Nixtla / mlforecast

Scalable machine 🤖 learning for time series forecasting.

https://nixtlaverse.nixtla.io/mlforecast

Nixtla/mlforecast Issues

[core] speed up date features calculation
Closed 19 days ago
Electricity load tutorial problem
Updated a month ago1
Found missing inputs in X_df. It should have one row per id and time for the complete forecasting horizon.
Closed a month ago12
[Custom Training] Add custom training for Cross Validation
Updated a month ago
Not enough models trained in cross_validation with fitted=True and horizon > 9
Closed a month ago
[MLForecast] Add the possibility to pass custom parameters to the fit function
Updated a month ago5
Cross validation with prediction_intervals and in-sample predictions enabled lacks folds
Closed a month ago
MLForecast LinearRegression Isn't Applied to Each Unique Id Time Series Seperately
Closed 2 months ago1
MLForecast - Support historic covariates out of the box
Closed 2 months ago3
get performance on training set
Closed 2 months ago2
[Core] getting an error module 'coreforecast.lag_transforms' has no attribute 'BaseLagTransform'
Closed 3 months ago3
Mlforecast + AutoDifferences + fitted=True
Updated 2 months ago2
Forecasting produces nearly horizontal results
Closed 3 months ago4
[distributed]: allow for .ts.update in DistributedMLForecast
Closed 3 months ago3
Unbale to do LogTransformation using target_transformation
Closed 3 months ago2
predict() and cross_validation() outputs inconsistency
Closed 3 months ago6
[Model] Distributed version of the model giving Arrow Capacity error
Closed 3 months ago6
mlforecast: can the number of exogenous variables be different for different unique_id?
Closed 3 months ago2
[Model] Distributed and Non-distributed version of the models giving different result
Closed 3 months ago8
[date features]: dayofweek_cat - day of week as a one hot encoding feature
Updated 3 months ago4
Lag feature: how initial values are treated or populated once the data has been shifted?
Closed 3 months ago2
AutoDifferences, AutoSeasonalityAndDifferences result in "AttributeError: 'NoneType' object has no attribute 'AutoDifferences'"
Closed 3 months ago2
Does mlforecast train a single global model or one model per serie?
Closed 3 months ago2
[MLForecast: test set eval and early stopping ]
Closed 3 months ago2
[Core] Saving of the model
Closed 4 months ago20
Mutli Step Forecast with Horizon Column Indicator
Updated 4 months ago3
Fcst.predict does not accept X_df with dynamic exogenous variables
Closed 4 months ago2
model.predict encounters an error
Closed 4 months ago3
MLforecast does not work with with PyArrow dates
Closed 4 months ago2
[forecast]: allow for "batched" forecasting
Updated 4 months ago2
Dynamic features for training
Closed 4 months ago1
[Core]`ts.update` method may not support target_transforms,lag_transforms and date_features
Closed 5 months ago1
[core]Using multiple models can cause `new_x` lag feature shift
Closed 5 months ago2
[MLForecast] lag_transforms with different features packages
Closed 5 months ago6
MLForecast: Core: Add prediction intervals to forecast_fitted_values
Closed 5 months ago
[Distributed] Adding external variable for distributed
Closed 5 months ago10
[Core] Support naming features
Closed 6 months ago
Do not re-calculate lags/transforms if the df already has them
Closed 6 months ago3
[Model] Getting issue with instantiating SparkLGBMForecast
Closed 6 months ago6
[utils] Move `backtest_splits` to `utilsforecast`
Closed 6 months ago
In predict Function, Found missing inputs in X_df. It should have one row per id and date for the complete forecasting horizon
Closed 7 months ago13
[ETC] No module name mlforecast.feature_engineering
Closed 7 months ago2
[Core] Inconsistent, and overlapping forecasting time steps generated, as a result of predict()
Closed 7 months ago2
[Forecast] Forecast Fitted Values Failing with NA
Closed 7 months ago
cross_validation of DistributedMLForecast not working when n_windows > 2
Closed 7 months ago
Better support for cross-validation on series with varying start times
Closed 7 months ago
In predict, ValueError: You are trying to merge on float64 and object columns. If you wish to proceed you should use pd.concat
Closed 7 months ago3
[Utils] Add "packaging" to the package dependencies
Closed 8 months ago
error in StatsForecast.plot
Closed 8 months ago2
[Core] keep_last_n has no effect in MLForecast.preprocess
Closed 8 months ago2