diff --git a/nbs/docs/capabilities/cross_validation.ipynb b/nbs/docs/capabilities/cross_validation.ipynb index 3cc664be6..eae8c8c56 100644 --- a/nbs/docs/capabilities/cross_validation.ipynb +++ b/nbs/docs/capabilities/cross_validation.ipynb @@ -78,7 +78,7 @@ "source": [ "## 2. Read the data\n", "\n", - "For this tutorial, we use part of the hourly M4 dataset. It is stored in a parquet file for efficiency. You can use ordinary pandas operations to read your data in other formats likes `.csv`. \n", + "For this tutorial, we use part of the hourly M4 dataset. It is stored in a parquet file for efficiency. However, you can use ordinary pandas operations to read your data in other formats likes `.csv`. \n", "\n", "The input to `NeuralForecast` is always a data frame in [long format](https://www.theanalysisfactor.com/wide-and-long-data/) with three columns: `unique_id`, `ds` and `y`:\n", "\n", @@ -180,7 +180,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "For simplicity, we use only a single series to explore in detail the cross-validation functionality. Also, let's use the first 700 time steps, such that we work with round numbers, making it easier to visualize and understand cross-validation." + "For simplicity, we focus on a single time series to explore the cross-validation functionality in detail. We also use only the first 700 time steps, which allows us to work with round numbers and makes the cross-validation process easier to visualize and understand." ] }, { @@ -449,7 +449,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "In the figure above, we see that we have 4 cutoff points, which correspond to our four cross-validation windows. Of course, notice that the windows are set from the end of the dataset. That way, the model trains on past data to predict future data. \n", + "In the figure above, we observe four cutoff points, each corresponding to a cross-validation window. Note that these windows are defined from the end of the dataset, ensuring that the model is trained on past data to predict future data.\n", "\n", ":::{.callout-warning collapse=\"true\"}\n", "## Important note\n", @@ -655,11 +655,17 @@ "metadata": {}, "source": [ "In the figure above, we see that our two folds overlap between time steps 601 and 650, since the step size is 50. This happens because:\n", + "\n", "- fold 1: model is trained using time steps 0 to 550 and predicts 551 to 650 (h=100)\n", "- fold 2: model is trained using time steps 0 to 600 (`step_size=50`) and predicts 601 to 700\n", "\n", "Be aware that when evaluating a model trained with overlapping cross-validation windows, some time steps have more than one prediction. This may bias your evaluation metric, as the repeated time steps are taken into account in the metric multiple times." ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [] } ], "metadata": { diff --git a/nbs/docs/capabilities/exogenous_variables.ipynb b/nbs/docs/capabilities/exogenous_variables.ipynb index 5b9d1667f..8e1b45e23 100644 --- a/nbs/docs/capabilities/exogenous_variables.ipynb +++ b/nbs/docs/capabilities/exogenous_variables.ipynb @@ -24,7 +24,7 @@ "All NeuralForecast models are capable of incorporating exogenous variables to model the following conditional predictive distribution:\n", "$$\\mathbb{P}(\\mathbf{y}_{t+1:t+H} \\;|\\; \\mathbf{y}_{[:t]},\\; \\mathbf{x}^{(h)}_{[:t]},\\; \\mathbf{x}^{(f)}_{[:t+H]},\\; \\mathbf{x}^{(s)} )$$\n", "\n", - "where the regressors are static exogenous $\\mathbf{x}^{(s)}$, historic exogenous $\\mathbf{x}^{(h)}_{[:t]}$, exogenous available at the time of the prediction $\\mathbf{x}^{(f)}_{[:t+H]}$ and autorregresive features $\\mathbf{y}_{[:t]}$. Depending on the [train loss](../../losses.pytorch), the model outputs can be point forecasts (location estimators) or uncertainty intervals (quantiles)." + "where the regressors are static exogenous $\\mathbf{x}^{(s)}$, historic exogenous $\\mathbf{x}^{(h)}_{[:t]}$, exogenous available at the time of the prediction $\\mathbf{x}^{(f)}_{[:t+H]}$ and autoregressive features $\\mathbf{y}_{[:t]}$. Depending on the [train loss](../../losses.pytorch), the model outputs can be point forecasts (location estimators) or uncertainty intervals (quantiles)." ] }, { diff --git a/nbs/docs/getting-started/datarequirements.ipynb b/nbs/docs/getting-started/datarequirements.ipynb index 50eed8e40..07edd831b 100644 --- a/nbs/docs/getting-started/datarequirements.ipynb +++ b/nbs/docs/getting-started/datarequirements.ipynb @@ -6,7 +6,7 @@ "metadata": {}, "source": [ "# Data Requirements\n", - "> Dataset input requirments" + "> Dataset input requirements" ] }, { @@ -352,7 +352,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "In this example `Y_df` only contains two columns: `timestamp`, and `value`. To use `NeuralForecast` we have to include the `unique_id` column and rename the previuos ones." + "In this example `Y_df` only contains two columns: `timestamp`, and `value`. To use `NeuralForecast` we have to include the `unique_id` column and rename the previous ones." ] }, { diff --git a/nbs/docs/getting-started/installation.ipynb b/nbs/docs/getting-started/installation.ipynb index 77a90f64f..dc7f94c5a 100644 --- a/nbs/docs/getting-started/installation.ipynb +++ b/nbs/docs/getting-started/installation.ipynb @@ -47,7 +47,7 @@ "* distributed training with spark: `pip install neuralforecast[spark]`\n", "* saving and loading from S3: `pip install neuralforecast[aws]`\n", "\n", - "#### User our env (optional)\n", + "#### Use our env (optional)\n", "\n", "If you don't have a Conda environment and need tools like Numba, Pandas, NumPy, Jupyter, Tune, and Nbdev you can use ours by following these steps:\n", "\n", diff --git a/nbs/docs/getting-started/quickstart.ipynb b/nbs/docs/getting-started/quickstart.ipynb index 7a336f908..4ecfad024 100644 --- a/nbs/docs/getting-started/quickstart.ipynb +++ b/nbs/docs/getting-started/quickstart.ipynb @@ -373,7 +373,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "Finally, we plot the forecasts of both models againts the real values." + "Finally, we plot the forecasts of both models against the real values." ] }, { diff --git a/nbs/docs/tutorials/adding_models.ipynb b/nbs/docs/tutorials/adding_models.ipynb index b0dc09b60..af3a47971 100644 --- a/nbs/docs/tutorials/adding_models.ipynb +++ b/nbs/docs/tutorials/adding_models.ipynb @@ -22,7 +22,7 @@ "\n", "We highly recommend reading first the Getting Started and the NeuralForecast Map tutorials!\n", "\n", - "Additionally, refer to the [CONTRIBUTING guide](https://github.com/Nixtla/neuralforecast/blob/main/CONTRIBUTING.md) for the basics how to contribute to NeuralForecast.\n", + "Additionally, refer to the [CONTRIBUTING guide](https://github.com/Nixtla/neuralforecast/blob/main/CONTRIBUTING.md) for the basics of how to contribute to NeuralForecast.\n", "\n", ":::" ] diff --git a/nbs/docs/tutorials/longhorizon_nhits.ipynb b/nbs/docs/tutorials/longhorizon_nhits.ipynb index 085970d3f..36c87bec0 100644 --- a/nbs/docs/tutorials/longhorizon_nhits.ipynb +++ b/nbs/docs/tutorials/longhorizon_nhits.ipynb @@ -61,7 +61,7 @@ "\n", "It return three Dataframes: `Y_df` contains the values for the target variables, `X_df` contains exogenous calendar features and `S_df` contains static features for each time-series (none for ETTm2). For this example we will only use `Y_df`.\n", "\n", - "If you want to use your own data just replace `Y_df`. Be sure to use a long format and have a simmilar structure than our data set." + "If you want to use your own data just replace `Y_df`. Be sure to use a long format and have a similar structure to our data set." ] }, { diff --git a/nbs/docs/tutorials/longhorizon_probabilistic.ipynb b/nbs/docs/tutorials/longhorizon_probabilistic.ipynb index 5ea12666a..dd06ff471 100644 --- a/nbs/docs/tutorials/longhorizon_probabilistic.ipynb +++ b/nbs/docs/tutorials/longhorizon_probabilistic.ipynb @@ -60,7 +60,7 @@ "\n", "It return three Dataframes: `Y_df` contains the values for the target variables, `X_df` contains exogenous calendar features and `S_df` contains static features for each time-series (none for ETTm2). For this example we will only use `Y_df`.\n", "\n", - "If you want to use your own data just replace `Y_df`. Be sure to use a long format and have a simmilar structure than our data set." + "If you want to use your own data just replace `Y_df`. Be sure to use a long format and have a similar structure to our data set." ] }, { diff --git a/nbs/docs/tutorials/longhorizon_transformers.ipynb b/nbs/docs/tutorials/longhorizon_transformers.ipynb index 63968d825..1b95cf63a 100644 --- a/nbs/docs/tutorials/longhorizon_transformers.ipynb +++ b/nbs/docs/tutorials/longhorizon_transformers.ipynb @@ -69,7 +69,7 @@ "\n", "It return three Dataframes: `Y_df` contains the values for the target variables, `X_df` contains exogenous calendar features and `S_df` contains static features for each time-series (none for ETTm2). For this example we will only use `Y_df`.\n", "\n", - "If you want to use your own data just replace `Y_df`. Be sure to use a long format and have a simmilar structure than our data set." + "If you want to use your own data just replace `Y_df`. Be sure to use a long format and have a similar structure to our data set." ] }, { diff --git a/nbs/docs/tutorials/robust_forecasting.ipynb b/nbs/docs/tutorials/robust_forecasting.ipynb index 1aa683dd6..28fcbbd47 100644 --- a/nbs/docs/tutorials/robust_forecasting.ipynb +++ b/nbs/docs/tutorials/robust_forecasting.ipynb @@ -687,7 +687,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "Finally, we plot the forecasts of both models againts the real values.\n", + "Finally, we plot the forecasts of both models against the real values.\n", "\n", "And evaluate the accuracy of the `NHITS-Huber` and `NHITS-Normal` forecasters." ] diff --git a/nbs/docs/tutorials/temporal_classification.ipynb b/nbs/docs/tutorials/temporal_classification.ipynb index d5c200b6a..b1fb41cd7 100644 --- a/nbs/docs/tutorials/temporal_classification.ipynb +++ b/nbs/docs/tutorials/temporal_classification.ipynb @@ -15,7 +15,7 @@ "source": [ "A logistic regression analyzes the relationship between a binary target variable and its predictor variables to estimate the probability of the dependent variable taking the value 1. In the presence of temporal data where observations along time aren't independent, the errors of the model will be correlated through time and incorporating autoregressive features or lags can capture temporal dependencies and enhance the predictive power of logistic regression.\n", "\n", - "
NHITS's inputs are static exogenous $\\mathbf{x}^{(s)}$, historic exogenous $\\mathbf{x}^{(h)}_{[:t]}$, exogenous available at the time of the prediction $\\mathbf{x}^{(f)}_{[:t+H]}$ and autorregresive features $\\mathbf{y}_{[:t]}$, each of these inputs is further decomposed into categorical and continuous. The network uses a multi-quantile regression to model the following conditional probability:$$\\mathbb{P}(\\mathbf{y}_{[t+1:t+H]}|\\;\\mathbf{y}_{[:t]},\\; \\mathbf{x}^{(h)}_{[:t]},\\; \\mathbf{x}^{(f)}_{[:t+H]},\\; \\mathbf{x}^{(s)})$$\n", + "
NHITS's inputs are static exogenous $\\mathbf{x}^{(s)}$, historic exogenous $\\mathbf{x}^{(h)}_{[:t]}$, exogenous available at the time of the prediction $\\mathbf{x}^{(f)}_{[:t+H]}$ and autoregressive features $\\mathbf{y}_{[:t]}$, each of these inputs is further decomposed into categorical and continuous. The network uses a multi-quantile regression to model the following conditional probability:$$\\mathbb{P}(\\mathbf{y}_{[t+1:t+H]}|\\;\\mathbf{y}_{[:t]},\\; \\mathbf{x}^{(h)}_{[:t]},\\; \\mathbf{x}^{(f)}_{[:t+H]},\\; \\mathbf{x}^{(s)})$$\n", "\n", "In this notebook we show how to fit NeuralForecast methods for binary sequences regression. We will:\n", "- Installing NeuralForecast.\n", @@ -975,7 +975,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "Finally, we plot the forecasts of both models againts the real values.\n", + "Finally, we plot the forecasts of both models against the real values.\n", "And evaluate the accuracy of the `MLP` and `NHITS` temporal classifiers." ] }, diff --git a/nbs/docs/tutorials/transfer_learning.ipynb b/nbs/docs/tutorials/transfer_learning.ipynb index f2ef6ab06..a28b998d6 100644 --- a/nbs/docs/tutorials/transfer_learning.ipynb +++ b/nbs/docs/tutorials/transfer_learning.ipynb @@ -14,7 +14,7 @@ "source": [ "Transfer learning refers to the process of pre-training a flexible model on a large dataset and using it later on other data with little to no training. It is one of the most outstanding 🚀 achievements in Machine Learning 🧠 and has many practical applications.\n", "\n", - "For time series forecasting, the technique allows you to get lightning-fast predictions ⚡ bypassing the tradeoff between accuracy and speed (more than 30 times faster than our alreadsy fast [autoARIMA](https://github.com/Nixtla/statsforecast) for a similar accuracy).\n", + "For time series forecasting, the technique allows you to get lightning-fast predictions ⚡ bypassing the tradeoff between accuracy and speed (more than 30 times faster than our already fast [autoARIMA](https://github.com/Nixtla/statsforecast) for a similar accuracy).\n", "\n", "This notebook shows how to generate a pre-trained model and store it in a checkpoint to make it available to forecast new time series never seen by the model. \n", "\n", @@ -84,7 +84,7 @@ "cell_type": "markdown", "metadata": {}, "source": [ - "This example will automatically run on GPUs if available. **Make sure** cuda is available. (If you need help to put this into production send us an email or join or community, we also offer a fully hosted solution)" + "This example will automatically run on GPUs if available. **Make sure** cuda is available. (If you need help to put this into production send us an email or join our community, we also offer a fully hosted solution)" ] }, { @@ -117,7 +117,7 @@ "\n", "It return three Dataframes: `Y_df` contains the values for the target variables, `X_df` contains exogenous calendar features and `S_df` contains static features for each time-series (none for M4). For this example we will only use `Y_df`.\n", "\n", - "If you want to use your own data just replace `Y_df`. Be sure to use a long format and have a simmilar structure than our data set." + "If you want to use your own data just replace `Y_df`. Be sure to use a long format and have a similar structure to our data set." ] }, { diff --git a/neuralforecast/core.py b/neuralforecast/core.py index 1d4aadc8d..a03b7b71c 100644 --- a/neuralforecast/core.py +++ b/neuralforecast/core.py @@ -225,7 +225,7 @@ def __init__( local_static_scaler_type: Optional[str] = None, ): """The `core.StatsForecast` class allows you to efficiently fit multiple `NeuralForecast` models - for large sets of time series. It operates with pandas DataFrame `df` that identifies series + for large sets of time series. It operates with a pandas DataFrame `df` that identifies series and datestamps with the `unique_id` and `ds` columns. The `y` column denotes the target time series variable. @@ -234,9 +234,9 @@ def __init__( see [collection here](./models). freq (str or int): Frequency of the data. Must be a valid pandas or polars offset alias, or an integer. local_scaler_type (str, optional): Scaler to apply per-serie to temporal features before fitting, which is inverted after predicting. - Can be 'standard', 'robust', 'robust-iqr', 'minmax' or 'boxcox'. Defaults to None. + Can be 'standard', 'robust', 'robust-iqr', 'minmax' or 'boxcox'. local_static_scaler_type (str, optional): Scaler to apply to static exogenous features before fitting. - Can be 'standard', 'robust', 'robust-iqr', 'minmax' or 'boxcox'. Defaults to None. + Can be 'standard', 'robust', 'robust-iqr', 'minmax' or 'boxcox'. Returns: NeuralForecast: Returns instantiated `NeuralForecast` class. @@ -449,23 +449,23 @@ def fit( distributed_config: Optional[DistributedConfig] = None, prediction_intervals: Optional[PredictionIntervals] = None, ) -> None: - """Fit the core.NeuralForecast. + """Fit the core.NeuralForecast - Fit `models` to a large set of time series from DataFrame `df`. + Fit `models` to a large set of time series from DataFrame `df` and store fitted models for later inspection. Args: df (pandas, polars or spark DataFrame, or a list of parquet files containing the series, optional): DataFrame with columns [`unique_id`, `ds`, `y`] and exogenous variables. - If None, a previously stored dataset is required. Defaults to None. - static_df (pandas, polars or spark DataFrame, optional): DataFrame with columns [`unique_id`] and static exogenous. Defaults to None. - val_size (int, optional): Size of validation set. Defaults to 0. - use_init_models (bool, optional): Use initial model passed when NeuralForecast object was instantiated. Defaults to False. - verbose (bool): Print processing steps. Defaults to False. - id_col (str): Column that identifies each serie. Defaults to 'unique_id'. - time_col (str): Column that identifies each timestep, its values can be timestamps or integers. Defaults to 'ds'. - target_col (str): Column that contains the target. Defaults to 'y'. + If None, a previously stored dataset is required. + static_df (pandas, polars or spark DataFrame, optional): DataFrame with columns [`unique_id`] and static exogenous. + val_size (int, optional): Size of validation set. + use_init_models (bool, optional): Use initial model passed when NeuralForecast object was instantiated. + verbose (bool): Print processing steps. + id_col (str): Column that identifies each serie. + time_col (str): Column that identifies each timestep, its values can be timestamps or integers. + target_col (str): Column that contains the target. distributed_config (neuralforecast.DistributedConfig): Configuration to use for DDP training. Currently only spark is supported. - prediction_intervals (PredictionIntervals, optional): Configuration to calibrate prediction intervals (Conformal Prediction). Defaults to None. + prediction_intervals (PredictionIntervals, optional): Configuration to calibrate prediction intervals (Conformal Prediction). Returns: NeuralForecast: Returns `NeuralForecast` class with fitted `models`. @@ -580,7 +580,7 @@ def make_future_dataframe( Args: df (pandas or polars DataFrame, optional): DataFrame with columns [`unique_id`, `ds`, `y`] and exogenous variables. - Only required if this is different than the one used in the fit step. Defaults to None. + Only required if this is different than the one used in the fit step. """ if not self._fitted: raise Exception("You must fit the model first.") @@ -821,14 +821,14 @@ def predict( Args: df (pandas, polars or spark DataFrame, optional): DataFrame with columns [`unique_id`, `ds`, `y`] and exogenous variables. - If a DataFrame is passed, it is used to generate forecasts. Defaults to None. - static_df (pandas, polars or spark DataFrame, optional): DataFrame with columns [`unique_id`] and static exogenous. Defaults to None. - futr_df (pandas, polars or spark DataFrame, optional): DataFrame with [`unique_id`, `ds`] columns and `df`'s future exogenous. Defaults to None. - verbose (bool): Print processing steps. Defaults to False. + If a DataFrame is passed, it is used to generate forecasts. + static_df (pandas, polars or spark DataFrame, optional): DataFrame with columns [`unique_id`] and static exogenous. + futr_df (pandas, polars or spark DataFrame, optional): DataFrame with [`unique_id`, `ds`] columns and `df`'s future exogenous. + verbose (bool): Print processing steps. engine (spark session): Distributed engine for inference. Only used if df is a spark dataframe or if fit was called on a spark dataframe. - level (list of ints or floats, optional): Confidence levels between 0 and 100. Defaults to None. - quantiles (list of floats, optional): Alternative to level, target quantiles to predict. Defaults to None. - h (int, optional): Forecasting horizon. If None, uses the horizon of the fitted models. Defaults to None. + level (list of ints or floats, optional): Confidence levels between 0 and 100. + quantiles (list of floats, optional): Alternative to level, target quantiles to predict. + h (int, optional): Forecasting horizon. If None, uses the horizon of the fitted models. data_kwargs (kwargs): Extra arguments to be passed to the dataset within each model. Returns: @@ -1031,14 +1031,14 @@ def explain( outputs (list of int, optional): List of outputs to explain for models with multiple outputs. Defaults to [0] (first output). explainer (str): Name of the explainer to use. Options are 'IntegratedGradients', 'ShapleyValueSampling', 'InputXGradient'. Defaults to 'IntegratedGradients'. df (pandas, polars or spark DataFrame, optional): DataFrame with columns [`unique_id`, `ds`, `y`] and exogenous variables. - If a DataFrame is passed, it is used to generate forecasts. Defaults to None. - static_df (pandas, polars or spark DataFrame, optional): DataFrame with columns [`unique_id`] and static exogenous. Defaults to None. - futr_df (pandas, polars or spark DataFrame, optional): DataFrame with [`unique_id`, `ds`] columns and `df`'s future exogenous. Defaults to None. + If a DataFrame is passed, it is used to generate forecasts. + static_df (pandas, polars or spark DataFrame, optional): DataFrame with columns [`unique_id`] and static exogenous. + futr_df (pandas, polars or spark DataFrame, optional): DataFrame with [`unique_id`, `ds`] columns and `df`'s future exogenous. h (int): The forecast horizon. Can be larger than the horizon set during training. - verbose (bool): Print processing steps. Defaults to False. + verbose (bool): Print processing steps. engine (spark session): Distributed engine for inference. Only used if df is a spark dataframe or if fit was called on a spark dataframe. - level (list of ints or floats, optional): Confidence levels between 0 and 100. Defaults to None. - quantiles (list of floats, optional): Alternative to level, target quantiles to predict. Defaults to None. + level (list of ints or floats, optional): Confidence levels between 0 and 100. + quantiles (list of floats, optional): Alternative to level, target quantiles to predict. data_kwargs (kwargs): Extra arguments to be passed to the dataset within each model. Returns: @@ -1360,24 +1360,24 @@ def cross_validation( Args: df (pandas or polars DataFrame, optional): DataFrame with columns [`unique_id`, `ds`, `y`] and exogenous variables. - If None, a previously stored dataset is required. Defaults to None. + If None, a previously stored dataset is required. static_df (pandas or polars DataFrame, optional): DataFrame with columns [`unique_id`] and static exogenous. Defaults to None. - n_windows (int): Number of windows used for cross validation. Defaults to 1. - step_size (int): Step size between each window. Defaults to 1. + n_windows (int): Number of windows used for cross validation. + step_size (int): Step size between each window. val_size (int, optional): Length of validation size. If passed, set `n_windows=None`. Defaults to 0. - test_size (int, optional): Length of test size. If passed, set `n_windows=None`. Defaults to None. - use_init_models (bool, optional): Use initial model passed when object was instantiated. Defaults to False. - verbose (bool): Print processing steps. Defaults to False. + test_size (int, optional): Length of test size. If passed, set `n_windows=None`. + use_init_models (bool, optional): Use initial model passed when object was instantiated. + verbose (bool): Print processing steps. refit (bool or int): Retrain model for each cross validation window. If False, the models are trained at the beginning and then used to predict each window. - If positive int, the models are retrained every `refit` windows. Defaults to False. - id_col (str): Column that identifies each serie. Defaults to 'unique_id'. + If positive int, the models are retrained every `refit` windows. + id_col (str): Column that identifies each serie. time_col (str): Column that identifies each timestep, its values can be timestamps or integers. Defaults to 'ds'. - target_col (str): Column that contains the target. Defaults to 'y'. + target_col (str): Column that contains the target. prediction_intervals (PredictionIntervals, optional): Configuration to calibrate prediction intervals (Conformal Prediction). Defaults to None. - level (list of ints or floats, optional): Confidence levels between 0 and 100. Defaults to None. - quantiles (list of floats, optional): Alternative to level, target quantiles to predict. Defaults to None. - h (int, optional): Forecasting horizon. If None, uses the horizon of the fitted models. Defaults to None. + level (list of ints or floats, optional): Confidence levels between 0 and 100. + quantiles (list of floats, optional): Alternative to level, target quantiles to predict. + h (int, optional): Forecasting horizon. If None, uses the horizon of the fitted models. data_kwargs (kwargs): Extra arguments to be passed to the dataset within each model. Returns: @@ -1533,9 +1533,9 @@ def predict_insample( to predict historic values of a time series from the stored dataframe. Args: - step_size (int): Step size between each window. Defaults to 1. - level (list of ints or floats, optional): Confidence levels between 0 and 100. Defaults to None. - quantiles (list of floats, optional): Alternative to level, target quantiles to predict. Defaults to None. + step_size (int): Step size between each window. + level (list of ints or floats, optional): Confidence levels between 0 and 100. + quantiles (list of floats, optional): Alternative to level, target quantiles to predict. Returns: fcsts_df (pandas.DataFrame): DataFrame with insample predictions for all fitted `models`. @@ -1705,9 +1705,9 @@ def save( Args: path (str): Directory to save current status. - model_index (list, optional): List to specify which models from list of self.models to save. Defaults to None. - save_dataset (bool): Whether to save dataset or not. Defaults to True. - overwrite (bool): Whether to overwrite files or not. Defaults to False. + model_index (list, optional): List to specify which models from list of self.models to save. + save_dataset (bool): Whether to save dataset or not. + overwrite (bool): Whether to overwrite files or not. """ # Standarize path without '/' if path[-1] == "/":