.max() - x.min()),\n ],\n )\n)\n\ndf_summarized\n\n\n\n\n\n\n\n\n\nid\ndate\nvalue_mean\nvalue_median\nvalue_min\nvalue_q25\nvalue_q75\nvalue_max\nvalue_range\n\n\n\n\n0\nD10\n2014-07-01\n1960.078889\n1979.90\n1781.6\n1915.225\n2002.575\n2076.2\n294.6\n\n\n1\nD10\n2014-10-01\n2184.586957\n2154.05\n2022.8\n2125.075\n2274.150\n2344.9\n322.1\n\n\n2\nD10\n2015-01-01\n2309.830000\n2312.30\n2209.6\n2284.575\n2342.150\n2392.4\n182.8\n\n\n3\nD10\n2015-04-01\n2344.481319\n2333.00\n2185.1\n2301.750\n2391.000\n2499.8\n314.7\n\n\n4\nD10\n2015-07-01\n2156.754348\n2186.70\n1856.6\n1997.250\n2289.425\n2368.1\n511.5\n\n\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n\n\n105\nD500\n2011-07-01\n9727.321739\n9745.55\n8964.5\n9534.125\n10003.900\n10463.9\n1499.4\n\n\n106\nD500\n2011-10-01\n8175.565217\n7897.00\n6755.0\n7669.875\n8592.575\n9860.0\n3105.0\n\n\n107\nD500\n2012-01-01\n8291.317582\n8412.60\n7471.5\n7814.800\n8677.850\n8980.7\n1509.2\n\n\n108\nD500\n2012-04-01\n8654.020879\n8471.10\n8245.6\n8389.850\n9017.250\n9349.2\n1103.6\n\n\n109\nD500\n2012-07-01\n8770.502353\n8690.50\n8348.1\n8604.400\n8846.000\n9545.3\n1197.2\n\n\n\n\n110 rows Ć 9 columns\n\n\n\n\n\n\n\n\n\nKey Takeaways: summarize_by_time()\n\n\n\n\n\n\nThe data must comply with the 3 core properties (date column, value column(s), and group column(s))\nThe aggregation functions were applied by combination of group (id) and resample (Quarter Start)\nThe result was a pandas DataFrame with group column, resampled date column, and summary values (mean, median, min, 25th-quantile, etc)\n\n\n\n\n\n\nAnother DataFrame Example: Creating 29 Engineered Features\nLetās examine another DataFrame function, tk.augment_timeseries_signature(). Feel free to inspect the documentation with help(tk.augment_timeseries_signature).\n\n\nCode\n# Creating 29 engineered features from the date column\n# Not run: help(tk.augment_timeseries_signature)\ndf_augmented = (\n m4_daily_df\n .augment_timeseries_signature(date_column = 'date')\n)\n\ndf_augmented.head()\n\n\n\n\n\n\n\n\n\nid\ndate\nvalue\ndate_index_num\ndate_year\ndate_year_iso\ndate_yearstart\ndate_yearend\ndate_leapyear\ndate_half\n...\ndate_mday\ndate_qday\ndate_yday\ndate_weekend\ndate_hour\ndate_minute\ndate_second\ndate_msecond\ndate_nsecond\ndate_am_pm\n\n\n\n\n0\nD10\n2014-07-03\n2076.2\n1404345600\n2014\n2014\n0\n0\n0\n2\n...\n3\n3\n184\n0\n0\n0\n0\n0\n0\nam\n\n\n1\nD10\n2014-07-04\n2073.4\n1404432000\n2014\n2014\n0\n0\n0\n2\n...\n4\n4\n185\n0\n0\n0\n0\n0\n0\nam\n\n\n2\nD10\n2014-07-05\n2048.7\n1404518400\n2014\n2014\n0\n0\n0\n2\n...\n5\n5\n186\n0\n0\n0\n0\n0\n0\nam\n\n\n3\nD10\n2014-07-06\n2048.9\n1404604800\n2014\n2014\n0\n0\n0\n2\n...\n6\n6\n187\n1\n0\n0\n0\n0\n0\nam\n\n\n4\nD10\n2014-07-07\n2006.4\n1404691200\n2014\n2014\n0\n0\n0\n2\n...\n7\n7\n188\n0\n0\n0\n0\n0\n0\nam\n\n\n\n\n5 rows Ć 32 columns\n\n\n\n\n\n\n\n\n\nKey Takeaways: augment_timeseries_signature()\n\n\n\n\n\n\nThe data must comply with the 1 of the 3 core properties (date column)\nThe result was a pandas DataFrame with 29 time series features that can be used for Machine Learning and Forecasting\n\n\n\n\n\n\nMaking Future Dates with Future Frame\nA common time series task before forecasting with machine learning models is to make a future DataFrame some length_out into the future. You can do this with tk.future_frame(). Hereās how.\n\n\nCode\n# Preparing a time series data set for Machine Learning Forecasting\nfull_augmented_df = (\n m4_daily_df \n .groupby('id')\n .future_frame('date', length_out = 365)\n .augment_timeseries_signature('date')\n)\nfull_augmented_df\n\n\n\n\n\n\n\n\n\n\n\n\nid\ndate\nvalue\ndate_index_num\ndate_year\ndate_year_iso\ndate_yearstart\ndate_yearend\ndate_leapyear\ndate_half\n...\ndate_mday\ndate_qday\ndate_yday\ndate_weekend\ndate_hour\ndate_minute\ndate_second\ndate_msecond\ndate_nsecond\ndate_am_pm\n\n\n\n\n0\nD10\n2014-07-03\n2076.2\n1404345600\n2014\n2014\n0\n0\n0\n2\n...\n3\n3\n184\n0\n0\n0\n0\n0\n0\nam\n\n\n1\nD10\n2014-07-04\n2073.4\n1404432000\n2014\n2014\n0\n0\n0\n2\n...\n4\n4\n185\n0\n0\n0\n0\n0\n0\nam\n\n\n2\nD10\n2014-07-05\n2048.7\n1404518400\n2014\n2014\n0\n0\n0\n2\n...\n5\n5\n186\n0\n0\n0\n0\n0\n0\nam\n\n\n3\nD10\n2014-07-06\n2048.9\n1404604800\n2014\n2014\n0\n0\n0\n2\n...\n6\n6\n187\n1\n0\n0\n0\n0\n0\nam\n\n\n4\nD10\n2014-07-07\n2006.4\n1404691200\n2014\n2014\n0\n0\n0\n2\n...\n7\n7\n188\n0\n0\n0\n0\n0\n0\nam\n\n\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n\n\n11198\nD500\n2013-09-19\nNaN\n1379548800\n2013\n2013\n0\n0\n0\n2\n...\n19\n81\n262\n0\n0\n0\n0\n0\n0\nam\n\n\n11199\nD500\n2013-09-20\nNaN\n1379635200\n2013\n2013\n0\n0\n0\n2\n...\n20\n82\n263\n0\n0\n0\n0\n0\n0\nam\n\n\n11200\nD500\n2013-09-21\nNaN\n1379721600\n2013\n2013\n0\n0\n0\n2\n...\n21\n83\n264\n0\n0\n0\n0\n0\n0\nam\n\n\n11201\nD500\n2013-09-22\nNaN\n1379808000\n2013\n2013\n0\n0\n0\n2\n...\n22\n84\n265\n1\n0\n0\n0\n0\n0\nam\n\n\n11202\nD500\n2013-09-23\nNaN\n1379894400\n2013\n2013\n0\n0\n0\n2\n...\n23\n85\n266\n0\n0\n0\n0\n0\n0\nam\n\n\n\n\n11203 rows Ć 32 columns\n\n\n\nWe can then get the future data by keying in on the data with value column that is missing (np.nan).\n\n\nCode\n# Get the future data (just the observations that haven't happened yet)\nfuture_df = (\n full_augmented_df\n .query('value.isna()')\n)\nfuture_df\n\n\n\n\n\n\n\n\n\nid\ndate\nvalue\ndate_index_num\ndate_year\ndate_year_iso\ndate_yearstart\ndate_yearend\ndate_leapyear\ndate_half\n...\ndate_mday\ndate_qday\ndate_yday\ndate_weekend\ndate_hour\ndate_minute\ndate_second\ndate_msecond\ndate_nsecond\ndate_am_pm\n\n\n\n\n9743\nD10\n2016-05-07\nNaN\n1462579200\n2016\n2016\n0\n0\n1\n1\n...\n7\n37\n128\n0\n0\n0\n0\n0\n0\nam\n\n\n9744\nD10\n2016-05-08\nNaN\n1462665600\n2016\n2016\n0\n0\n1\n1\n...\n8\n38\n129\n1\n0\n0\n0\n0\n0\nam\n\n\n9745\nD10\n2016-05-09\nNaN\n1462752000\n2016\n2016\n0\n0\n1\n1\n...\n9\n39\n130\n0\n0\n0\n0\n0\n0\nam\n\n\n9746\nD10\n2016-05-10\nNaN\n1462838400\n2016\n2016\n0\n0\n1\n1\n...\n10\n40\n131\n0\n0\n0\n0\n0\n0\nam\n\n\n9747\nD10\n2016-05-11\nNaN\n1462924800\n2016\n2016\n0\n0\n1\n1\n...\n11\n41\n132\n0\n0\n0\n0\n0\n0\nam\n\n\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n\n\n11198\nD500\n2013-09-19\nNaN\n1379548800\n2013\n2013\n0\n0\n0\n2\n...\n19\n81\n262\n0\n0\n0\n0\n0\n0\nam\n\n\n11199\nD500\n2013-09-20\nNaN\n1379635200\n2013\n2013\n0\n0\n0\n2\n...\n20\n82\n263\n0\n0\n0\n0\n0\n0\nam\n\n\n11200\nD500\n2013-09-21\nNaN\n1379721600\n2013\n2013\n0\n0\n0\n2\n...\n21\n83\n264\n0\n0\n0\n0\n0\n0\nam\n\n\n11201\nD500\n2013-09-22\nNaN\n1379808000\n2013\n2013\n0\n0\n0\n2\n...\n22\n84\n265\n1\n0\n0\n0\n0\n0\nam\n\n\n11202\nD500\n2013-09-23\nNaN\n1379894400\n2013\n2013\n0\n0\n0\n2\n...\n23\n85\n266\n0\n0\n0\n0\n0\n0\nam\n\n\n\n\n1460 rows Ć 32 columns"
+ "objectID": "guides/07_timeseries_crossvalidation.html#step-4-plot-the-cross-validation-splits",
+ "href": "guides/07_timeseries_crossvalidation.html#step-4-plot-the-cross-validation-splits",
+ "title": "Time Series Cross Validation",
+ "section": "2.4 Step 4: Plot the Cross-Validation Splits",
+ "text": "2.4 Step 4: Plot the Cross-Validation Splits\nYou can visualize how the data is split for training and testing:\n\n# Plot the cross-validation splits\ntscv.plot(\n walmart_sales_df['Weekly_Sales'], \n time_series=walmart_sales_df['Date']\n)\n\n\n \n\n\nThis plot will show each fold, illustrating which weeks are used for training and which weeks are used for forecasting."
},
{
- "objectID": "guides/02_timetk_concepts.html#type-2-pandas-series-operations",
- "href": "guides/02_timetk_concepts.html#type-2-pandas-series-operations",
- "title": "PyTimeTK Basics",
- "section": "2.2 Type 2: Pandas Series Operations",
- "text": "2.2 Type 2: Pandas Series Operations\nThe main difference between a DataFrame operation and a Series operation is that we are operating on an array of values from typically one of the following dtypes:\n\nTimestamps (datetime64)\nNumeric (float64 or int64)\n\nThe first argument of Series operations that operate on Timestamps will always be idx.\nLetās take a look at one shall we? Weāll start with a common action: Making future time series from an existing time series with a regular frequency.\n\nThe Make Future Time Series Function\nSay we have a monthly sequence of timestamps. What if we want to create a forecast where we predict 12 months into the future? Well, we will need to create 12 future timestamps. Hereās how.\nFirst create a pd.date_range() with dates starting at the beginning of each month.\n\n\nCode\n# Make a monthly date range\ndates_dt = pd.date_range(\"2023-01\", \"2024-01\", freq=\"MS\")\ndates_dt\n\n\nDatetimeIndex(['2023-01-01', '2023-02-01', '2023-03-01', '2023-04-01',\n '2023-05-01', '2023-06-01', '2023-07-01', '2023-08-01',\n '2023-09-01', '2023-10-01', '2023-11-01', '2023-12-01',\n '2024-01-01'],\n dtype='datetime64[ns]', freq='MS')\n\n\nNext, use tk.make_future_timeseries() to create the next 12 timestamps in the sequence.\n\nPandas SeriesDateTimeIndex\n\n\n\n\nCode\n# Pandas Series: Future Dates\nfuture_series = pd.Series(dates_dt).make_future_timeseries(12)\nfuture_series\n\n\n0 2024-02-01\n1 2024-03-01\n2 2024-04-01\n3 2024-05-01\n4 2024-06-01\n5 2024-07-01\n6 2024-08-01\n7 2024-09-01\n8 2024-10-01\n9 2024-11-01\n10 2024-12-01\n11 2025-01-01\ndtype: datetime64[ns]\n\n\n\n\n\n\nCode\n# DateTimeIndex: Future Dates\nfuture_dt = tk.make_future_timeseries(\n idx = dates_dt,\n length_out = 12\n)\nfuture_dt\n\n\n0 2024-02-01\n1 2024-03-01\n2 2024-04-01\n3 2024-05-01\n4 2024-06-01\n5 2024-07-01\n6 2024-08-01\n7 2024-09-01\n8 2024-10-01\n9 2024-11-01\n10 2024-12-01\n11 2025-01-01\ndtype: datetime64[ns]\n\n\n\n\n\nWe can combine the actual and future timestamps into one combined timeseries.\n\n\nCode\n# Combining the 2 series and resetting the index\ncombined_timeseries = (\n pd.concat(\n [pd.Series(dates_dt), pd.Series(future_dt)],\n axis=0\n )\n .reset_index(drop = True)\n)\n\ncombined_timeseries\n\n\n0 2023-01-01\n1 2023-02-01\n2 2023-03-01\n3 2023-04-01\n4 2023-05-01\n5 2023-06-01\n6 2023-07-01\n7 2023-08-01\n8 2023-09-01\n9 2023-10-01\n10 2023-11-01\n11 2023-12-01\n12 2024-01-01\n13 2024-02-01\n14 2024-03-01\n15 2024-04-01\n16 2024-05-01\n17 2024-06-01\n18 2024-07-01\n19 2024-08-01\n20 2024-09-01\n21 2024-10-01\n22 2024-11-01\n23 2024-12-01\n24 2025-01-01\ndtype: datetime64[ns]\n\n\nNext, weāll take a look at how to go from an irregular time series to a regular time series.\n\n\nFlooring Dates\nAn example is tk.floor_date, which is used to round down dates. See help(tk.floor_date).\nFlooring dates is often used as part of a strategy to go from an irregular time series to regular by combining with an aggregation. Often summarize_by_time() is used (Iāll share why shortly). But conceptually, date flooring is the secret.\n\nWith FlooringWithout Flooring\n\n\n\n\nCode\n# Monthly flooring rounds dates down to 1st of the month\nm4_daily_df['date'].floor_date(unit = \"M\")\n\n\n0 2014-07-01\n1 2014-07-01\n2 2014-07-01\n3 2014-07-01\n4 2014-07-01\n ... \n9738 2014-07-01\n9739 2014-07-01\n9740 2014-07-01\n9741 2014-07-01\n9742 2014-07-01\nName: date, Length: 9743, dtype: datetime64[ns]\n\n\n\n\n\n\nCode\n# Before Flooring\nm4_daily_df['date']\n\n\n0 2014-07-03\n1 2014-07-03\n2 2014-07-03\n3 2014-07-03\n4 2014-07-03\n ... \n9738 2014-07-03\n9739 2014-07-03\n9740 2014-07-03\n9741 2014-07-03\n9742 2014-07-03\nName: date, Length: 9743, dtype: datetime64[ns]\n\n\n\n\n\nThis ādate flooringā operation can be useful for creating date groupings.\n\n\nCode\n# Adding a date group with floor_date()\ndates_grouped_by_month = (\n m4_daily_df\n .assign(date_group = lambda x: x['date'].floor_date(\"M\"))\n)\n\ndates_grouped_by_month\n\n\n\n\n\n\n\n\n\nid\ndate\nvalue\ndate_group\n\n\n\n\n0\nD10\n2014-07-03\n2076.2\n2014-07-01\n\n\n1\nD10\n2014-07-03\n2073.4\n2014-07-01\n\n\n2\nD10\n2014-07-03\n2048.7\n2014-07-01\n\n\n3\nD10\n2014-07-03\n2048.9\n2014-07-01\n\n\n4\nD10\n2014-07-03\n2006.4\n2014-07-01\n\n\n...\n...\n...\n...\n...\n\n\n9738\nD500\n2014-07-03\n9418.8\n2014-07-01\n\n\n9739\nD500\n2014-07-03\n9365.7\n2014-07-01\n\n\n9740\nD500\n2014-07-03\n9445.9\n2014-07-01\n\n\n9741\nD500\n2014-07-03\n9497.9\n2014-07-01\n\n\n9742\nD500\n2014-07-03\n9545.3\n2014-07-01\n\n\n\n\n9743 rows Ć 4 columns\n\n\n\nWe can then do grouped operations.\n\n\nCode\n# Example of a grouped operation with floored dates\nsummary_df = (\n dates_grouped_by_month\n .drop('date', axis=1) \\\n .groupby(['id', 'date_group'])\n .mean() \\\n .reset_index()\n)\n\nsummary_df\n\n\n\n\n\n\n\n\n\nid\ndate_group\nvalue\n\n\n\n\n0\nD10\n2014-07-01\n2261.606825\n\n\n1\nD160\n2014-07-01\n9243.155254\n\n\n2\nD410\n2014-07-01\n8259.786346\n\n\n3\nD500\n2014-07-01\n8287.728789\n\n\n\n\n\n\n\nOf course for this operation, we can do it faster with summarize_by_time() (and itās much more flexible).\n\n\nCode\n# Summarize by time is less code and more flexible\n(\n m4_daily_df \n .groupby('id')\n .summarize_by_time(\n 'date', 'value', \n freq = \"MS\",\n agg_func = ['mean', 'median', 'min', 'max']\n )\n)\n\n\n\n\n\n\n\n\n\nid\ndate\nvalue_mean\nvalue_median\nvalue_min\nvalue_max\n\n\n\n\n0\nD10\n2014-07-01\n2261.606825\n2302.30\n1781.60\n2649.30\n\n\n1\nD160\n2014-07-01\n9243.155254\n10097.30\n1734.90\n19432.50\n\n\n2\nD410\n2014-07-01\n8259.786346\n8382.81\n6309.38\n9540.62\n\n\n3\nD500\n2014-07-01\n8287.728789\n7662.10\n4172.10\n14954.10\n\n\n\n\n\n\n\nAnd thatās the core idea behind pytimetk, writing less code and getting more.\nNext, letās do one more function. The brother of augment_timeseries_signature()ā¦\n\n\nThe Get Time Series Signature Function\nThis function takes a pandas Series or DateTimeIndex and returns a DataFrame containing the 29 engineered features.\nStart with either a DateTimeIndexā¦\n\n\nCode\ntimestamps_dt = pd.date_range(\"2023\", \"2024\", freq = \"D\")\ntimestamps_dt\n\n\nDatetimeIndex(['2023-01-01', '2023-01-02', '2023-01-03', '2023-01-04',\n '2023-01-05', '2023-01-06', '2023-01-07', '2023-01-08',\n '2023-01-09', '2023-01-10',\n ...\n '2023-12-23', '2023-12-24', '2023-12-25', '2023-12-26',\n '2023-12-27', '2023-12-28', '2023-12-29', '2023-12-30',\n '2023-12-31', '2024-01-01'],\n dtype='datetime64[ns]', length=366, freq='D')\n\n\nā¦ Or a Pandas Series.\n\n\nCode\ntimestamps_series = pd.Series(timestamps_dt)\ntimestamps_series\n\n\n0 2023-01-01\n1 2023-01-02\n2 2023-01-03\n3 2023-01-04\n4 2023-01-05\n ... \n361 2023-12-28\n362 2023-12-29\n363 2023-12-30\n364 2023-12-31\n365 2024-01-01\nLength: 366, dtype: datetime64[ns]\n\n\nAnd you can use the pandas Series function, tk.get_timeseries_signature() to create 29 features from the date sequence.\n\nPandas SeriesDateTimeIndex\n\n\n\n\nCode\n# Pandas series: get_timeseries_signature\ntimestamps_series.get_timeseries_signature()\n\n\n\n\n\n\n\n\n\nidx\nidx_index_num\nidx_year\nidx_year_iso\nidx_yearstart\nidx_yearend\nidx_leapyear\nidx_half\nidx_quarter\nidx_quarteryear\n...\nidx_mday\nidx_qday\nidx_yday\nidx_weekend\nidx_hour\nidx_minute\nidx_second\nidx_msecond\nidx_nsecond\nidx_am_pm\n\n\n\n\n0\n2023-01-01\n1672531200\n2023\n2022\n1\n0\n0\n1\n1\n2023Q1\n...\n1\n1\n1\n1\n0\n0\n0\n0\n0\nam\n\n\n1\n2023-01-02\n1672617600\n2023\n2023\n0\n0\n0\n1\n1\n2023Q1\n...\n2\n2\n2\n0\n0\n0\n0\n0\n0\nam\n\n\n2\n2023-01-03\n1672704000\n2023\n2023\n0\n0\n0\n1\n1\n2023Q1\n...\n3\n3\n3\n0\n0\n0\n0\n0\n0\nam\n\n\n3\n2023-01-04\n1672790400\n2023\n2023\n0\n0\n0\n1\n1\n2023Q1\n...\n4\n4\n4\n0\n0\n0\n0\n0\n0\nam\n\n\n4\n2023-01-05\n1672876800\n2023\n2023\n0\n0\n0\n1\n1\n2023Q1\n...\n5\n5\n5\n0\n0\n0\n0\n0\n0\nam\n\n\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n\n\n361\n2023-12-28\n1703721600\n2023\n2023\n0\n0\n0\n2\n4\n2023Q4\n...\n28\n89\n362\n0\n0\n0\n0\n0\n0\nam\n\n\n362\n2023-12-29\n1703808000\n2023\n2023\n0\n0\n0\n2\n4\n2023Q4\n...\n29\n90\n363\n0\n0\n0\n0\n0\n0\nam\n\n\n363\n2023-12-30\n1703894400\n2023\n2023\n0\n0\n0\n2\n4\n2023Q4\n...\n30\n91\n364\n0\n0\n0\n0\n0\n0\nam\n\n\n364\n2023-12-31\n1703980800\n2023\n2023\n0\n1\n0\n2\n4\n2023Q4\n...\n31\n92\n365\n1\n0\n0\n0\n0\n0\nam\n\n\n365\n2024-01-01\n1704067200\n2024\n2024\n1\n0\n1\n1\n1\n2024Q1\n...\n1\n1\n1\n0\n0\n0\n0\n0\n0\nam\n\n\n\n\n366 rows Ć 30 columns\n\n\n\n\n\n\n\nCode\n# DateTimeIndex: get_timeseries_signature\ntk.get_timeseries_signature(timestamps_dt)\n\n\n\n\n\n\n\n\n\nidx\nidx_index_num\nidx_year\nidx_year_iso\nidx_yearstart\nidx_yearend\nidx_leapyear\nidx_half\nidx_quarter\nidx_quarteryear\n...\nidx_mday\nidx_qday\nidx_yday\nidx_weekend\nidx_hour\nidx_minute\nidx_second\nidx_msecond\nidx_nsecond\nidx_am_pm\n\n\n\n\n0\n2023-01-01\n1672531200\n2023\n2022\n1\n0\n0\n1\n1\n2023Q1\n...\n1\n1\n1\n1\n0\n0\n0\n0\n0\nam\n\n\n1\n2023-01-02\n1672617600\n2023\n2023\n0\n0\n0\n1\n1\n2023Q1\n...\n2\n2\n2\n0\n0\n0\n0\n0\n0\nam\n\n\n2\n2023-01-03\n1672704000\n2023\n2023\n0\n0\n0\n1\n1\n2023Q1\n...\n3\n3\n3\n0\n0\n0\n0\n0\n0\nam\n\n\n3\n2023-01-04\n1672790400\n2023\n2023\n0\n0\n0\n1\n1\n2023Q1\n...\n4\n4\n4\n0\n0\n0\n0\n0\n0\nam\n\n\n4\n2023-01-05\n1672876800\n2023\n2023\n0\n0\n0\n1\n1\n2023Q1\n...\n5\n5\n5\n0\n0\n0\n0\n0\n0\nam\n\n\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n\n\n361\n2023-12-28\n1703721600\n2023\n2023\n0\n0\n0\n2\n4\n2023Q4\n...\n28\n89\n362\n0\n0\n0\n0\n0\n0\nam\n\n\n362\n2023-12-29\n1703808000\n2023\n2023\n0\n0\n0\n2\n4\n2023Q4\n...\n29\n90\n363\n0\n0\n0\n0\n0\n0\nam\n\n\n363\n2023-12-30\n1703894400\n2023\n2023\n0\n0\n0\n2\n4\n2023Q4\n...\n30\n91\n364\n0\n0\n0\n0\n0\n0\nam\n\n\n364\n2023-12-31\n1703980800\n2023\n2023\n0\n1\n0\n2\n4\n2023Q4\n...\n31\n92\n365\n1\n0\n0\n0\n0\n0\nam\n\n\n365\n2024-01-01\n1704067200\n2024\n2024\n1\n0\n1\n1\n1\n2024Q1\n...\n1\n1\n1\n0\n0\n0\n0\n0\n0\nam\n\n\n\n\n366 rows Ć 30 columns"
+ "objectID": "guides/07_timeseries_crossvalidation.html#step-1-setting-up-the-timeseriescvsplitter",
+ "href": "guides/07_timeseries_crossvalidation.html#step-1-setting-up-the-timeseriescvsplitter",
+ "title": "Time Series Cross Validation",
+ "section": "3.1 Step 1: Setting Up the TimeSeriesCVSplitter",
+ "text": "3.1 Step 1: Setting Up the TimeSeriesCVSplitter\nThe TimeSeriesCVSplitter helps us divide our dataset into training and forecast sets in a rolling window fashion. Hereās how we configure it:\n\nfrom pytimetk.crossvalidation import TimeSeriesCVSplitter\nfrom sklearn.ensemble import RandomForestRegressor\nfrom sklearn.model_selection import cross_val_score\n\n# Set up TimeSeriesCVSplitter\ncv_splitter = TimeSeriesCVSplitter(\n time_series=walmart_sales_df['Date'],\n frequency=\"weeks\",\n train_size=52*2,\n forecast_horizon=12,\n gap=0,\n stride=4,\n window=\"rolling\",\n mode=\"backward\",\n split_limit = 5\n)\n\n# Visualize the TSCV Strategy\ncv_splitter.splitter.plot(walmart_sales_df['Weekly_Sales'], walmart_sales_df['Date'])\n\n\n \n\n\nThe TimeSeriesCVSplitter creates multiple splits of the time series data, allowing us to validate the model across different periods. By visualizing the cross-validation strategy, we can see how the training and forecast sets are structured."
},
{
- "objectID": "getting-started/02_quick_start.html",
- "href": "getting-started/02_quick_start.html",
- "title": "Quick Start",
- "section": "",
- "text": "This is a simple exercise to showcase the power of our 2 most popular function:\n\nsummarize_by_time()\nplot_timeseries()\n\n\n\nFirst, import pytimetk as tk. This gets you access to the most important functions. Use tk.load_dataset() to load the ābike_sales_sampleā dataset.\n\n\n\n\n\n\nAbout the Bike Sales Sample Dataset\n\n\n\n\n\nThis dataset contains āorderlinesā for orders recieved. The order_date column contains timestamps. We can use this column to peform sales aggregations (e.g.Ā total revenue).\n\n\n\n\n\nCode\nimport pytimetk as tk\nimport pandas as pd\n\ndf = tk.load_dataset('bike_sales_sample')\ndf['order_date'] = pd.to_datetime(df['order_date'])\n\ndf \n\n\n\n\n\n\n\n\n\norder_id\norder_line\norder_date\nquantity\nprice\ntotal_price\nmodel\ncategory_1\ncategory_2\nframe_material\nbikeshop_name\ncity\nstate\n\n\n\n\n0\n1\n1\n2011-01-07\n1\n6070\n6070\nJekyll Carbon 2\nMountain\nOver Mountain\nCarbon\nIthaca Mountain Climbers\nIthaca\nNY\n\n\n1\n1\n2\n2011-01-07\n1\n5970\n5970\nTrigger Carbon 2\nMountain\nOver Mountain\nCarbon\nIthaca Mountain Climbers\nIthaca\nNY\n\n\n2\n2\n1\n2011-01-10\n1\n2770\n2770\nBeast of the East 1\nMountain\nTrail\nAluminum\nKansas City 29ers\nKansas City\nKS\n\n\n3\n2\n2\n2011-01-10\n1\n5970\n5970\nTrigger Carbon 2\nMountain\nOver Mountain\nCarbon\nKansas City 29ers\nKansas City\nKS\n\n\n4\n3\n1\n2011-01-10\n1\n10660\n10660\nSupersix Evo Hi-Mod Team\nRoad\nElite Road\nCarbon\nLouisville Race Equipment\nLouisville\nKY\n\n\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n\n\n2461\n321\n3\n2011-12-22\n1\n1410\n1410\nCAAD8 105\nRoad\nElite Road\nAluminum\nMiami Race Equipment\nMiami\nFL\n\n\n2462\n322\n1\n2011-12-28\n1\n1250\n1250\nSynapse Disc Tiagra\nRoad\nEndurance Road\nAluminum\nPhoenix Bi-peds\nPhoenix\nAZ\n\n\n2463\n322\n2\n2011-12-28\n1\n2660\n2660\nBad Habit 2\nMountain\nTrail\nAluminum\nPhoenix Bi-peds\nPhoenix\nAZ\n\n\n2464\n322\n3\n2011-12-28\n1\n2340\n2340\nF-Si 1\nMountain\nCross Country Race\nAluminum\nPhoenix Bi-peds\nPhoenix\nAZ\n\n\n2465\n322\n4\n2011-12-28\n1\n5860\n5860\nSynapse Hi-Mod Dura Ace\nRoad\nEndurance Road\nCarbon\nPhoenix Bi-peds\nPhoenix\nAZ\n\n\n\n\n2466 rows Ć 13 columns\n\n\n\n\n\n\nYour company might be interested in sales patterns for various categories of bicycles. We can obtain a grouped monthly sales aggregation by category_1 in two lines of code:\n\nFirst use pandasās groupby() method to group the DataFrame on category_1\nNext, use timetkās summarize_by_time() method to apply the sum function my month start (āMSā) and use wide_format = 'False' to return the dataframe in a long format (Note long format is the default).\n\nThe result is the total revenue for Mountain and Road bikes by month.\n\n\nCode\nsummary_category_1_df = df \\\n .groupby(\"category_1\") \\\n .summarize_by_time(\n date_column = 'order_date', \n value_column = 'total_price',\n freq = \"MS\",\n agg_func = 'sum',\n wide_format = False\n )\n\n# First 5 rows shown\nsummary_category_1_df.head()\n\n\n\n\n\n\n\n\n\ncategory_1\norder_date\ntotal_price\n\n\n\n\n0\nMountain\n2011-01-01\n221490\n\n\n1\nMountain\n2011-02-01\n660555\n\n\n2\nMountain\n2011-03-01\n358855\n\n\n3\nMountain\n2011-04-01\n1075975\n\n\n4\nMountain\n2011-05-01\n450440\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nNow available: plot_timeseries().\n\n\n\n\n\nPlot time series is a quick and easy way to visualize time series and make professional time series plots.\n\n\n\nWith the data summarized by time, we can visualize with plot_timeseries(). pytimetk functions are groupby() aware meaning they understand if your data is grouped to do things by group. This is useful in time series where we often deal with 100s of time series groups.\n\n\nCode\nsummary_category_1_df \\\n .groupby('category_1') \\\n .plot_timeseries(\n date_column = 'order_date',\n value_column = 'total_price',\n smooth_frac = 0.8\n )"
+ "objectID": "guides/07_timeseries_crossvalidation.html#step-2-feature-engineering-for-time-series-data",
+ "href": "guides/07_timeseries_crossvalidation.html#step-2-feature-engineering-for-time-series-data",
+ "title": "Time Series Cross Validation",
+ "section": "3.2 Step 2: Feature Engineering for Time Series Data",
+ "text": "3.2 Step 2: Feature Engineering for Time Series Data\nEffective feature engineering can significantly impact the performance of a time series model. Using pytimetk, we extract a variety of features from the Date column.\n\nGenerating Time Series Features\nWe use get_timeseries_signature to generate useful features, such as year, quarter, month, and day-of-week indicators.\n\n# Prepare data for modeling\n\n# Extract time series features from the 'Date' column\nX_time_features = tk.get_timeseries_signature(walmart_sales_df['Date'])\n\n# Select features to dummy encode\nfeatures_to_dummy = ['Date_quarteryear', 'Date_month_lbl', 'Date_wday_lbl', 'Date_am_pm']\n\n# Dummy encode the selected features\nX_time_dummies = pd.get_dummies(X_time_features[features_to_dummy], drop_first=True)\n\n# Dummy encode the 'id' column\nX_id_dummies = pd.get_dummies(walmart_sales_df['id'], prefix='store')\n\n# Combine the time series features, dummy-encoded features, and the 'id' dummies\nX = pd.concat([X_time_features, X_time_dummies, X_id_dummies], axis=1)\n\n# Drop the original categorical columns that were dummy encoded\nX = X.drop(columns=features_to_dummy).drop('Date', axis=1)\n\n# Set the target variable\ny = walmart_sales_df['Weekly_Sales'].values"
},
{
- "objectID": "getting-started/02_quick_start.html#import-libraries-data",
- "href": "getting-started/02_quick_start.html#import-libraries-data",
- "title": "Quick Start",
- "section": "",
- "text": "First, import pytimetk as tk. This gets you access to the most important functions. Use tk.load_dataset() to load the ābike_sales_sampleā dataset.\n\n\n\n\n\n\nAbout the Bike Sales Sample Dataset\n\n\n\n\n\nThis dataset contains āorderlinesā for orders recieved. The order_date column contains timestamps. We can use this column to peform sales aggregations (e.g.Ā total revenue).\n\n\n\n\n\nCode\nimport pytimetk as tk\nimport pandas as pd\n\ndf = tk.load_dataset('bike_sales_sample')\ndf['order_date'] = pd.to_datetime(df['order_date'])\n\ndf \n\n\n\n\n\n\n\n\n\norder_id\norder_line\norder_date\nquantity\nprice\ntotal_price\nmodel\ncategory_1\ncategory_2\nframe_material\nbikeshop_name\ncity\nstate\n\n\n\n\n0\n1\n1\n2011-01-07\n1\n6070\n6070\nJekyll Carbon 2\nMountain\nOver Mountain\nCarbon\nIthaca Mountain Climbers\nIthaca\nNY\n\n\n1\n1\n2\n2011-01-07\n1\n5970\n5970\nTrigger Carbon 2\nMountain\nOver Mountain\nCarbon\nIthaca Mountain Climbers\nIthaca\nNY\n\n\n2\n2\n1\n2011-01-10\n1\n2770\n2770\nBeast of the East 1\nMountain\nTrail\nAluminum\nKansas City 29ers\nKansas City\nKS\n\n\n3\n2\n2\n2011-01-10\n1\n5970\n5970\nTrigger Carbon 2\nMountain\nOver Mountain\nCarbon\nKansas City 29ers\nKansas City\nKS\n\n\n4\n3\n1\n2011-01-10\n1\n10660\n10660\nSupersix Evo Hi-Mod Team\nRoad\nElite Road\nCarbon\nLouisville Race Equipment\nLouisville\nKY\n\n\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n\n\n2461\n321\n3\n2011-12-22\n1\n1410\n1410\nCAAD8 105\nRoad\nElite Road\nAluminum\nMiami Race Equipment\nMiami\nFL\n\n\n2462\n322\n1\n2011-12-28\n1\n1250\n1250\nSynapse Disc Tiagra\nRoad\nEndurance Road\nAluminum\nPhoenix Bi-peds\nPhoenix\nAZ\n\n\n2463\n322\n2\n2011-12-28\n1\n2660\n2660\nBad Habit 2\nMountain\nTrail\nAluminum\nPhoenix Bi-peds\nPhoenix\nAZ\n\n\n2464\n322\n3\n2011-12-28\n1\n2340\n2340\nF-Si 1\nMountain\nCross Country Race\nAluminum\nPhoenix Bi-peds\nPhoenix\nAZ\n\n\n2465\n322\n4\n2011-12-28\n1\n5860\n5860\nSynapse Hi-Mod Dura Ace\nRoad\nEndurance Road\nCarbon\nPhoenix Bi-peds\nPhoenix\nAZ\n\n\n\n\n2466 rows Ć 13 columns"
+ "objectID": "guides/07_timeseries_crossvalidation.html#step-3-model-training-and-evaluation-with-random-forest",
+ "href": "guides/07_timeseries_crossvalidation.html#step-3-model-training-and-evaluation-with-random-forest",
+ "title": "Time Series Cross Validation",
+ "section": "3.3 Step 3: Model Training and Evaluation with Random Forest",
+ "text": "3.3 Step 3: Model Training and Evaluation with Random Forest\nFor this example, we use RandomForestRegressor from scikit-learn to model the time series data. A random forest is a robust, ensemble-based model that can handle a wide range of regression tasks.\n\n# Initialize the RandomForestRegressor model\nmodel = RandomForestRegressor(\n n_estimators=100, # Number of trees in the forest\n max_depth=None, # Maximum depth of the trees (None means nodes are expanded until all leaves are pure)\n random_state=42 # Set a random state for reproducibility\n)\n\n# Evaluate the model using cross-validation scores\nscores = cross_val_score(model, X, y, cv=cv_splitter, scoring='neg_mean_squared_error')\n\n# Print cross-validation scores\nprint(\"Cross-Validation Scores (Negative MSE):\", scores)\n\nCross-Validation Scores (Negative MSE): [-23761708.80112538 -23107644.58461143 -21728878.18790144\n -25113860.93913386 -86192034.48953015]"
},
{
- "objectID": "getting-started/02_quick_start.html#using-summarize_by_time-for-a-sales-analysis",
- "href": "getting-started/02_quick_start.html#using-summarize_by_time-for-a-sales-analysis",
- "title": "Quick Start",
- "section": "",
- "text": "Your company might be interested in sales patterns for various categories of bicycles. We can obtain a grouped monthly sales aggregation by category_1 in two lines of code:\n\nFirst use pandasās groupby() method to group the DataFrame on category_1\nNext, use timetkās summarize_by_time() method to apply the sum function my month start (āMSā) and use wide_format = 'False' to return the dataframe in a long format (Note long format is the default).\n\nThe result is the total revenue for Mountain and Road bikes by month.\n\n\nCode\nsummary_category_1_df = df \\\n .groupby(\"category_1\") \\\n .summarize_by_time(\n date_column = 'order_date', \n value_column = 'total_price',\n freq = \"MS\",\n agg_func = 'sum',\n wide_format = False\n )\n\n# First 5 rows shown\nsummary_category_1_df.head()\n\n\n\n\n\n\n\n\n\ncategory_1\norder_date\ntotal_price\n\n\n\n\n0\nMountain\n2011-01-01\n221490\n\n\n1\nMountain\n2011-02-01\n660555\n\n\n2\nMountain\n2011-03-01\n358855\n\n\n3\nMountain\n2011-04-01\n1075975\n\n\n4\nMountain\n2011-05-01\n450440"
- },
- {
- "objectID": "getting-started/02_quick_start.html#visualizing-sales-patterns",
- "href": "getting-started/02_quick_start.html#visualizing-sales-patterns",
- "title": "Quick Start",
- "section": "",
- "text": "Now available: plot_timeseries().\n\n\n\n\n\nPlot time series is a quick and easy way to visualize time series and make professional time series plots.\n\n\n\nWith the data summarized by time, we can visualize with plot_timeseries(). pytimetk functions are groupby() aware meaning they understand if your data is grouped to do things by group. This is useful in time series where we often deal with 100s of time series groups.\n\n\nCode\nsummary_category_1_df \\\n .groupby('category_1') \\\n .plot_timeseries(\n date_column = 'order_date',\n value_column = 'total_price',\n smooth_frac = 0.8\n )"
- },
- {
- "objectID": "index.html",
- "href": "index.html",
- "title": "PyTimeTK ",
- "section": "",
- "text": "Time series easier, faster, more fun. Pytimetk.\n\nPyTimetkās Mission: To make time series analysis easier, faster, and more enjoyable in Python.\nPlease ā us on GitHub (it takes 2-seconds and means a lot).\n\n1 Introducing pytimetk: Simplifying Time Series Analysis for Everyone\nTime series analysis is fundamental in many fields, from business forecasting to scientific research. While the Python ecosystem offers tools like pandas, they sometimes can be verbose and not optimized for all operations, especially for complex time-based aggregations and visualizations.\nEnter pytimetk. Crafted with a blend of ease-of-use and computational efficiency, pytimetk significantly simplifies the process of time series manipulation and visualization. By leveraging the polars backend, you can experience speed improvements ranging from 3X to a whopping 3500X. Letās dive into a comparative analysis.\n\n\n\n\n\n\n\n\nFeatures/Properties\npytimetk\npandas (+matplotlib)\n\n\n\n\nSpeed\nš 3X to 500X Faster\nš¢ Standard\n\n\nCode Simplicity\nš Concise, readable syntax\nš Often verbose\n\n\nplot_timeseries()\nšØ 2 lines, no customization\nšØ 16 lines, customization needed\n\n\nsummarize_by_time()\nš 2 lines, 13.4X faster\nš 6 lines, 2 for-loops\n\n\npad_by_time()\nā³ 2 lines, fills gaps in timeseries\nā No equivalent\n\n\nanomalize()\nš 2 lines, detects and corrects anomalies\nā No equivalent\n\n\naugment_timeseries_signature()\nš 1 line, all calendar features\nš 30 lines of dt extractors\n\n\naugment_rolling()\nšļø 10X to 3500X faster\nš¢ Slow Rolling Operations\n\n\n\nAs evident from the table, pytimetk is not just about speed; it also simplifies your codebase. For example, summarize_by_time(), converts a 6-line, double for-loop routine in pandas into a concise 2-line operation. And with the polars engine, get results 13.4X faster than pandas!\nSimilarly, plot_timeseries() dramatically streamlines the plotting process, encapsulating what would typically require 16 lines of matplotlib code into a mere 2-line command in pytimetk, without sacrificing customization or quality. And with plotly and plotnine engines, you can create interactive plots and beautiful static visualizations with just a few lines of code.\nFor calendar features, pytimetk offers augment_timeseries_signature() which cuts down on over 30 lines of pandas dt extractions. For rolling features, pytimetk offers augment_rolling(), which is 10X to 3500X faster than pandas. It also offers pad_by_time() to fill gaps in your time series data, and anomalize() to detect and correct anomalies in your time series data.\nJoin the revolution in time series analysis. Reduce your code complexity, increase your productivity, and harness the speed that pytimetk brings to your workflows.\nExplore more at our pytimetk homepage.\n\n\n2 š Installation\nInstall the Latest Stable Version:\npip install pytimetk\nAlternatively, install the Development GitHub Version:\npip install git+https://github.com/business-science/pytimetk.git\n\n\n3 š Quick Start: A Monthly Sales Analysis\nThis is a simple exercise to showcase the power of summarize_by_time():\n\nImport Libraries & Data\nFirst, import pytimetk as tk. This gets you access to the most important functions. Use tk.load_dataset() to load the ābike_sales_sampleā dataset.\n\n\n\n\n\n\nAbout the Bike Sales Sample Dataset\n\n\n\n\n\nThis dataset contains āorderlinesā for orders recieved. The order_date column contains timestamps. We can use this column to peform sales aggregations (e.g.Ā total revenue).\n\n\n\n\nimport pytimetk as tk\nimport pandas as pd\n\ndf = tk.load_dataset('bike_sales_sample')\ndf['order_date'] = pd.to_datetime(df['order_date'])\n\ndf \n\n\n\n\n\n\n\n\norder_id\norder_line\norder_date\nquantity\nprice\ntotal_price\nmodel\ncategory_1\ncategory_2\nframe_material\nbikeshop_name\ncity\nstate\n\n\n\n\n0\n1\n1\n2011-01-07\n1\n6070\n6070\nJekyll Carbon 2\nMountain\nOver Mountain\nCarbon\nIthaca Mountain Climbers\nIthaca\nNY\n\n\n1\n1\n2\n2011-01-07\n1\n5970\n5970\nTrigger Carbon 2\nMountain\nOver Mountain\nCarbon\nIthaca Mountain Climbers\nIthaca\nNY\n\n\n2\n2\n1\n2011-01-10\n1\n2770\n2770\nBeast of the East 1\nMountain\nTrail\nAluminum\nKansas City 29ers\nKansas City\nKS\n\n\n3\n2\n2\n2011-01-10\n1\n5970\n5970\nTrigger Carbon 2\nMountain\nOver Mountain\nCarbon\nKansas City 29ers\nKansas City\nKS\n\n\n4\n3\n1\n2011-01-10\n1\n10660\n10660\nSupersix Evo Hi-Mod Team\nRoad\nElite Road\nCarbon\nLouisville Race Equipment\nLouisville\nKY\n\n\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n\n\n2461\n321\n3\n2011-12-22\n1\n1410\n1410\nCAAD8 105\nRoad\nElite Road\nAluminum\nMiami Race Equipment\nMiami\nFL\n\n\n2462\n322\n1\n2011-12-28\n1\n1250\n1250\nSynapse Disc Tiagra\nRoad\nEndurance Road\nAluminum\nPhoenix Bi-peds\nPhoenix\nAZ\n\n\n2463\n322\n2\n2011-12-28\n1\n2660\n2660\nBad Habit 2\nMountain\nTrail\nAluminum\nPhoenix Bi-peds\nPhoenix\nAZ\n\n\n2464\n322\n3\n2011-12-28\n1\n2340\n2340\nF-Si 1\nMountain\nCross Country Race\nAluminum\nPhoenix Bi-peds\nPhoenix\nAZ\n\n\n2465\n322\n4\n2011-12-28\n1\n5860\n5860\nSynapse Hi-Mod Dura Ace\nRoad\nEndurance Road\nCarbon\nPhoenix Bi-peds\nPhoenix\nAZ\n\n\n\n\n2466 rows Ć 13 columns\n\n\n\n\n\nUsing summarize_by_time() for a Sales Analysis\nYour company might be interested in sales patterns for various categories of bicycles. We can obtain a grouped monthly sales aggregation by category_1 in two lines of code:\n\nFirst use pandasās groupby() method to group the DataFrame on category_1\nNext, use timetkās summarize_by_time() method to apply the sum function my month start (āMSā) and use wide_format = 'False' to return the dataframe in a long format (Note long format is the default). The default engine is \"pandas\". Selecting engine = \"polars\" allows us to improve the speed of the function.\n\nThe result is the total revenue for Mountain and Road bikes by month.\n\nsummary_category_1_df = df \\\n .groupby(\"category_1\") \\\n .summarize_by_time(\n date_column = 'order_date', \n value_column = 'total_price',\n freq = \"MS\",\n agg_func = 'sum',\n wide_format = False,\n engine = \"polars\"\n )\n\n# Quickly examine each column\nsummary_category_1_df.glimpse()\n\n<class 'pandas.core.frame.DataFrame'>: 24 rows of 3 columns\ncategory_1: object ['Mountain', 'Mountain', 'Mountain', ...\norder_date: datetime64[ns] [Timestamp('2011-01-01 00:00:00'), T ...\ntotal_price_sum: int64 [221490, 660555, 358855, 1075975, 45 ...\n\n\n\n\nVisualizing Sales Patterns\n\n\n\n\n\n\nNow available: plot_timeseries().\n\n\n\n\n\nPlot time series is a quick and easy way to visualize time series and make professional time series plots.\n\n\n\nWith the data summarized by time, we can visualize with plot_timeseries(). pytimetk functions are groupby() aware meaning they understand if your data is grouped to do things by group. This is useful in time series where we often deal with 100s of time series groups.\nThe default engine in āplotnineā for static plotting. Setting the engine = \"plotly\" returns an interactive plot.\n\nsummary_category_1_df \\\n .groupby('category_1') \\\n .plot_timeseries(\n date_column = 'order_date',\n value_column = 'total_price_sum',\n smooth_frac = 0.8,\n engine = \"plotly\"\n )\n\n\n \n\n\n\n\n\n4 š Documentation\nNext step? Learn more with the pytimetk documentation\n\nš Overview\nš Getting Started\nšŗļø Beginner Guides\nšApplied Data Science Tutorials with PyTimeTK\nšļøSpeed Comparisons\nš API Reference\n\n\n\n5 š» Contributing\nInterested in helping us make this the best Python package for time series analysis? Weād love your help.\nFollow these instructions to Contribute.\n\n\n6 š More Coming Soonā¦\nWe are in the early stages of development. But itās obvious the potential for pytimetk now in Python. š\n\nPlease ā us on GitHub (it takes 2-seconds and means a lot).\nTo make requests, please see our Project Roadmap GH Issue #2. You can make requests there.\nWant to contribute? See our contributing guide here."
- },
- {
- "objectID": "contributing.html",
- "href": "contributing.html",
- "title": "Contributing (Developer Setup)",
- "section": "",
- "text": "Interested in contributing?\n\n\n\n\n\nMake sure to Fork the GitHub Repo. Clone your fork. Then use poetry to install the pytimetk package.\n\n\n\n\n1 GitHub\nTo contribute, youāll need to have a GitHub account. Then:\n\n1. Fork our pytimetk repository\nHead to our GitHub Repo and select āforkā. This makes a copied version of pytimetk for your personal use.\n\n\n2. Clone your forked version\nCloning will put your own personal version of pytimetk on your local machine. Make sure to replace [your_user_name] with your user name.\ngit clone https://github.com/[your_user_name]/pytimetk\n\n\n\n2 Poetry Environment Setup\nTo install pytimetk using Poetry, follow these steps:\n\n1. Prerequisites\nMake sure you have Python 3.9 or later installed on your system.\n\n\n2. Install Poetry\nTo install Poetry, you can use the official installer provided by Poetry. Do not use pip.\n\n\n3. Install Dependencies\nUse Poetry to install the package and its dependencies:\npoetry install\nor you can create a virtualenv with poetry and install the dependencies\npoetry shell\npoetry install\n\n\n\n3 Submit a Pull Request\n\n1. Make changes on a Branch\nMake changes in your local version on a branch where my-feature-branch is a branch youād like to create that contains modifications.\ngit checkout -b my-feature-branch\n\n\n2. Push to your forked version of pytimetk\ngit push origin my-feature-branch\n\n\n3. Create a Pull Request\n\nGo to your forked repository on GitHub and switch to your branch.\nClick on āNew pull requestā and compare the changes you made with the original repository.\nFill out the pull request template with the necessary information, explaining your changes, the reason for them, and any other relevant information.\n\n\n\n4. Submit the Pull Request\n\nReview your changes and submit the pull request.\n\n\n\n\n4 Next Steps š»\nWe will review your PR. If all goes well, weāll merge! And then youāve just helped the community. š»"
- },
- {
- "objectID": "getting-started/01_installation.html",
- "href": "getting-started/01_installation.html",
- "title": "Install",
- "section": "",
- "text": "1 Quick Install\nLetās get you up and running with pytimetk fast with the latest stable release.\npip install pytimetk\nYou can install from GitHub with this code.\npip install git+https://github.com/business-science/pytimetk.git\n\n\n2 Next steps\nCheck out the Quick Start Guide Next.\n\n\n3 More Coming Soonā¦\nWe are in the early stages of development. But itās obvious the potential for pytimetk now in Python. š\n\nPlease ā us on GitHub (it takes 2-seconds and means a lot).\nTo make requests, please see our Project Roadmap GH Issue #2. You can make requests there.\nWant to contribute? See our contributing guide here."
- },
- {
- "objectID": "guides/03_pandas_frequency.html",
- "href": "guides/03_pandas_frequency.html",
- "title": "Pandas Frequencies",
- "section": "",
- "text": "How this guide benefits you\n\n\n\n\n\nThis guide covers how to use the pandas frequency strings within pytimetk. Once you understand key frequencies, you can apply them to manipulate time series data like a pro.\n\n\n\n\n1 Pandas Frequencies\nPandas offers a variety of frequency strings, also known as offset aliases, to define the frequency of a time series. Here are some common frequency strings used in pandas:\n\nāBā: Business Day\nāDā: Calendar day\nāWā: Weekly\nāMā: Month end\nāBMā: Business month end\nāMSā: Month start\nāBMSā: Business month start\nāQā: Quarter end\nāBQā: Business quarter end\nāQSā: Quarter start\nāBQSā: Business quarter start\nāAā or āYā: Year end\nāBAā or āBYā: Business year end\nāASā or āYSā: Year start\nāBASā or āBYSā: Business year start\nāHā: Hourly\nāTā or āminā: Minutely\nāSā: Secondly\nāLā or āmsā: Milliseconds\nāUā: Microseconds\nāNā: Nanoseconds\n\n\nCustom Frequencies:\n\nYou can also create custom frequencies by combining base frequencies, like:\n\nā2Dā: Every 2 days\nā3Wā: Every 3 weeks\nā4Hā: Every 4 hours\nā1H30Tā: Every 1 hour and 30 minutes\n\n\n\n\nCompound Frequencies:\n\nYou can combine multiple frequencies by adding them together.\n\nā1D1Hā: 1 day and 1 hour\nā1H30Tā: 1 hour and 30 minutes\n\n\n\n\nExample:\n\n\nCode\nimport pandas as pd\n\n# Creating a date range with daily frequency\ndate_range_daily = pd.date_range(start='2023-01-01', end='2023-01-10', freq='D')\n\ndate_range_daily\n\n\nDatetimeIndex(['2023-01-01', '2023-01-02', '2023-01-03', '2023-01-04',\n '2023-01-05', '2023-01-06', '2023-01-07', '2023-01-08',\n '2023-01-09', '2023-01-10'],\n dtype='datetime64[ns]', freq='D')\n\n\n\n\nCode\n# Creating a date range with 2 days frequency\ndate_range_two_days = pd.date_range(start='2023-01-01', end='2023-01-10', freq='2D')\n\ndate_range_two_days\n\n\nDatetimeIndex(['2023-01-01', '2023-01-03', '2023-01-05', '2023-01-07',\n '2023-01-09'],\n dtype='datetime64[ns]', freq='2D')\n\n\nThese frequency strings help in resampling, creating date ranges, and handling time-series data efficiently in pandas.\n\n\n\n2 Timetk Incorporates Pandas Frequencies\nNow that youāve seen pandas frequencies, youāll see them pop up in many of the pytimetk functions.\n\nExample: Padding Dates\nThis example shows how to use Pandas frequencies inside of pytimetk functions.\nWeāll use pad_by_time to show how to use freq to fill in missing dates.\n\n\nCode\n# DataFrame with missing dates\nimport pandas as pd\n\ndata = {\n # '2023-09-05' is missing\n 'datetime': ['2023-09-01', '2023-09-02', '2023-09-03', '2023-09-04', '2023-09-06'], \n 'value': [10, 30, 40, 50, 60]\n}\n\ndf = pd.DataFrame(data)\ndf['datetime'] = pd.to_datetime(df['datetime'])\ndf\n\n\n\n\n\n\n\n\n\ndatetime\nvalue\n\n\n\n\n0\n2023-09-01\n10\n\n\n1\n2023-09-02\n30\n\n\n2\n2023-09-03\n40\n\n\n3\n2023-09-04\n50\n\n\n4\n2023-09-06\n60\n\n\n\n\n\n\n\nWe can resample to fill in the missing day using pad_by_time with freq = 'D'.\n\n\nCode\nimport pytimetk as tk\n\ndf.pad_by_time('datetime', freq = 'D')\n\n\n\n\n\n\n\n\n\ndatetime\nvalue\n\n\n\n\n0\n2023-09-01\n10.0\n\n\n1\n2023-09-02\n30.0\n\n\n2\n2023-09-03\n40.0\n\n\n3\n2023-09-04\n50.0\n\n\n4\n2023-09-05\nNaN\n\n\n5\n2023-09-06\n60.0\n\n\n\n\n\n\n\nWhat about resampling every 12 hours? Just set `freq = ā12Hā.\n\n\nCode\nimport pytimetk as tk\n\ndf.pad_by_time('datetime', freq = '12H')\n\n\n\n\n\n\n\n\n\ndatetime\nvalue\n\n\n\n\n0\n2023-09-01 00:00:00\n10.0\n\n\n1\n2023-09-01 12:00:00\nNaN\n\n\n2\n2023-09-02 00:00:00\n30.0\n\n\n3\n2023-09-02 12:00:00\nNaN\n\n\n4\n2023-09-03 00:00:00\n40.0\n\n\n5\n2023-09-03 12:00:00\nNaN\n\n\n6\n2023-09-04 00:00:00\n50.0\n\n\n7\n2023-09-04 12:00:00\nNaN\n\n\n8\n2023-09-05 00:00:00\nNaN\n\n\n9\n2023-09-05 12:00:00\nNaN\n\n\n10\n2023-09-06 00:00:00\n60.0\n\n\n\n\n\n\n\nYouāll see these pandas frequencies come up as the parameter freq in many pytimetk functions.\n\n\n\n3 Next Steps\nCheck out the Data Wrangling Guide next.\n\n\n4 More Coming Soonā¦\nWe are in the early stages of development. But itās obvious the potential for pytimetk now in Python. š\n\nPlease ā us on GitHub (it takes 2-seconds and means a lot).\nTo make requests, please see our Project Roadmap GH Issue #2. You can make requests there.\nWant to contribute? See our contributing guide here."
+ "objectID": "guides/07_timeseries_crossvalidation.html#step-4-visualizing-the-forecast",
+ "href": "guides/07_timeseries_crossvalidation.html#step-4-visualizing-the-forecast",
+ "title": "Time Series Cross Validation",
+ "section": "3.4 Step 4: Visualizing the Forecast",
+ "text": "3.4 Step 4: Visualizing the Forecast\nVisualization is crucial to understand how well the model predicts future values. We collect the actual and predicted values for each fold and combine them for easy plotting.\n\n# Lists to store the combined data\ncombined_data = []\n\n# Iterate through each fold and collect the data\nfor i, (train_index, test_index) in enumerate(cv_splitter.split(X, y), start=1):\n # Get the training and forecast data from the original DataFrame\n train_df = walmart_sales_df.iloc[train_index].copy()\n test_df = walmart_sales_df.iloc[test_index].copy()\n \n # Fit the model on the training data\n model.fit(X.iloc[train_index], y[train_index])\n \n # Predict on the test set\n y_pred = model.predict(X.iloc[test_index])\n \n # Add the actual and predicted values\n train_df['Actual'] = y[train_index]\n train_df['Predicted'] = None # No predictions for training data\n train_df['Fold'] = i # Indicate the current fold\n \n test_df['Actual'] = y[test_index]\n test_df['Predicted'] = y_pred # Predictions for the test data\n test_df['Fold'] = i # Indicate the current fold\n \n # Append both the training and forecast DataFrames to the combined data list\n combined_data.extend([train_df, test_df])\n\n# Combine all the data into a single DataFrame\nfull_forecast_df = pd.concat(combined_data, ignore_index=True)\n\nfull_forecast_df = full_forecast_df[['id', 'Date', 'Actual', 'Predicted', 'Fold']]\n\nfull_forecast_df.glimpse()\n\n<class 'pandas.core.frame.DataFrame'>: 4060 rows of 5 columns\nid: object ['1_1', '1_1', '1_1', '1_1', '1_1', '1_1', ...\nDate: datetime64[ns] [Timestamp('2010-08-06 00:00:00'), Timesta ...\nActual: float64 [17508.41, 15536.4, 15740.13, 15793.87, 16 ...\nPredicted: float64 [nan, nan, nan, nan, nan, nan, nan, nan, n ...\nFold: int64 [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...\n\n\n\nPreparing Data for Visualization\nTo make the data easier to plot, we use pd.melt() to transform the Actual and Predicted columns into a long format.\n\n# Melt the Actual and Predicted columns\nmelted_df = pd.melt(\n full_forecast_df,\n id_vars=['id', 'Date', 'Fold'], # Columns to keep\n value_vars=['Actual', 'Predicted'], # Columns to melt\n var_name='Type', # Name for the new column indicating 'Actual' or 'Predicted'\n value_name='Value' # Name for the new column with the values\n)\n\nmelted_df[\"unique_id\"] = \"ID_\" + melted_df['id'] + \"-Fold_\" + melted_df[\"Fold\"].astype(str)\n\nmelted_df.glimpse()\n\n<class 'pandas.core.frame.DataFrame'>: 8120 rows of 6 columns\nid: object ['1_1', '1_1', '1_1', '1_1', '1_1', '1_1', ...\nDate: datetime64[ns] [Timestamp('2010-08-06 00:00:00'), Timesta ...\nFold: int64 [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...\nType: object ['Actual', 'Actual', 'Actual', 'Actual', ' ...\nValue: float64 [17508.41, 15536.4, 15740.13, 15793.87, 16 ...\nunique_id: object ['ID_1_1-Fold_1', 'ID_1_1-Fold_1', 'ID_1_1 ...\n\n\n\n\nPlotting the Forecasts\nFinally, we use plot_timeseries() to visualize the forecasts, comparing the actual and predicted values for each fold.\n\nmelted_df \\\n .groupby('unique_id') \\\n .plot_timeseries(\n \"Date\", \"Value\",\n color_column = \"Type\",\n smooth=False, \n plotly_dropdown=True\n )"
},
{
"objectID": "guides/06_anomalize.html",
@@ -1546,6 +1511,104 @@
"section": "1.6 Changing Parameters",
"text": "1.6 Changing Parameters\nSome important parameters to hightlight in the anomalize() function include iqr_alpha.\n\n\n\n\n\n\nImportant\n\n\n\n\n\niqr_alpha controls the threshold for detecting outliers. It is the significance level used in the interquartile range (IQR) method for outlier detection. The default value is 0.05, which corresponds to a 5% significance level. A lower significance level will result in a higher threshold, which means fewer outliers will be detected. A higher significance level will result in a lower threshold, which means more outliers will be detected.\n\n\n\nLets visualize the effect of changing the iqr_alpha parameter;\n\nChanging iqr_alpha\nFirst, lets get a dataframe with multiple values for iqr_alpha;\n\n# Anomalized data with multiple iqr_alpha values\n\n# - Alpha values\niqr_alpha_values = [0.05, 0.10, 0.15, 0.20]\n\n# - Empty dataframes list\ndfs = []\n\nfor alpha in iqr_alpha_values:\n\n # - Run anomalize function\n anomalize_df = tk.anomalize(\n data = df,\n date_column = 'date',\n value_column = 'value',\n period = 7,\n iqr_alpha = alpha\n )\n\n # - Add the iqr_alpha column\n anomalize_df['iqr_alpha'] = f'iqr_alpha value of {alpha}'\n\n # - Append to the list\n dfs.append(anomalize_df)\n\n# - Concatenate all dataframes\nfinal_df = pd.concat(dfs)\n\nNow we can visualize the anomalies:\n\n\nVisualizing Grouped Anomalies (Facets)\n\n# Visualize\n(\n final_df\n .groupby('iqr_alpha')\n .plot_anomalies(\n date_column = 'date',\n engine = 'plotly',\n facet_ncol = 2\n )\n)\n\n\n \n\n\n\n\nVisualizing Grouped Anomalies (Plotly Dropdown)\n\n# Visualize\n(\n final_df\n .groupby('iqr_alpha')\n .plot_anomalies(\n date_column = 'date',\n engine = 'plotly',\n plotly_dropdown = True,\n plotly_dropdown_x = 1,\n plotly_dropdown_y = 0.60\n )\n)"
},
+ {
+ "objectID": "guides/03_pandas_frequency.html",
+ "href": "guides/03_pandas_frequency.html",
+ "title": "Pandas Frequencies",
+ "section": "",
+ "text": "How this guide benefits you\n\n\n\n\n\nThis guide covers how to use the pandas frequency strings within pytimetk. Once you understand key frequencies, you can apply them to manipulate time series data like a pro.\n\n\n\n\n1 Pandas Frequencies\nPandas offers a variety of frequency strings, also known as offset aliases, to define the frequency of a time series. Here are some common frequency strings used in pandas:\n\nāBā: Business Day\nāDā: Calendar day\nāWā: Weekly\nāMā: Month end\nāBMā: Business month end\nāMSā: Month start\nāBMSā: Business month start\nāQā: Quarter end\nāBQā: Business quarter end\nāQSā: Quarter start\nāBQSā: Business quarter start\nāAā or āYā: Year end\nāBAā or āBYā: Business year end\nāASā or āYSā: Year start\nāBASā or āBYSā: Business year start\nāHā: Hourly\nāTā or āminā: Minutely\nāSā: Secondly\nāLā or āmsā: Milliseconds\nāUā: Microseconds\nāNā: Nanoseconds\n\n\nCustom Frequencies:\n\nYou can also create custom frequencies by combining base frequencies, like:\n\nā2Dā: Every 2 days\nā3Wā: Every 3 weeks\nā4Hā: Every 4 hours\nā1H30Tā: Every 1 hour and 30 minutes\n\n\n\n\nCompound Frequencies:\n\nYou can combine multiple frequencies by adding them together.\n\nā1D1Hā: 1 day and 1 hour\nā1H30Tā: 1 hour and 30 minutes\n\n\n\n\nExample:\n\n\nCode\nimport pandas as pd\n\n# Creating a date range with daily frequency\ndate_range_daily = pd.date_range(start='2023-01-01', end='2023-01-10', freq='D')\n\ndate_range_daily\n\n\nDatetimeIndex(['2023-01-01', '2023-01-02', '2023-01-03', '2023-01-04',\n '2023-01-05', '2023-01-06', '2023-01-07', '2023-01-08',\n '2023-01-09', '2023-01-10'],\n dtype='datetime64[ns]', freq='D')\n\n\n\n\nCode\n# Creating a date range with 2 days frequency\ndate_range_two_days = pd.date_range(start='2023-01-01', end='2023-01-10', freq='2D')\n\ndate_range_two_days\n\n\nDatetimeIndex(['2023-01-01', '2023-01-03', '2023-01-05', '2023-01-07',\n '2023-01-09'],\n dtype='datetime64[ns]', freq='2D')\n\n\nThese frequency strings help in resampling, creating date ranges, and handling time-series data efficiently in pandas.\n\n\n\n2 Timetk Incorporates Pandas Frequencies\nNow that youāve seen pandas frequencies, youāll see them pop up in many of the pytimetk functions.\n\nExample: Padding Dates\nThis example shows how to use Pandas frequencies inside of pytimetk functions.\nWeāll use pad_by_time to show how to use freq to fill in missing dates.\n\n\nCode\n# DataFrame with missing dates\nimport pandas as pd\n\ndata = {\n # '2023-09-05' is missing\n 'datetime': ['2023-09-01', '2023-09-02', '2023-09-03', '2023-09-04', '2023-09-06'], \n 'value': [10, 30, 40, 50, 60]\n}\n\ndf = pd.DataFrame(data)\ndf['datetime'] = pd.to_datetime(df['datetime'])\ndf\n\n\n\n\n\n\n\n\n\ndatetime\nvalue\n\n\n\n\n0\n2023-09-01\n10\n\n\n1\n2023-09-02\n30\n\n\n2\n2023-09-03\n40\n\n\n3\n2023-09-04\n50\n\n\n4\n2023-09-06\n60\n\n\n\n\n\n\n\nWe can resample to fill in the missing day using pad_by_time with freq = 'D'.\n\n\nCode\nimport pytimetk as tk\n\ndf.pad_by_time('datetime', freq = 'D')\n\n\n\n\n\n\n\n\n\ndatetime\nvalue\n\n\n\n\n0\n2023-09-01\n10.0\n\n\n1\n2023-09-02\n30.0\n\n\n2\n2023-09-03\n40.0\n\n\n3\n2023-09-04\n50.0\n\n\n4\n2023-09-05\nNaN\n\n\n5\n2023-09-06\n60.0\n\n\n\n\n\n\n\nWhat about resampling every 12 hours? Just set `freq = ā12Hā.\n\n\nCode\nimport pytimetk as tk\n\ndf.pad_by_time('datetime', freq = '12H')\n\n\n\n\n\n\n\n\n\ndatetime\nvalue\n\n\n\n\n0\n2023-09-01 00:00:00\n10.0\n\n\n1\n2023-09-01 12:00:00\nNaN\n\n\n2\n2023-09-02 00:00:00\n30.0\n\n\n3\n2023-09-02 12:00:00\nNaN\n\n\n4\n2023-09-03 00:00:00\n40.0\n\n\n5\n2023-09-03 12:00:00\nNaN\n\n\n6\n2023-09-04 00:00:00\n50.0\n\n\n7\n2023-09-04 12:00:00\nNaN\n\n\n8\n2023-09-05 00:00:00\nNaN\n\n\n9\n2023-09-05 12:00:00\nNaN\n\n\n10\n2023-09-06 00:00:00\n60.0\n\n\n\n\n\n\n\nYouāll see these pandas frequencies come up as the parameter freq in many pytimetk functions.\n\n\n\n3 Next Steps\nCheck out the Data Wrangling Guide next.\n\n\n4 More Coming Soonā¦\nWe are in the early stages of development. But itās obvious the potential for pytimetk now in Python. š\n\nPlease ā us on GitHub (it takes 2-seconds and means a lot).\nTo make requests, please see our Project Roadmap GH Issue #2. You can make requests there.\nWant to contribute? See our contributing guide here."
+ },
+ {
+ "objectID": "getting-started/01_installation.html",
+ "href": "getting-started/01_installation.html",
+ "title": "Install",
+ "section": "",
+ "text": "1 Quick Install\nLetās get you up and running with pytimetk fast with the latest stable release.\npip install pytimetk\nYou can install from GitHub with this code.\npip install git+https://github.com/business-science/pytimetk.git\n\n\n2 Next steps\nCheck out the Quick Start Guide Next.\n\n\n3 More Coming Soonā¦\nWe are in the early stages of development. But itās obvious the potential for pytimetk now in Python. š\n\nPlease ā us on GitHub (it takes 2-seconds and means a lot).\nTo make requests, please see our Project Roadmap GH Issue #2. You can make requests there.\nWant to contribute? See our contributing guide here."
+ },
+ {
+ "objectID": "contributing.html",
+ "href": "contributing.html",
+ "title": "Contributing (Developer Setup)",
+ "section": "",
+ "text": "Interested in contributing?\n\n\n\n\n\nMake sure to Fork the GitHub Repo. Clone your fork. Then use poetry to install the pytimetk package.\n\n\n\n\n1 GitHub\nTo contribute, youāll need to have a GitHub account. Then:\n\n1. Fork our pytimetk repository\nHead to our GitHub Repo and select āforkā. This makes a copied version of pytimetk for your personal use.\n\n\n2. Clone your forked version\nCloning will put your own personal version of pytimetk on your local machine. Make sure to replace [your_user_name] with your user name.\ngit clone https://github.com/[your_user_name]/pytimetk\n\n\n\n2 Poetry Environment Setup\nTo install pytimetk using Poetry, follow these steps:\n\n1. Prerequisites\nMake sure you have Python 3.9 or later installed on your system.\n\n\n2. Install Poetry\nTo install Poetry, you can use the official installer provided by Poetry. Do not use pip.\n\n\n3. Install Dependencies\nUse Poetry to install the package and its dependencies:\npoetry install\nor you can create a virtualenv with poetry and install the dependencies\npoetry shell\npoetry install\n\n\n\n3 Submit a Pull Request\n\n1. Make changes on a Branch\nMake changes in your local version on a branch where my-feature-branch is a branch youād like to create that contains modifications.\ngit checkout -b my-feature-branch\n\n\n2. Push to your forked version of pytimetk\ngit push origin my-feature-branch\n\n\n3. Create a Pull Request\n\nGo to your forked repository on GitHub and switch to your branch.\nClick on āNew pull requestā and compare the changes you made with the original repository.\nFill out the pull request template with the necessary information, explaining your changes, the reason for them, and any other relevant information.\n\n\n\n4. Submit the Pull Request\n\nReview your changes and submit the pull request.\n\n\n\n\n4 Next Steps š»\nWe will review your PR. If all goes well, weāll merge! And then youāve just helped the community. š»"
+ },
+ {
+ "objectID": "index.html",
+ "href": "index.html",
+ "title": "PyTimeTK ",
+ "section": "",
+ "text": "Time series easier, faster, more fun. Pytimetk.\n\nPyTimetkās Mission: To make time series analysis easier, faster, and more enjoyable in Python.\nPlease ā us on GitHub (it takes 2-seconds and means a lot).\n\n1 Introducing pytimetk: Simplifying Time Series Analysis for Everyone\nTime series analysis is fundamental in many fields, from business forecasting to scientific research. While the Python ecosystem offers tools like pandas, they sometimes can be verbose and not optimized for all operations, especially for complex time-based aggregations and visualizations.\nEnter pytimetk. Crafted with a blend of ease-of-use and computational efficiency, pytimetk significantly simplifies the process of time series manipulation and visualization. By leveraging the polars backend, you can experience speed improvements ranging from 3X to a whopping 3500X. Letās dive into a comparative analysis.\n\n\n\n\n\n\n\n\nFeatures/Properties\npytimetk\npandas (+matplotlib)\n\n\n\n\nSpeed\nš 3X to 500X Faster\nš¢ Standard\n\n\nCode Simplicity\nš Concise, readable syntax\nš Often verbose\n\n\nplot_timeseries()\nšØ 2 lines, no customization\nšØ 16 lines, customization needed\n\n\nsummarize_by_time()\nš 2 lines, 13.4X faster\nš 6 lines, 2 for-loops\n\n\npad_by_time()\nā³ 2 lines, fills gaps in timeseries\nā No equivalent\n\n\nanomalize()\nš 2 lines, detects and corrects anomalies\nā No equivalent\n\n\naugment_timeseries_signature()\nš 1 line, all calendar features\nš 30 lines of dt extractors\n\n\naugment_rolling()\nšļø 10X to 3500X faster\nš¢ Slow Rolling Operations\n\n\n\nAs evident from the table, pytimetk is not just about speed; it also simplifies your codebase. For example, summarize_by_time(), converts a 6-line, double for-loop routine in pandas into a concise 2-line operation. And with the polars engine, get results 13.4X faster than pandas!\nSimilarly, plot_timeseries() dramatically streamlines the plotting process, encapsulating what would typically require 16 lines of matplotlib code into a mere 2-line command in pytimetk, without sacrificing customization or quality. And with plotly and plotnine engines, you can create interactive plots and beautiful static visualizations with just a few lines of code.\nFor calendar features, pytimetk offers augment_timeseries_signature() which cuts down on over 30 lines of pandas dt extractions. For rolling features, pytimetk offers augment_rolling(), which is 10X to 3500X faster than pandas. It also offers pad_by_time() to fill gaps in your time series data, and anomalize() to detect and correct anomalies in your time series data.\nJoin the revolution in time series analysis. Reduce your code complexity, increase your productivity, and harness the speed that pytimetk brings to your workflows.\nExplore more at our pytimetk homepage.\n\n\n2 š Installation\nInstall the Latest Stable Version:\npip install pytimetk\nAlternatively, install the Development GitHub Version:\npip install git+https://github.com/business-science/pytimetk.git\n\n\n3 š Quick Start: A Monthly Sales Analysis\nThis is a simple exercise to showcase the power of summarize_by_time():\n\nImport Libraries & Data\nFirst, import pytimetk as tk. This gets you access to the most important functions. Use tk.load_dataset() to load the ābike_sales_sampleā dataset.\n\n\n\n\n\n\nAbout the Bike Sales Sample Dataset\n\n\n\n\n\nThis dataset contains āorderlinesā for orders recieved. The order_date column contains timestamps. We can use this column to peform sales aggregations (e.g.Ā total revenue).\n\n\n\n\nimport pytimetk as tk\nimport pandas as pd\n\ndf = tk.load_dataset('bike_sales_sample')\ndf['order_date'] = pd.to_datetime(df['order_date'])\n\ndf \n\n\n\n\n\n\n\n\norder_id\norder_line\norder_date\nquantity\nprice\ntotal_price\nmodel\ncategory_1\ncategory_2\nframe_material\nbikeshop_name\ncity\nstate\n\n\n\n\n0\n1\n1\n2011-01-07\n1\n6070\n6070\nJekyll Carbon 2\nMountain\nOver Mountain\nCarbon\nIthaca Mountain Climbers\nIthaca\nNY\n\n\n1\n1\n2\n2011-01-07\n1\n5970\n5970\nTrigger Carbon 2\nMountain\nOver Mountain\nCarbon\nIthaca Mountain Climbers\nIthaca\nNY\n\n\n2\n2\n1\n2011-01-10\n1\n2770\n2770\nBeast of the East 1\nMountain\nTrail\nAluminum\nKansas City 29ers\nKansas City\nKS\n\n\n3\n2\n2\n2011-01-10\n1\n5970\n5970\nTrigger Carbon 2\nMountain\nOver Mountain\nCarbon\nKansas City 29ers\nKansas City\nKS\n\n\n4\n3\n1\n2011-01-10\n1\n10660\n10660\nSupersix Evo Hi-Mod Team\nRoad\nElite Road\nCarbon\nLouisville Race Equipment\nLouisville\nKY\n\n\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n\n\n2461\n321\n3\n2011-12-22\n1\n1410\n1410\nCAAD8 105\nRoad\nElite Road\nAluminum\nMiami Race Equipment\nMiami\nFL\n\n\n2462\n322\n1\n2011-12-28\n1\n1250\n1250\nSynapse Disc Tiagra\nRoad\nEndurance Road\nAluminum\nPhoenix Bi-peds\nPhoenix\nAZ\n\n\n2463\n322\n2\n2011-12-28\n1\n2660\n2660\nBad Habit 2\nMountain\nTrail\nAluminum\nPhoenix Bi-peds\nPhoenix\nAZ\n\n\n2464\n322\n3\n2011-12-28\n1\n2340\n2340\nF-Si 1\nMountain\nCross Country Race\nAluminum\nPhoenix Bi-peds\nPhoenix\nAZ\n\n\n2465\n322\n4\n2011-12-28\n1\n5860\n5860\nSynapse Hi-Mod Dura Ace\nRoad\nEndurance Road\nCarbon\nPhoenix Bi-peds\nPhoenix\nAZ\n\n\n\n\n2466 rows Ć 13 columns\n\n\n\n\n\nUsing summarize_by_time() for a Sales Analysis\nYour company might be interested in sales patterns for various categories of bicycles. We can obtain a grouped monthly sales aggregation by category_1 in two lines of code:\n\nFirst use pandasās groupby() method to group the DataFrame on category_1\nNext, use timetkās summarize_by_time() method to apply the sum function my month start (āMSā) and use wide_format = 'False' to return the dataframe in a long format (Note long format is the default). The default engine is \"pandas\". Selecting engine = \"polars\" allows us to improve the speed of the function.\n\nThe result is the total revenue for Mountain and Road bikes by month.\n\nsummary_category_1_df = df \\\n .groupby(\"category_1\") \\\n .summarize_by_time(\n date_column = 'order_date', \n value_column = 'total_price',\n freq = \"MS\",\n agg_func = 'sum',\n wide_format = False,\n engine = \"polars\"\n )\n\n# Quickly examine each column\nsummary_category_1_df.glimpse()\n\n<class 'pandas.core.frame.DataFrame'>: 24 rows of 3 columns\ncategory_1: object ['Mountain', 'Mountain', 'Mountain', ...\norder_date: datetime64[ns] [Timestamp('2011-01-01 00:00:00'), T ...\ntotal_price_sum: int64 [221490, 660555, 358855, 1075975, 45 ...\n\n\n\n\nVisualizing Sales Patterns\n\n\n\n\n\n\nNow available: plot_timeseries().\n\n\n\n\n\nPlot time series is a quick and easy way to visualize time series and make professional time series plots.\n\n\n\nWith the data summarized by time, we can visualize with plot_timeseries(). pytimetk functions are groupby() aware meaning they understand if your data is grouped to do things by group. This is useful in time series where we often deal with 100s of time series groups.\nThe default engine in āplotnineā for static plotting. Setting the engine = \"plotly\" returns an interactive plot.\n\nsummary_category_1_df \\\n .groupby('category_1') \\\n .plot_timeseries(\n date_column = 'order_date',\n value_column = 'total_price_sum',\n smooth_frac = 0.8,\n engine = \"plotly\"\n )\n\n\n \n\n\n\n\n\n4 š Documentation\nNext step? Learn more with the pytimetk documentation\n\nš Overview\nš Getting Started\nšŗļø Beginner Guides\nšApplied Data Science Tutorials with PyTimeTK\nšļøSpeed Comparisons\nš API Reference\n\n\n\n5 š» Contributing\nInterested in helping us make this the best Python package for time series analysis? Weād love your help.\nFollow these instructions to Contribute.\n\n\n6 š More Coming Soonā¦\nWe are in the early stages of development. But itās obvious the potential for pytimetk now in Python. š\n\nPlease ā us on GitHub (it takes 2-seconds and means a lot).\nTo make requests, please see our Project Roadmap GH Issue #2. You can make requests there.\nWant to contribute? See our contributing guide here."
+ },
+ {
+ "objectID": "getting-started/02_quick_start.html",
+ "href": "getting-started/02_quick_start.html",
+ "title": "Quick Start",
+ "section": "",
+ "text": "This is a simple exercise to showcase the power of our 2 most popular function:\n\nsummarize_by_time()\nplot_timeseries()\n\n\n\nFirst, import pytimetk as tk. This gets you access to the most important functions. Use tk.load_dataset() to load the ābike_sales_sampleā dataset.\n\n\n\n\n\n\nAbout the Bike Sales Sample Dataset\n\n\n\n\n\nThis dataset contains āorderlinesā for orders recieved. The order_date column contains timestamps. We can use this column to peform sales aggregations (e.g.Ā total revenue).\n\n\n\n\n\nCode\nimport pytimetk as tk\nimport pandas as pd\n\ndf = tk.load_dataset('bike_sales_sample')\ndf['order_date'] = pd.to_datetime(df['order_date'])\n\ndf \n\n\n\n\n\n\n\n\n\norder_id\norder_line\norder_date\nquantity\nprice\ntotal_price\nmodel\ncategory_1\ncategory_2\nframe_material\nbikeshop_name\ncity\nstate\n\n\n\n\n0\n1\n1\n2011-01-07\n1\n6070\n6070\nJekyll Carbon 2\nMountain\nOver Mountain\nCarbon\nIthaca Mountain Climbers\nIthaca\nNY\n\n\n1\n1\n2\n2011-01-07\n1\n5970\n5970\nTrigger Carbon 2\nMountain\nOver Mountain\nCarbon\nIthaca Mountain Climbers\nIthaca\nNY\n\n\n2\n2\n1\n2011-01-10\n1\n2770\n2770\nBeast of the East 1\nMountain\nTrail\nAluminum\nKansas City 29ers\nKansas City\nKS\n\n\n3\n2\n2\n2011-01-10\n1\n5970\n5970\nTrigger Carbon 2\nMountain\nOver Mountain\nCarbon\nKansas City 29ers\nKansas City\nKS\n\n\n4\n3\n1\n2011-01-10\n1\n10660\n10660\nSupersix Evo Hi-Mod Team\nRoad\nElite Road\nCarbon\nLouisville Race Equipment\nLouisville\nKY\n\n\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n\n\n2461\n321\n3\n2011-12-22\n1\n1410\n1410\nCAAD8 105\nRoad\nElite Road\nAluminum\nMiami Race Equipment\nMiami\nFL\n\n\n2462\n322\n1\n2011-12-28\n1\n1250\n1250\nSynapse Disc Tiagra\nRoad\nEndurance Road\nAluminum\nPhoenix Bi-peds\nPhoenix\nAZ\n\n\n2463\n322\n2\n2011-12-28\n1\n2660\n2660\nBad Habit 2\nMountain\nTrail\nAluminum\nPhoenix Bi-peds\nPhoenix\nAZ\n\n\n2464\n322\n3\n2011-12-28\n1\n2340\n2340\nF-Si 1\nMountain\nCross Country Race\nAluminum\nPhoenix Bi-peds\nPhoenix\nAZ\n\n\n2465\n322\n4\n2011-12-28\n1\n5860\n5860\nSynapse Hi-Mod Dura Ace\nRoad\nEndurance Road\nCarbon\nPhoenix Bi-peds\nPhoenix\nAZ\n\n\n\n\n2466 rows Ć 13 columns\n\n\n\n\n\n\nYour company might be interested in sales patterns for various categories of bicycles. We can obtain a grouped monthly sales aggregation by category_1 in two lines of code:\n\nFirst use pandasās groupby() method to group the DataFrame on category_1\nNext, use timetkās summarize_by_time() method to apply the sum function my month start (āMSā) and use wide_format = 'False' to return the dataframe in a long format (Note long format is the default).\n\nThe result is the total revenue for Mountain and Road bikes by month.\n\n\nCode\nsummary_category_1_df = df \\\n .groupby(\"category_1\") \\\n .summarize_by_time(\n date_column = 'order_date', \n value_column = 'total_price',\n freq = \"MS\",\n agg_func = 'sum',\n wide_format = False\n )\n\n# First 5 rows shown\nsummary_category_1_df.head()\n\n\n\n\n\n\n\n\n\ncategory_1\norder_date\ntotal_price\n\n\n\n\n0\nMountain\n2011-01-01\n221490\n\n\n1\nMountain\n2011-02-01\n660555\n\n\n2\nMountain\n2011-03-01\n358855\n\n\n3\nMountain\n2011-04-01\n1075975\n\n\n4\nMountain\n2011-05-01\n450440\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nNow available: plot_timeseries().\n\n\n\n\n\nPlot time series is a quick and easy way to visualize time series and make professional time series plots.\n\n\n\nWith the data summarized by time, we can visualize with plot_timeseries(). pytimetk functions are groupby() aware meaning they understand if your data is grouped to do things by group. This is useful in time series where we often deal with 100s of time series groups.\n\n\nCode\nsummary_category_1_df \\\n .groupby('category_1') \\\n .plot_timeseries(\n date_column = 'order_date',\n value_column = 'total_price',\n smooth_frac = 0.8\n )"
+ },
+ {
+ "objectID": "getting-started/02_quick_start.html#import-libraries-data",
+ "href": "getting-started/02_quick_start.html#import-libraries-data",
+ "title": "Quick Start",
+ "section": "",
+ "text": "First, import pytimetk as tk. This gets you access to the most important functions. Use tk.load_dataset() to load the ābike_sales_sampleā dataset.\n\n\n\n\n\n\nAbout the Bike Sales Sample Dataset\n\n\n\n\n\nThis dataset contains āorderlinesā for orders recieved. The order_date column contains timestamps. We can use this column to peform sales aggregations (e.g.Ā total revenue).\n\n\n\n\n\nCode\nimport pytimetk as tk\nimport pandas as pd\n\ndf = tk.load_dataset('bike_sales_sample')\ndf['order_date'] = pd.to_datetime(df['order_date'])\n\ndf \n\n\n\n\n\n\n\n\n\norder_id\norder_line\norder_date\nquantity\nprice\ntotal_price\nmodel\ncategory_1\ncategory_2\nframe_material\nbikeshop_name\ncity\nstate\n\n\n\n\n0\n1\n1\n2011-01-07\n1\n6070\n6070\nJekyll Carbon 2\nMountain\nOver Mountain\nCarbon\nIthaca Mountain Climbers\nIthaca\nNY\n\n\n1\n1\n2\n2011-01-07\n1\n5970\n5970\nTrigger Carbon 2\nMountain\nOver Mountain\nCarbon\nIthaca Mountain Climbers\nIthaca\nNY\n\n\n2\n2\n1\n2011-01-10\n1\n2770\n2770\nBeast of the East 1\nMountain\nTrail\nAluminum\nKansas City 29ers\nKansas City\nKS\n\n\n3\n2\n2\n2011-01-10\n1\n5970\n5970\nTrigger Carbon 2\nMountain\nOver Mountain\nCarbon\nKansas City 29ers\nKansas City\nKS\n\n\n4\n3\n1\n2011-01-10\n1\n10660\n10660\nSupersix Evo Hi-Mod Team\nRoad\nElite Road\nCarbon\nLouisville Race Equipment\nLouisville\nKY\n\n\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n\n\n2461\n321\n3\n2011-12-22\n1\n1410\n1410\nCAAD8 105\nRoad\nElite Road\nAluminum\nMiami Race Equipment\nMiami\nFL\n\n\n2462\n322\n1\n2011-12-28\n1\n1250\n1250\nSynapse Disc Tiagra\nRoad\nEndurance Road\nAluminum\nPhoenix Bi-peds\nPhoenix\nAZ\n\n\n2463\n322\n2\n2011-12-28\n1\n2660\n2660\nBad Habit 2\nMountain\nTrail\nAluminum\nPhoenix Bi-peds\nPhoenix\nAZ\n\n\n2464\n322\n3\n2011-12-28\n1\n2340\n2340\nF-Si 1\nMountain\nCross Country Race\nAluminum\nPhoenix Bi-peds\nPhoenix\nAZ\n\n\n2465\n322\n4\n2011-12-28\n1\n5860\n5860\nSynapse Hi-Mod Dura Ace\nRoad\nEndurance Road\nCarbon\nPhoenix Bi-peds\nPhoenix\nAZ\n\n\n\n\n2466 rows Ć 13 columns"
+ },
+ {
+ "objectID": "getting-started/02_quick_start.html#using-summarize_by_time-for-a-sales-analysis",
+ "href": "getting-started/02_quick_start.html#using-summarize_by_time-for-a-sales-analysis",
+ "title": "Quick Start",
+ "section": "",
+ "text": "Your company might be interested in sales patterns for various categories of bicycles. We can obtain a grouped monthly sales aggregation by category_1 in two lines of code:\n\nFirst use pandasās groupby() method to group the DataFrame on category_1\nNext, use timetkās summarize_by_time() method to apply the sum function my month start (āMSā) and use wide_format = 'False' to return the dataframe in a long format (Note long format is the default).\n\nThe result is the total revenue for Mountain and Road bikes by month.\n\n\nCode\nsummary_category_1_df = df \\\n .groupby(\"category_1\") \\\n .summarize_by_time(\n date_column = 'order_date', \n value_column = 'total_price',\n freq = \"MS\",\n agg_func = 'sum',\n wide_format = False\n )\n\n# First 5 rows shown\nsummary_category_1_df.head()\n\n\n\n\n\n\n\n\n\ncategory_1\norder_date\ntotal_price\n\n\n\n\n0\nMountain\n2011-01-01\n221490\n\n\n1\nMountain\n2011-02-01\n660555\n\n\n2\nMountain\n2011-03-01\n358855\n\n\n3\nMountain\n2011-04-01\n1075975\n\n\n4\nMountain\n2011-05-01\n450440"
+ },
+ {
+ "objectID": "getting-started/02_quick_start.html#visualizing-sales-patterns",
+ "href": "getting-started/02_quick_start.html#visualizing-sales-patterns",
+ "title": "Quick Start",
+ "section": "",
+ "text": "Now available: plot_timeseries().\n\n\n\n\n\nPlot time series is a quick and easy way to visualize time series and make professional time series plots.\n\n\n\nWith the data summarized by time, we can visualize with plot_timeseries(). pytimetk functions are groupby() aware meaning they understand if your data is grouped to do things by group. This is useful in time series where we often deal with 100s of time series groups.\n\n\nCode\nsummary_category_1_df \\\n .groupby('category_1') \\\n .plot_timeseries(\n date_column = 'order_date',\n value_column = 'total_price',\n smooth_frac = 0.8\n )"
+ },
+ {
+ "objectID": "guides/02_timetk_concepts.html",
+ "href": "guides/02_timetk_concepts.html",
+ "title": "PyTimeTK Basics",
+ "section": "",
+ "text": "PyTimeTK has one mission: To make time series analysis simpler, easier, and faster in Python. This goal requires some opinionated ways of treating time series in Python. We will conceptually lay out how pytimetk can help.\nLetās first start with how to think about time series data conceptually. Time series data has 3 core properties."
+ },
+ {
+ "objectID": "guides/02_timetk_concepts.html#type-1-pandas-dataframe-operations",
+ "href": "guides/02_timetk_concepts.html#type-1-pandas-dataframe-operations",
+ "title": "PyTimeTK Basics",
+ "section": "2.1 Type 1: Pandas DataFrame Operations",
+ "text": "2.1 Type 1: Pandas DataFrame Operations\nBefore we start using pytimetk, letās make sure our data is set up properly.\n\nTimetk Data Format Compliance\n\n\n\n\n\n\n3 Core Properties Must Be Upheald\n\n\n\n\n\nA pytimetk-Compliant Pandas DataFrame must have:\n\nTime Series Index: A Time Stamp column containing datetime64 values\nValue Column(s): The value column(s) containing float or int values\nGroup Column(s): Optionally for grouped time series analysis, one or more columns containg str or categorical values (shown as an object)\n\nIf these are NOT upheld, this will impact your ability to use pytimetk DataFrame operations.\n\n\n\n\n\n\n\n\n\nInspect the DataFrame\n\n\n\n\n\nUse the tk.glimpse() method to check compliance.\n\n\n\nUsing pytimetk glimpse() method, we can see that we have a compliant data frame with a date column containing datetime64 and a value column containing float64. For grouped analysis we have the id column containing object dtype.\n\n\nCode\n# Tip: Inspect for compliance with glimpse()\nm4_daily_df.glimpse()\n\n\n<class 'pandas.core.frame.DataFrame'>: 9743 rows of 3 columns\nid: object ['D10', 'D10', 'D10', 'D10', 'D10', 'D10', 'D1 ...\ndate: datetime64[ns] [Timestamp('2014-07-03 00:00:00'), Timestamp(' ...\nvalue: float64 [2076.2, 2073.4, 2048.7, 2048.9, 2006.4, 2017. ...\n\n\n\n\nGrouped Time Series Analysis with Summarize By Time\nFirst, inspect how the summarize_by_time function works by calling help().\n\n\nCode\n# Review the summarize_by_time documentation (output not shown)\nhelp(tk.summarize_by_time)\n\n\n\n\n\n\n\n\nHelp Doc Info: summarize_by_time()\n\n\n\n\n\n\nThe first parameter is data, indicating this is a DataFrame operation.\nThe Examples show different use cases for how to apply the function on a DataFrame\n\n\n\n\nLetās test the summarize_by_time() DataFrame operation out using the grouped approach with method chaining. DataFrame operations can be used as Pandas methods with method-chaining, which allows us to more succinctly apply time series operations.\n\n\nCode\n# Grouped Summarize By Time with Method Chaining\ndf_summarized = (\n m4_daily_df\n .groupby('id')\n .summarize_by_time(\n date_column = 'date',\n value_column = 'value',\n freq = 'QS', # QS = Quarter Start\n agg_func = [\n 'mean', \n 'median', \n 'min',\n ('q25', lambda x: np.quantile(x, 0.25)),\n ('q75', lambda x: np.quantile(x, 0.75)),\n 'max',\n ('range',lambda x: x.max() - x.min()),\n ],\n )\n)\n\ndf_summarized\n\n\n\n\n\n\n\n\n\nid\ndate\nvalue_mean\nvalue_median\nvalue_min\nvalue_q25\nvalue_q75\nvalue_max\nvalue_range\n\n\n\n\n0\nD10\n2014-07-01\n1960.078889\n1979.90\n1781.6\n1915.225\n2002.575\n2076.2\n294.6\n\n\n1\nD10\n2014-10-01\n2184.586957\n2154.05\n2022.8\n2125.075\n2274.150\n2344.9\n322.1\n\n\n2\nD10\n2015-01-01\n2309.830000\n2312.30\n2209.6\n2284.575\n2342.150\n2392.4\n182.8\n\n\n3\nD10\n2015-04-01\n2344.481319\n2333.00\n2185.1\n2301.750\n2391.000\n2499.8\n314.7\n\n\n4\nD10\n2015-07-01\n2156.754348\n2186.70\n1856.6\n1997.250\n2289.425\n2368.1\n511.5\n\n\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n\n\n105\nD500\n2011-07-01\n9727.321739\n9745.55\n8964.5\n9534.125\n10003.900\n10463.9\n1499.4\n\n\n106\nD500\n2011-10-01\n8175.565217\n7897.00\n6755.0\n7669.875\n8592.575\n9860.0\n3105.0\n\n\n107\nD500\n2012-01-01\n8291.317582\n8412.60\n7471.5\n7814.800\n8677.850\n8980.7\n1509.2\n\n\n108\nD500\n2012-04-01\n8654.020879\n8471.10\n8245.6\n8389.850\n9017.250\n9349.2\n1103.6\n\n\n109\nD500\n2012-07-01\n8770.502353\n8690.50\n8348.1\n8604.400\n8846.000\n9545.3\n1197.2\n\n\n\n\n110 rows Ć 9 columns\n\n\n\n\n\n\n\n\n\nKey Takeaways: summarize_by_time()\n\n\n\n\n\n\nThe data must comply with the 3 core properties (date column, value column(s), and group column(s))\nThe aggregation functions were applied by combination of group (id) and resample (Quarter Start)\nThe result was a pandas DataFrame with group column, resampled date column, and summary values (mean, median, min, 25th-quantile, etc)\n\n\n\n\n\n\nAnother DataFrame Example: Creating 29 Engineered Features\nLetās examine another DataFrame function, tk.augment_timeseries_signature(). Feel free to inspect the documentation with help(tk.augment_timeseries_signature).\n\n\nCode\n# Creating 29 engineered features from the date column\n# Not run: help(tk.augment_timeseries_signature)\ndf_augmented = (\n m4_daily_df\n .augment_timeseries_signature(date_column = 'date')\n)\n\ndf_augmented.head()\n\n\n\n\n\n\n\n\n\nid\ndate\nvalue\ndate_index_num\ndate_year\ndate_year_iso\ndate_yearstart\ndate_yearend\ndate_leapyear\ndate_half\n...\ndate_mday\ndate_qday\ndate_yday\ndate_weekend\ndate_hour\ndate_minute\ndate_second\ndate_msecond\ndate_nsecond\ndate_am_pm\n\n\n\n\n0\nD10\n2014-07-03\n2076.2\n1404345600\n2014\n2014\n0\n0\n0\n2\n...\n3\n3\n184\n0\n0\n0\n0\n0\n0\nam\n\n\n1\nD10\n2014-07-04\n2073.4\n1404432000\n2014\n2014\n0\n0\n0\n2\n...\n4\n4\n185\n0\n0\n0\n0\n0\n0\nam\n\n\n2\nD10\n2014-07-05\n2048.7\n1404518400\n2014\n2014\n0\n0\n0\n2\n...\n5\n5\n186\n0\n0\n0\n0\n0\n0\nam\n\n\n3\nD10\n2014-07-06\n2048.9\n1404604800\n2014\n2014\n0\n0\n0\n2\n...\n6\n6\n187\n1\n0\n0\n0\n0\n0\nam\n\n\n4\nD10\n2014-07-07\n2006.4\n1404691200\n2014\n2014\n0\n0\n0\n2\n...\n7\n7\n188\n0\n0\n0\n0\n0\n0\nam\n\n\n\n\n5 rows Ć 32 columns\n\n\n\n\n\n\n\n\n\nKey Takeaways: augment_timeseries_signature()\n\n\n\n\n\n\nThe data must comply with the 1 of the 3 core properties (date column)\nThe result was a pandas DataFrame with 29 time series features that can be used for Machine Learning and Forecasting\n\n\n\n\n\n\nMaking Future Dates with Future Frame\nA common time series task before forecasting with machine learning models is to make a future DataFrame some length_out into the future. You can do this with tk.future_frame(). Hereās how.\n\n\nCode\n# Preparing a time series data set for Machine Learning Forecasting\nfull_augmented_df = (\n m4_daily_df \n .groupby('id')\n .future_frame('date', length_out = 365)\n .augment_timeseries_signature('date')\n)\nfull_augmented_df\n\n\n\n\n\n\n\n\n\n\n\n\nid\ndate\nvalue\ndate_index_num\ndate_year\ndate_year_iso\ndate_yearstart\ndate_yearend\ndate_leapyear\ndate_half\n...\ndate_mday\ndate_qday\ndate_yday\ndate_weekend\ndate_hour\ndate_minute\ndate_second\ndate_msecond\ndate_nsecond\ndate_am_pm\n\n\n\n\n0\nD10\n2014-07-03\n2076.2\n1404345600\n2014\n2014\n0\n0\n0\n2\n...\n3\n3\n184\n0\n0\n0\n0\n0\n0\nam\n\n\n1\nD10\n2014-07-04\n2073.4\n1404432000\n2014\n2014\n0\n0\n0\n2\n...\n4\n4\n185\n0\n0\n0\n0\n0\n0\nam\n\n\n2\nD10\n2014-07-05\n2048.7\n1404518400\n2014\n2014\n0\n0\n0\n2\n...\n5\n5\n186\n0\n0\n0\n0\n0\n0\nam\n\n\n3\nD10\n2014-07-06\n2048.9\n1404604800\n2014\n2014\n0\n0\n0\n2\n...\n6\n6\n187\n1\n0\n0\n0\n0\n0\nam\n\n\n4\nD10\n2014-07-07\n2006.4\n1404691200\n2014\n2014\n0\n0\n0\n2\n...\n7\n7\n188\n0\n0\n0\n0\n0\n0\nam\n\n\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n\n\n11198\nD500\n2013-09-19\nNaN\n1379548800\n2013\n2013\n0\n0\n0\n2\n...\n19\n81\n262\n0\n0\n0\n0\n0\n0\nam\n\n\n11199\nD500\n2013-09-20\nNaN\n1379635200\n2013\n2013\n0\n0\n0\n2\n...\n20\n82\n263\n0\n0\n0\n0\n0\n0\nam\n\n\n11200\nD500\n2013-09-21\nNaN\n1379721600\n2013\n2013\n0\n0\n0\n2\n...\n21\n83\n264\n0\n0\n0\n0\n0\n0\nam\n\n\n11201\nD500\n2013-09-22\nNaN\n1379808000\n2013\n2013\n0\n0\n0\n2\n...\n22\n84\n265\n1\n0\n0\n0\n0\n0\nam\n\n\n11202\nD500\n2013-09-23\nNaN\n1379894400\n2013\n2013\n0\n0\n0\n2\n...\n23\n85\n266\n0\n0\n0\n0\n0\n0\nam\n\n\n\n\n11203 rows Ć 32 columns\n\n\n\nWe can then get the future data by keying in on the data with value column that is missing (np.nan).\n\n\nCode\n# Get the future data (just the observations that haven't happened yet)\nfuture_df = (\n full_augmented_df\n .query('value.isna()')\n)\nfuture_df\n\n\n\n\n\n\n\n\n\nid\ndate\nvalue\ndate_index_num\ndate_year\ndate_year_iso\ndate_yearstart\ndate_yearend\ndate_leapyear\ndate_half\n...\ndate_mday\ndate_qday\ndate_yday\ndate_weekend\ndate_hour\ndate_minute\ndate_second\ndate_msecond\ndate_nsecond\ndate_am_pm\n\n\n\n\n9743\nD10\n2016-05-07\nNaN\n1462579200\n2016\n2016\n0\n0\n1\n1\n...\n7\n37\n128\n0\n0\n0\n0\n0\n0\nam\n\n\n9744\nD10\n2016-05-08\nNaN\n1462665600\n2016\n2016\n0\n0\n1\n1\n...\n8\n38\n129\n1\n0\n0\n0\n0\n0\nam\n\n\n9745\nD10\n2016-05-09\nNaN\n1462752000\n2016\n2016\n0\n0\n1\n1\n...\n9\n39\n130\n0\n0\n0\n0\n0\n0\nam\n\n\n9746\nD10\n2016-05-10\nNaN\n1462838400\n2016\n2016\n0\n0\n1\n1\n...\n10\n40\n131\n0\n0\n0\n0\n0\n0\nam\n\n\n9747\nD10\n2016-05-11\nNaN\n1462924800\n2016\n2016\n0\n0\n1\n1\n...\n11\n41\n132\n0\n0\n0\n0\n0\n0\nam\n\n\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n\n\n11198\nD500\n2013-09-19\nNaN\n1379548800\n2013\n2013\n0\n0\n0\n2\n...\n19\n81\n262\n0\n0\n0\n0\n0\n0\nam\n\n\n11199\nD500\n2013-09-20\nNaN\n1379635200\n2013\n2013\n0\n0\n0\n2\n...\n20\n82\n263\n0\n0\n0\n0\n0\n0\nam\n\n\n11200\nD500\n2013-09-21\nNaN\n1379721600\n2013\n2013\n0\n0\n0\n2\n...\n21\n83\n264\n0\n0\n0\n0\n0\n0\nam\n\n\n11201\nD500\n2013-09-22\nNaN\n1379808000\n2013\n2013\n0\n0\n0\n2\n...\n22\n84\n265\n1\n0\n0\n0\n0\n0\nam\n\n\n11202\nD500\n2013-09-23\nNaN\n1379894400\n2013\n2013\n0\n0\n0\n2\n...\n23\n85\n266\n0\n0\n0\n0\n0\n0\nam\n\n\n\n\n1460 rows Ć 32 columns"
+ },
+ {
+ "objectID": "guides/02_timetk_concepts.html#type-2-pandas-series-operations",
+ "href": "guides/02_timetk_concepts.html#type-2-pandas-series-operations",
+ "title": "PyTimeTK Basics",
+ "section": "2.2 Type 2: Pandas Series Operations",
+ "text": "2.2 Type 2: Pandas Series Operations\nThe main difference between a DataFrame operation and a Series operation is that we are operating on an array of values from typically one of the following dtypes:\n\nTimestamps (datetime64)\nNumeric (float64 or int64)\n\nThe first argument of Series operations that operate on Timestamps will always be idx.\nLetās take a look at one shall we? Weāll start with a common action: Making future time series from an existing time series with a regular frequency.\n\nThe Make Future Time Series Function\nSay we have a monthly sequence of timestamps. What if we want to create a forecast where we predict 12 months into the future? Well, we will need to create 12 future timestamps. Hereās how.\nFirst create a pd.date_range() with dates starting at the beginning of each month.\n\n\nCode\n# Make a monthly date range\ndates_dt = pd.date_range(\"2023-01\", \"2024-01\", freq=\"MS\")\ndates_dt\n\n\nDatetimeIndex(['2023-01-01', '2023-02-01', '2023-03-01', '2023-04-01',\n '2023-05-01', '2023-06-01', '2023-07-01', '2023-08-01',\n '2023-09-01', '2023-10-01', '2023-11-01', '2023-12-01',\n '2024-01-01'],\n dtype='datetime64[ns]', freq='MS')\n\n\nNext, use tk.make_future_timeseries() to create the next 12 timestamps in the sequence.\n\nPandas SeriesDateTimeIndex\n\n\n\n\nCode\n# Pandas Series: Future Dates\nfuture_series = pd.Series(dates_dt).make_future_timeseries(12)\nfuture_series\n\n\n0 2024-02-01\n1 2024-03-01\n2 2024-04-01\n3 2024-05-01\n4 2024-06-01\n5 2024-07-01\n6 2024-08-01\n7 2024-09-01\n8 2024-10-01\n9 2024-11-01\n10 2024-12-01\n11 2025-01-01\ndtype: datetime64[ns]\n\n\n\n\n\n\nCode\n# DateTimeIndex: Future Dates\nfuture_dt = tk.make_future_timeseries(\n idx = dates_dt,\n length_out = 12\n)\nfuture_dt\n\n\n0 2024-02-01\n1 2024-03-01\n2 2024-04-01\n3 2024-05-01\n4 2024-06-01\n5 2024-07-01\n6 2024-08-01\n7 2024-09-01\n8 2024-10-01\n9 2024-11-01\n10 2024-12-01\n11 2025-01-01\ndtype: datetime64[ns]\n\n\n\n\n\nWe can combine the actual and future timestamps into one combined timeseries.\n\n\nCode\n# Combining the 2 series and resetting the index\ncombined_timeseries = (\n pd.concat(\n [pd.Series(dates_dt), pd.Series(future_dt)],\n axis=0\n )\n .reset_index(drop = True)\n)\n\ncombined_timeseries\n\n\n0 2023-01-01\n1 2023-02-01\n2 2023-03-01\n3 2023-04-01\n4 2023-05-01\n5 2023-06-01\n6 2023-07-01\n7 2023-08-01\n8 2023-09-01\n9 2023-10-01\n10 2023-11-01\n11 2023-12-01\n12 2024-01-01\n13 2024-02-01\n14 2024-03-01\n15 2024-04-01\n16 2024-05-01\n17 2024-06-01\n18 2024-07-01\n19 2024-08-01\n20 2024-09-01\n21 2024-10-01\n22 2024-11-01\n23 2024-12-01\n24 2025-01-01\ndtype: datetime64[ns]\n\n\nNext, weāll take a look at how to go from an irregular time series to a regular time series.\n\n\nFlooring Dates\nAn example is tk.floor_date, which is used to round down dates. See help(tk.floor_date).\nFlooring dates is often used as part of a strategy to go from an irregular time series to regular by combining with an aggregation. Often summarize_by_time() is used (Iāll share why shortly). But conceptually, date flooring is the secret.\n\nWith FlooringWithout Flooring\n\n\n\n\nCode\n# Monthly flooring rounds dates down to 1st of the month\nm4_daily_df['date'].floor_date(unit = \"M\")\n\n\n0 2014-07-01\n1 2014-07-01\n2 2014-07-01\n3 2014-07-01\n4 2014-07-01\n ... \n9738 2014-07-01\n9739 2014-07-01\n9740 2014-07-01\n9741 2014-07-01\n9742 2014-07-01\nName: date, Length: 9743, dtype: datetime64[ns]\n\n\n\n\n\n\nCode\n# Before Flooring\nm4_daily_df['date']\n\n\n0 2014-07-03\n1 2014-07-03\n2 2014-07-03\n3 2014-07-03\n4 2014-07-03\n ... \n9738 2014-07-03\n9739 2014-07-03\n9740 2014-07-03\n9741 2014-07-03\n9742 2014-07-03\nName: date, Length: 9743, dtype: datetime64[ns]\n\n\n\n\n\nThis ādate flooringā operation can be useful for creating date groupings.\n\n\nCode\n# Adding a date group with floor_date()\ndates_grouped_by_month = (\n m4_daily_df\n .assign(date_group = lambda x: x['date'].floor_date(\"M\"))\n)\n\ndates_grouped_by_month\n\n\n\n\n\n\n\n\n\nid\ndate\nvalue\ndate_group\n\n\n\n\n0\nD10\n2014-07-03\n2076.2\n2014-07-01\n\n\n1\nD10\n2014-07-03\n2073.4\n2014-07-01\n\n\n2\nD10\n2014-07-03\n2048.7\n2014-07-01\n\n\n3\nD10\n2014-07-03\n2048.9\n2014-07-01\n\n\n4\nD10\n2014-07-03\n2006.4\n2014-07-01\n\n\n...\n...\n...\n...\n...\n\n\n9738\nD500\n2014-07-03\n9418.8\n2014-07-01\n\n\n9739\nD500\n2014-07-03\n9365.7\n2014-07-01\n\n\n9740\nD500\n2014-07-03\n9445.9\n2014-07-01\n\n\n9741\nD500\n2014-07-03\n9497.9\n2014-07-01\n\n\n9742\nD500\n2014-07-03\n9545.3\n2014-07-01\n\n\n\n\n9743 rows Ć 4 columns\n\n\n\nWe can then do grouped operations.\n\n\nCode\n# Example of a grouped operation with floored dates\nsummary_df = (\n dates_grouped_by_month\n .drop('date', axis=1) \\\n .groupby(['id', 'date_group'])\n .mean() \\\n .reset_index()\n)\n\nsummary_df\n\n\n\n\n\n\n\n\n\nid\ndate_group\nvalue\n\n\n\n\n0\nD10\n2014-07-01\n2261.606825\n\n\n1\nD160\n2014-07-01\n9243.155254\n\n\n2\nD410\n2014-07-01\n8259.786346\n\n\n3\nD500\n2014-07-01\n8287.728789\n\n\n\n\n\n\n\nOf course for this operation, we can do it faster with summarize_by_time() (and itās much more flexible).\n\n\nCode\n# Summarize by time is less code and more flexible\n(\n m4_daily_df \n .groupby('id')\n .summarize_by_time(\n 'date', 'value', \n freq = \"MS\",\n agg_func = ['mean', 'median', 'min', 'max']\n )\n)\n\n\n\n\n\n\n\n\n\nid\ndate\nvalue_mean\nvalue_median\nvalue_min\nvalue_max\n\n\n\n\n0\nD10\n2014-07-01\n2261.606825\n2302.30\n1781.60\n2649.30\n\n\n1\nD160\n2014-07-01\n9243.155254\n10097.30\n1734.90\n19432.50\n\n\n2\nD410\n2014-07-01\n8259.786346\n8382.81\n6309.38\n9540.62\n\n\n3\nD500\n2014-07-01\n8287.728789\n7662.10\n4172.10\n14954.10\n\n\n\n\n\n\n\nAnd thatās the core idea behind pytimetk, writing less code and getting more.\nNext, letās do one more function. The brother of augment_timeseries_signature()ā¦\n\n\nThe Get Time Series Signature Function\nThis function takes a pandas Series or DateTimeIndex and returns a DataFrame containing the 29 engineered features.\nStart with either a DateTimeIndexā¦\n\n\nCode\ntimestamps_dt = pd.date_range(\"2023\", \"2024\", freq = \"D\")\ntimestamps_dt\n\n\nDatetimeIndex(['2023-01-01', '2023-01-02', '2023-01-03', '2023-01-04',\n '2023-01-05', '2023-01-06', '2023-01-07', '2023-01-08',\n '2023-01-09', '2023-01-10',\n ...\n '2023-12-23', '2023-12-24', '2023-12-25', '2023-12-26',\n '2023-12-27', '2023-12-28', '2023-12-29', '2023-12-30',\n '2023-12-31', '2024-01-01'],\n dtype='datetime64[ns]', length=366, freq='D')\n\n\nā¦ Or a Pandas Series.\n\n\nCode\ntimestamps_series = pd.Series(timestamps_dt)\ntimestamps_series\n\n\n0 2023-01-01\n1 2023-01-02\n2 2023-01-03\n3 2023-01-04\n4 2023-01-05\n ... \n361 2023-12-28\n362 2023-12-29\n363 2023-12-30\n364 2023-12-31\n365 2024-01-01\nLength: 366, dtype: datetime64[ns]\n\n\nAnd you can use the pandas Series function, tk.get_timeseries_signature() to create 29 features from the date sequence.\n\nPandas SeriesDateTimeIndex\n\n\n\n\nCode\n# Pandas series: get_timeseries_signature\ntimestamps_series.get_timeseries_signature()\n\n\n\n\n\n\n\n\n\nidx\nidx_index_num\nidx_year\nidx_year_iso\nidx_yearstart\nidx_yearend\nidx_leapyear\nidx_half\nidx_quarter\nidx_quarteryear\n...\nidx_mday\nidx_qday\nidx_yday\nidx_weekend\nidx_hour\nidx_minute\nidx_second\nidx_msecond\nidx_nsecond\nidx_am_pm\n\n\n\n\n0\n2023-01-01\n1672531200\n2023\n2022\n1\n0\n0\n1\n1\n2023Q1\n...\n1\n1\n1\n1\n0\n0\n0\n0\n0\nam\n\n\n1\n2023-01-02\n1672617600\n2023\n2023\n0\n0\n0\n1\n1\n2023Q1\n...\n2\n2\n2\n0\n0\n0\n0\n0\n0\nam\n\n\n2\n2023-01-03\n1672704000\n2023\n2023\n0\n0\n0\n1\n1\n2023Q1\n...\n3\n3\n3\n0\n0\n0\n0\n0\n0\nam\n\n\n3\n2023-01-04\n1672790400\n2023\n2023\n0\n0\n0\n1\n1\n2023Q1\n...\n4\n4\n4\n0\n0\n0\n0\n0\n0\nam\n\n\n4\n2023-01-05\n1672876800\n2023\n2023\n0\n0\n0\n1\n1\n2023Q1\n...\n5\n5\n5\n0\n0\n0\n0\n0\n0\nam\n\n\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n\n\n361\n2023-12-28\n1703721600\n2023\n2023\n0\n0\n0\n2\n4\n2023Q4\n...\n28\n89\n362\n0\n0\n0\n0\n0\n0\nam\n\n\n362\n2023-12-29\n1703808000\n2023\n2023\n0\n0\n0\n2\n4\n2023Q4\n...\n29\n90\n363\n0\n0\n0\n0\n0\n0\nam\n\n\n363\n2023-12-30\n1703894400\n2023\n2023\n0\n0\n0\n2\n4\n2023Q4\n...\n30\n91\n364\n0\n0\n0\n0\n0\n0\nam\n\n\n364\n2023-12-31\n1703980800\n2023\n2023\n0\n1\n0\n2\n4\n2023Q4\n...\n31\n92\n365\n1\n0\n0\n0\n0\n0\nam\n\n\n365\n2024-01-01\n1704067200\n2024\n2024\n1\n0\n1\n1\n1\n2024Q1\n...\n1\n1\n1\n0\n0\n0\n0\n0\n0\nam\n\n\n\n\n366 rows Ć 30 columns\n\n\n\n\n\n\n\nCode\n# DateTimeIndex: get_timeseries_signature\ntk.get_timeseries_signature(timestamps_dt)\n\n\n\n\n\n\n\n\n\nidx\nidx_index_num\nidx_year\nidx_year_iso\nidx_yearstart\nidx_yearend\nidx_leapyear\nidx_half\nidx_quarter\nidx_quarteryear\n...\nidx_mday\nidx_qday\nidx_yday\nidx_weekend\nidx_hour\nidx_minute\nidx_second\nidx_msecond\nidx_nsecond\nidx_am_pm\n\n\n\n\n0\n2023-01-01\n1672531200\n2023\n2022\n1\n0\n0\n1\n1\n2023Q1\n...\n1\n1\n1\n1\n0\n0\n0\n0\n0\nam\n\n\n1\n2023-01-02\n1672617600\n2023\n2023\n0\n0\n0\n1\n1\n2023Q1\n...\n2\n2\n2\n0\n0\n0\n0\n0\n0\nam\n\n\n2\n2023-01-03\n1672704000\n2023\n2023\n0\n0\n0\n1\n1\n2023Q1\n...\n3\n3\n3\n0\n0\n0\n0\n0\n0\nam\n\n\n3\n2023-01-04\n1672790400\n2023\n2023\n0\n0\n0\n1\n1\n2023Q1\n...\n4\n4\n4\n0\n0\n0\n0\n0\n0\nam\n\n\n4\n2023-01-05\n1672876800\n2023\n2023\n0\n0\n0\n1\n1\n2023Q1\n...\n5\n5\n5\n0\n0\n0\n0\n0\n0\nam\n\n\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n...\n\n\n361\n2023-12-28\n1703721600\n2023\n2023\n0\n0\n0\n2\n4\n2023Q4\n...\n28\n89\n362\n0\n0\n0\n0\n0\n0\nam\n\n\n362\n2023-12-29\n1703808000\n2023\n2023\n0\n0\n0\n2\n4\n2023Q4\n...\n29\n90\n363\n0\n0\n0\n0\n0\n0\nam\n\n\n363\n2023-12-30\n1703894400\n2023\n2023\n0\n0\n0\n2\n4\n2023Q4\n...\n30\n91\n364\n0\n0\n0\n0\n0\n0\nam\n\n\n364\n2023-12-31\n1703980800\n2023\n2023\n0\n1\n0\n2\n4\n2023Q4\n...\n31\n92\n365\n1\n0\n0\n0\n0\n0\nam\n\n\n365\n2024-01-01\n1704067200\n2024\n2024\n1\n0\n1\n1\n1\n2024Q1\n...\n1\n1\n1\n0\n0\n0\n0\n0\n0\nam\n\n\n\n\n366 rows Ć 30 columns"
+ },
+ {
+ "objectID": "guides/01_visualization.html",
+ "href": "guides/01_visualization.html",
+ "title": "Data Visualization",
+ "section": "",
+ "text": "How this guide benefits you\n\n\n\n\n\nThis guide covers how to use the plot_timeseries() for data visualization. Once you understand how it works, you can apply explore time series data easier than ever.\nThis tutorial focuses on, plot_timeseries(), a workhorse time-series plotting function that:"
+ },
+ {
+ "objectID": "guides/01_visualization.html#plotting-a-single-time-series",
+ "href": "guides/01_visualization.html#plotting-a-single-time-series",
+ "title": "Data Visualization",
+ "section": "2.1 Plotting a Single Time Series",
+ "text": "2.1 Plotting a Single Time Series\nLetās start with a popular time series, taylor_30_min, which includes energy demand in megawatts at a sampling interval of 30-minutes. This is a single time series.\n\n\nCode\n# Import a Time Series Data Set\ntaylor_30_min = tk.load_dataset(\"taylor_30_min\", parse_dates = ['date'])\ntaylor_30_min\n\n\n\n\n\n\n\n\n\ndate\nvalue\n\n\n\n\n0\n2000-06-05 00:00:00+00:00\n22262\n\n\n1\n2000-06-05 00:30:00+00:00\n21756\n\n\n2\n2000-06-05 01:00:00+00:00\n22247\n\n\n3\n2000-06-05 01:30:00+00:00\n22759\n\n\n4\n2000-06-05 02:00:00+00:00\n22549\n\n\n...\n...\n...\n\n\n4027\n2000-08-27 21:30:00+00:00\n27946\n\n\n4028\n2000-08-27 22:00:00+00:00\n27133\n\n\n4029\n2000-08-27 22:30:00+00:00\n25996\n\n\n4030\n2000-08-27 23:00:00+00:00\n24610\n\n\n4031\n2000-08-27 23:30:00+00:00\n23132\n\n\n\n\n4032 rows Ć 2 columns\n\n\n\nThe plot_timeseries() function generates an interactive plotly chart by default.\n\nSimply provide the date variable (time-based column, date_column) and the numeric variable (value_column) that changes over time as the first 2 arguments.\nBy default, the plotting engine is plotly, which is interactive and excellent for data exploration and apps. However, if you require static plots for reports, you can set the engine to engine = āplotnineā or engine = āmatplotlibā.\n\nInteractive plot\n\n\nCode\ntaylor_30_min.plot_timeseries('date', 'value')\n\n\n\n \n\n\nStatic plot\n\n\nCode\ntaylor_30_min.plot_timeseries(\n 'date', 'value',\n engine = 'plotnine'\n)\n\n\n\n\n\n<Figure Size: (700 x 500)>"
+ },
+ {
+ "objectID": "guides/01_visualization.html#plotting-groups",
+ "href": "guides/01_visualization.html#plotting-groups",
+ "title": "Data Visualization",
+ "section": "2.2 Plotting Groups",
+ "text": "2.2 Plotting Groups\nNext, letās move on to a dataset with time series groups, m4_monthly, which is a sample of 4 time series from the M4 competition that are sampled at a monthly frequency.\n\n\nCode\n# Import a Time Series Data Set\nm4_monthly = tk.load_dataset(\"m4_monthly\", parse_dates = ['date'])\nm4_monthly\n\n\n\n\n\n\n\n\n\nid\ndate\nvalue\n\n\n\n\n0\nM1\n1976-06-01\n8000\n\n\n1\nM1\n1976-07-01\n8350\n\n\n2\nM1\n1976-08-01\n8570\n\n\n3\nM1\n1976-09-01\n7700\n\n\n4\nM1\n1976-10-01\n7080\n\n\n...\n...\n...\n...\n\n\n1569\nM1000\n2015-02-01\n880\n\n\n1570\nM1000\n2015-03-01\n800\n\n\n1571\nM1000\n2015-04-01\n1140\n\n\n1572\nM1000\n2015-05-01\n970\n\n\n1573\nM1000\n2015-06-01\n1430\n\n\n\n\n1574 rows Ć 3 columns\n\n\n\nVisualizing grouped data is as simple as grouping the data set with groupby() before run it into the plot_timeseries() function. There are 2 methods:\n\nFacets\nPlotly Dropdown\n\n\nFacets (Subgroups on one plot)\nThis is great to see all time series in one plot. Here are the key points:\n\nGroups can be added using the pandas groupby().\nThese groups are then converted into facets.\nUsing facet_ncol = 2 returns a 2-column faceted plot.\nSetting facet_scales = \"free\" allows the x and y-axes of each plot to scale independently of the other plots.\n\n\n\nCode\nm4_monthly.groupby('id').plot_timeseries(\n 'date', 'value', \n facet_ncol = 2, \n facet_scales = \"free\"\n)\n\n\n\n \n\n\n\n\nPlotly Dropdown\nSometimes you have many groups and would prefer to see one plot per group. This can be accomplished with plotly_dropdown. You can adjust the x and y position as follows:\n\n\nCode\nm4_monthly.groupby('id').plot_timeseries(\n 'date', 'value', \n plotly_dropdown=True,\n plotly_dropdown_x=0,\n plotly_dropdown_y=1\n)\n\n\n\n \n\n\nThe groups can also be vizualized in the same plot using color_column paramenter. Letās come back to taylor_30_min dataframe.\n\n\nCode\n# load data\ntaylor_30_min = tk.load_dataset(\"taylor_30_min\", parse_dates = ['date'])\n\n# extract the month using pandas\ntaylor_30_min['month'] = pd.to_datetime(taylor_30_min['date']).dt.month\n\n# plot groups\ntaylor_30_min.plot_timeseries(\n 'date', 'value', \n color_column = 'month'\n)"
+ },
{
"objectID": "guides/05_augmenting.html",
"href": "guides/05_augmenting.html",
diff --git a/docs/_site/sitemap.xml b/docs/_site/sitemap.xml
index a62a7dea..2348c247 100644
--- a/docs/_site/sitemap.xml
+++ b/docs/_site/sitemap.xml
@@ -2,350 +2,354 @@
https://business-science.github.io/pytimetk/reference/ts_features.html
- 2024-11-06T02:17:26.592Z
+ 2024-11-06T04:06:27.207Zhttps://business-science.github.io/pytimetk/reference/floor_date.html
- 2024-11-06T02:17:23.752Z
+ 2024-11-06T04:06:24.296Zhttps://business-science.github.io/pytimetk/reference/ts_summary.html
- 2024-11-06T02:17:21.443Z
+ 2024-11-06T04:06:21.750Zhttps://business-science.github.io/pytimetk/reference/timeseries_unit_frequency_table.html
- 2024-11-06T02:17:18.802Z
+ 2024-11-06T04:06:18.937Zhttps://business-science.github.io/pytimetk/reference/transform_columns.html
- 2024-11-06T02:17:17.103Z
+ 2024-11-06T04:06:17.208Zhttps://business-science.github.io/pytimetk/reference/augment_timeseries_signature.html
- 2024-11-06T02:17:15.629Z
+ 2024-11-06T04:06:15.752Zhttps://business-science.github.io/pytimetk/reference/plot_anomalies.html
- 2024-11-06T02:17:13.709Z
+ 2024-11-06T04:06:13.741Zhttps://business-science.github.io/pytimetk/reference/augment_ewm.html
- 2024-11-06T02:17:11.023Z
+ 2024-11-06T04:06:10.896Zhttps://business-science.github.io/pytimetk/reference/make_future_timeseries.html
- 2024-11-06T02:17:09.301Z
+ 2024-11-06T04:06:09.148Zhttps://business-science.github.io/pytimetk/reference/augment_roc.html
- 2024-11-06T02:17:07.942Z
+ 2024-11-06T04:06:07.765Zhttps://business-science.github.io/pytimetk/reference/get_frequency_summary.html
- 2024-11-06T02:17:06.337Z
+ 2024-11-06T04:06:06.111Zhttps://business-science.github.io/pytimetk/reference/augment_qsmomentum.html
- 2024-11-06T02:17:04.634Z
+ 2024-11-06T04:06:04.379Zhttps://business-science.github.io/pytimetk/reference/augment_hilbert.html
- 2024-11-06T02:17:02.496Z
+ 2024-11-06T04:06:02.148Zhttps://business-science.github.io/pytimetk/reference/progress_apply.html
- 2024-11-06T02:16:59.750Z
+ 2024-11-06T04:05:59.291Zhttps://business-science.github.io/pytimetk/reference/flatten_multiindex_column_names.html
- 2024-11-06T02:16:58.156Z
+ 2024-11-06T04:05:57.647Zhttps://business-science.github.io/pytimetk/reference/make_weekday_sequence.html
- 2024-11-06T02:16:45.978Z
+ 2024-11-06T04:05:55.699Zhttps://business-science.github.io/pytimetk/reference/augment_pct_change.html
- 2024-11-06T02:16:44.613Z
+ 2024-11-06T04:05:54.385Zhttps://business-science.github.io/pytimetk/reference/get_trend_frequency.html
- 2024-11-06T02:16:41.896Z
+ 2024-11-06T04:05:51.553Zhttps://business-science.github.io/pytimetk/reference/augment_fourier.html
- 2024-11-06T02:16:40.338Z
+ 2024-11-06T04:05:49.970Zhttps://business-science.github.io/pytimetk/reference/correlate.html
- 2024-11-06T02:16:16.482Z
+ 2024-11-06T04:05:48.009Zhttps://business-science.github.io/pytimetk/reference/filter_by_time.html
- 2024-11-06T02:16:14.373Z
+ 2024-11-06T04:05:45.857Zhttps://business-science.github.io/pytimetk/reference/summarize_by_time.html
- 2024-11-06T02:16:11.976Z
+ 2024-11-06T04:05:43.444Zhttps://business-science.github.io/pytimetk/reference/drop_zero_variance.html
- 2024-11-06T02:16:08.761Z
+ 2024-11-06T04:05:40.163Zhttps://business-science.github.io/pytimetk/reference/index.html
- 2024-11-06T02:16:07.024Z
+ 2024-11-06T04:05:38.350Zhttps://business-science.github.io/pytimetk/reference/get_pandas_frequency.html
- 2024-11-06T02:16:05.021Z
+ 2024-11-06T04:05:36.219Zhttps://business-science.github.io/pytimetk/reference/palette_timetk.html
- 2024-11-06T02:16:03.568Z
+ 2024-11-06T04:05:34.753Zhttps://business-science.github.io/pytimetk/reference/get_date_summary.html
- 2024-11-06T02:16:02.070Z
+ 2024-11-06T04:05:33.247Zhttps://business-science.github.io/pytimetk/reference/is_holiday.html
- 2024-11-06T02:16:00.398Z
+ 2024-11-06T04:05:31.473Zhttps://business-science.github.io/pytimetk/reference/get_available_datasets.html
- 2024-11-06T02:15:58.808Z
+ 2024-11-06T04:05:29.878Zhttps://business-science.github.io/pytimetk/reference/augment_wavelet.html
- 2024-11-06T02:15:57.063Z
+ 2024-11-06T04:05:28.145Zhttps://business-science.github.io/pytimetk/reference/augment_cmo.html
- 2024-11-06T02:15:54.877Z
+ 2024-11-06T04:05:25.860Zhttps://business-science.github.io/pytimetk/reference/augment_expanding.html
- 2024-11-06T02:15:51.885Z
+ 2024-11-06T04:05:22.709Zhttps://business-science.github.io/pytimetk/reference/ceil_date.html
- 2024-11-06T02:15:49.821Z
+ 2024-11-06T04:05:20.598Zhttps://business-science.github.io/pytimetk/reference/augment_rolling_apply.html
- 2024-11-06T02:15:47.928Z
+ 2024-11-06T04:05:18.669Zhttps://business-science.github.io/pytimetk/reference/augment_rolling.html
- 2024-11-06T02:15:45.551Z
+ 2024-11-06T04:05:16.091Zhttps://business-science.github.io/pytimetk/performance/01_speed_comparisons.html
- 2024-11-06T02:15:43.076Z
+ 2024-11-06T04:05:13.479Zhttps://business-science.github.io/pytimetk/tutorials/01_sales_crm.html
- 2024-11-06T02:15:33.832Z
+ 2024-11-06T04:05:04.339Zhttps://business-science.github.io/pytimetk/tutorials/05_clustering.html
- 2024-11-06T02:15:25.874Z
+ 2024-11-06T04:04:56.757Zhttps://business-science.github.io/pytimetk/tutorials/04_anomaly_detection.html
- 2024-11-06T02:15:24.081Z
+ 2024-11-06T04:04:54.880Zhttps://business-science.github.io/pytimetk/guides/04_wrangling.html
- 2024-11-06T02:15:22.036Z
+ 2024-11-06T04:04:52.761Z
- https://business-science.github.io/pytimetk/guides/01_visualization.html
- 2024-11-06T02:15:16.637Z
+ https://business-science.github.io/pytimetk/guides/07_timeseries_crossvalidation.html
+ 2024-11-06T04:04:46.853Z
- https://business-science.github.io/pytimetk/guides/02_timetk_concepts.html
- 2024-11-06T02:15:14.326Z
+ https://business-science.github.io/pytimetk/guides/06_anomalize.html
+ 2024-11-06T04:04:25.699Z
- https://business-science.github.io/pytimetk/getting-started/02_quick_start.html
- 2024-11-06T02:15:09.441Z
+ https://business-science.github.io/pytimetk/guides/03_pandas_frequency.html
+ 2024-11-06T04:04:20.180Z
- https://business-science.github.io/pytimetk/index.html
- 2024-11-06T02:15:07.644Z
+ https://business-science.github.io/pytimetk/getting-started/01_installation.html
+ 2024-11-06T04:04:17.628Zhttps://business-science.github.io/pytimetk/contributing.html
- 2024-11-06T02:15:06.410Z
+ 2024-11-06T04:04:15.441Z
- https://business-science.github.io/pytimetk/getting-started/01_installation.html
- 2024-11-06T02:15:08.343Z
+ https://business-science.github.io/pytimetk/index.html
+ 2024-11-06T04:04:16.888Z
- https://business-science.github.io/pytimetk/guides/03_pandas_frequency.html
- 2024-11-06T02:15:10.423Z
+ https://business-science.github.io/pytimetk/getting-started/02_quick_start.html
+ 2024-11-06T04:04:18.984Z
- https://business-science.github.io/pytimetk/guides/06_anomalize.html
- 2024-11-06T02:15:15.418Z
+ https://business-science.github.io/pytimetk/guides/02_timetk_concepts.html
+ 2024-11-06T04:04:24.521Z
+
+
+ https://business-science.github.io/pytimetk/guides/01_visualization.html
+ 2024-11-06T04:04:26.961Zhttps://business-science.github.io/pytimetk/guides/05_augmenting.html
- 2024-11-06T02:15:20.041Z
+ 2024-11-06T04:04:50.441Zhttps://business-science.github.io/pytimetk/tutorials/06_correlationfunnel.html
- 2024-11-06T02:15:25.125Z
+ 2024-11-06T04:04:56.039Zhttps://business-science.github.io/pytimetk/tutorials/03_demand_forecasting.html
- 2024-11-06T02:15:29.467Z
+ 2024-11-06T04:05:00.096Zhttps://business-science.github.io/pytimetk/tutorials/02_finance.html
- 2024-11-06T02:15:41.916Z
+ 2024-11-06T04:05:12.252Zhttps://business-science.github.io/pytimetk/changelog-news.html
- 2024-11-06T02:15:44.115Z
+ 2024-11-06T04:05:14.548Zhttps://business-science.github.io/pytimetk/reference/plot_timeseries.html
- 2024-11-06T02:15:46.943Z
+ 2024-11-06T04:05:17.609Zhttps://business-science.github.io/pytimetk/reference/augment_diffs.html
- 2024-11-06T02:15:49.038Z
+ 2024-11-06T04:05:19.815Zhttps://business-science.github.io/pytimetk/reference/time_scale_template.html
- 2024-11-06T02:15:50.577Z
+ 2024-11-06T04:05:21.411Zhttps://business-science.github.io/pytimetk/reference/pad_by_time.html
- 2024-11-06T02:15:53.354Z
+ 2024-11-06T04:05:24.274Zhttps://business-science.github.io/pytimetk/reference/plot_anomalies_cleaned.html
- 2024-11-06T02:15:55.902Z
+ 2024-11-06T04:05:26.909Zhttps://business-science.github.io/pytimetk/reference/plot_anomaly_decomp.html
- 2024-11-06T02:15:58.035Z
+ 2024-11-06T04:05:29.147Zhttps://business-science.github.io/pytimetk/reference/get_seasonal_frequency.html
- 2024-11-06T02:15:59.583Z
+ 2024-11-06T04:05:30.665Zhttps://business-science.github.io/pytimetk/reference/plot_correlation_funnel.html
- 2024-11-06T02:16:01.332Z
+ 2024-11-06T04:05:32.468Zhttps://business-science.github.io/pytimetk/reference/glimpse.html
- 2024-11-06T02:16:02.842Z
+ 2024-11-06T04:05:34.014Zhttps://business-science.github.io/pytimetk/reference/augment_ppo.html
- 2024-11-06T02:16:04.533Z
+ 2024-11-06T04:05:35.728Zhttps://business-science.github.io/pytimetk/reference/augment_bbands.html
- 2024-11-06T02:16:06.008Z
+ 2024-11-06T04:05:37.328Zhttps://business-science.github.io/pytimetk/reference/apply_by_time.html
- 2024-11-06T02:16:08.268Z
+ 2024-11-06T04:05:39.680Zhttps://business-science.github.io/pytimetk/reference/plot_anomalies_decomp.html
- 2024-11-06T02:16:09.793Z
+ 2024-11-06T04:05:41.206Zhttps://business-science.github.io/pytimetk/reference/parallel_apply.html
- 2024-11-06T02:16:13.030Z
+ 2024-11-06T04:05:44.446Zhttps://business-science.github.io/pytimetk/reference/augment_expanding_apply.html
- 2024-11-06T02:16:15.411Z
+ 2024-11-06T04:05:46.928Zhttps://business-science.github.io/pytimetk/reference/TimeSeriesCVSplitter.html
- 2024-11-06T02:16:39.207Z
+ 2024-11-06T04:05:48.833Zhttps://business-science.github.io/pytimetk/reference/week_of_month.html
- 2024-11-06T02:16:41.132Z
+ 2024-11-06T04:05:50.766Zhttps://business-science.github.io/pytimetk/reference/future_frame.html
- 2024-11-06T02:16:43.539Z
+ 2024-11-06T04:05:53.261Zhttps://business-science.github.io/pytimetk/reference/get_frequency.html
- 2024-11-06T02:16:45.155Z
+ 2024-11-06T04:05:54.890Zhttps://business-science.github.io/pytimetk/reference/TimeSeriesCV.html
- 2024-11-06T02:16:57.347Z
+ 2024-11-06T04:05:56.874Zhttps://business-science.github.io/pytimetk/reference/theme_timetk.html
- 2024-11-06T02:16:58.969Z
+ 2024-11-06T04:05:58.497Zhttps://business-science.github.io/pytimetk/reference/augment_rsi.html
- 2024-11-06T02:17:01.235Z
+ 2024-11-06T04:06:00.808Zhttps://business-science.github.io/pytimetk/reference/augment_leads.html
- 2024-11-06T02:17:03.796Z
+ 2024-11-06T04:06:03.533Zhttps://business-science.github.io/pytimetk/reference/binarize.html
- 2024-11-06T02:17:05.527Z
+ 2024-11-06T04:06:05.313Zhttps://business-science.github.io/pytimetk/reference/load_dataset.html
- 2024-11-06T02:17:06.885Z
+ 2024-11-06T04:06:06.686Zhttps://business-science.github.io/pytimetk/reference/get_diff_summary.html
- 2024-11-06T02:17:08.489Z
+ 2024-11-06T04:06:08.293Zhttps://business-science.github.io/pytimetk/reference/make_weekend_sequence.html
- 2024-11-06T02:17:10.120Z
+ 2024-11-06T04:06:09.972Zhttps://business-science.github.io/pytimetk/reference/get_holiday_signature.html
- 2024-11-06T02:17:12.423Z
+ 2024-11-06T04:06:12.344Zhttps://business-science.github.io/pytimetk/reference/augment_atr.html
- 2024-11-06T02:17:14.767Z
+ 2024-11-06T04:06:14.835Zhttps://business-science.github.io/pytimetk/reference/augment_macd.html
- 2024-11-06T02:17:16.615Z
+ 2024-11-06T04:06:16.717Zhttps://business-science.github.io/pytimetk/reference/get_timeseries_signature.html
- 2024-11-06T02:17:17.999Z
+ 2024-11-06T04:06:18.108Zhttps://business-science.github.io/pytimetk/reference/augment_lags.html
- 2024-11-06T02:17:20.071Z
+ 2024-11-06T04:06:20.329Zhttps://business-science.github.io/pytimetk/reference/augment_holiday_signature.html
- 2024-11-06T02:17:22.983Z
+ 2024-11-06T04:06:23.464Zhttps://business-science.github.io/pytimetk/reference/anomalize.html
- 2024-11-06T02:17:25.624Z
+ 2024-11-06T04:06:26.246Z
diff --git a/docs/_site/tutorials/01_sales_crm.html b/docs/_site/tutorials/01_sales_crm.html
index a22436c6..7357a844 100644
--- a/docs/_site/tutorials/01_sales_crm.html
+++ b/docs/_site/tutorials/01_sales_crm.html
@@ -64,7 +64,7 @@
-
+
@@ -266,6 +266,12 @@
Anomaly Detection
+
+