Repository Issues
unit8co/darts
A python library for user-friendly forecasting and anomaly detection on time series.
Issues
Open
[BUG] Optuna-Integration no longer depends on pytorch-lightning
dependenciesgood first issuepr_welcome
3 comments0 reactions0 assignees
Open
Add `save()` and `load()` methods to`KMeansScorer`
feature requestgood first issue
6 comments0 reactions0 assignees
Open
[Feature] Improve handling & documentation of encoders and categorical static covariates in .fit_from_dataset()
documentationgood first issue
3 comments1 reaction0 assignees
Open
Improve error message when future covariates are missing during backtest() call
good first issueimprovement
4 comments0 reactions0 assignees
Open
TFTExplainer: getting explanations when trained on multiple time series
buggood first issue
5 comments0 reactions0 assignees
Open
Passing kwargs to the underlying fitting function
good first issueimprovementpr_welcome
5 comments1 reaction0 assignees
Open
mc_dropout with predict_likelihood_parameters
documentationgood first issue
3 comments0 reactions0 assignees
Open
[BUG] gridsearch with RegressionModel
buggood first issue
1 comment0 reactions0 assignees
Open
Dirichlet Likelihood TFT model
buggood first issue
6 comments0 reactions0 assignees
Open
Retrieve last hidden state for RNNModel and BlockRNNModel
good first issueimprovement
2 comments1 reaction0 assignees
Open
Gridsearch doesn't work with multiple timeseries
good first issueimprovement
3 comments0 reactions0 assignees
Open
Improve RNNmodel and blockRNN
good first issueimprovement
2 comments0 reactions0 assignees
Open
Allow historical forecast to work on time series with nan
good first issueimprovement
0 comments3 reactions0 assignees
Open
Write an `NBEATSExplainer` class for the interpretable version
good first issueimprovement
1 comment0 reactions0 assignees
Open
[BUG] Implementation errors on `TransformerModel`.
buggood first issue
0 comments0 reactions0 assignees
Open
NegativeBinomialLikelihood - Invalid values during training
buggood first issue
0 comments0 reactions0 assignees
Open
[New models - placeholder] More new models
good first issuenew model
0 comments0 reactions0 assignees
Open
[New model] DeepVAR
good first issuenew model
0 comments2 reactions0 assignees
Open
Future covariates for exponential smoothing
good first issueimprovement
4 comments0 reactions0 assignees
Open
Test N-BEATS with HorizonBasedDataset and reproduce paper's results
good first issueimprovement
0 comments0 reactions0 assignees