Papers › N-BEATS: Neural basis expansion analysis for interpretable time series forecasting

N-BEATS: Neural basis expansion analysis for interpretable time series forecasting

24 May 2019ICLR 2020 1arXiv:1905.10437archive 2025-07-28

Boris N. Oreshkin, Dmitri Carpov, Nicolas Chapados, Yoshua Bengio

We focus on solving the univariate times series point forecasting problem using deep learning. We propose a deep neural architecture based on backward and forward residual links and a very deep stack of fully-connected layers. The architecture has a number of desirable properties, being interpretable, applicable without modification to a wide array of target domains, and fast to train. We test the proposed architecture on several well-known datasets, including M3, M4 and TOURISM competition datasets containing time series from diverse domains. We demonstrate state-of-the-art performance for two configurations of N-BEATS for all the datasets, improving forecast accuracy by 11% over a statistical benchmark and by 3% over last year's winner of the M4 competition, a domain-adjusted hand-crafted hybrid between neural network and statistical time series models. The first configuration of our model does not employ any time-series-specific components and its performance on heterogeneous datasets strongly suggests that, contrarily to received wisdom, deep learning primitives such as residual blocks are by themselves sufficient to solve a wide range of forecasting problems. Finally, we demonstrate how the proposed architecture can be augmented to provide outputs that are interpretable without considerable loss in accuracy.

PaperPDFConference PDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 39 code samples harvested from 6 repositories linked to this paper: 34 ran (of which 0 constructed an object rather than computing a result; 29 with no instrument failure: 5 honoured, 0 violated, 24 with no contract checked; 5 where Syntology's instrument failed) · 5 unverified. Where Syntology's instrument failed: 5 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: community (archive-listed): 39 samples from 6 repositories, 34 ran (29 with no instrument failure; 5 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

No sample is linked to this paper from 4 of its repositories that Syntology harvested for other papers:

Levels written between 2026-09-22 and 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 1 of the 34 samples that ran has no recorded date.

18 repositories listed; official and paper-mentioned ones first.

Chasm4359/ProTS mentioned on GitHubpytorchBSD-3-ClauseSyntology: 10 samples, 8 ran (0 instrument-failed) · 2 unverified report
DFrolova/ml-sk-project mentioned on GitHubpytorchSyntology: not harvested report
ElementAI/N-BEATS mentioned on GitHubpytorchNOASSERTIONSyntology: not harvested report
MLD3/Deep-Residual-Time-Series-Forecasting mentioned on GitHubpytorchSyntology: 3 samples, 3 ran (3 instrument-failed) · 0 unverified report
Nixtla/neuralforecast mentioned on GitHubpytorchSyntology: 3 samples, 3 ran (0 instrument-failed) · 0 unverified report
Nixtla/nixtlats mentioned on GitHubSyntology: not harvested report
Y9008/NBEATS mentioned on GitHubpytorchMITSyntology: not harvested report
amitesh863/nbeats_forecast mentioned on GitHubpytorchMITSyntology: not harvested report
flaviagiammarino/nbeats-tensorflow mentioned on GitHubtfMITSyntology: 6 samples, 6 ran (0 instrument-failed) · 0 unverified report
jdb78/pytorch-forecasting mentioned on GitHubpytorchMITSyntology: not harvested report
philipperemy/n-beats mentioned on GitHubpytorchSyntology: 3 samples, 3 ran (2 instrument-failed) · 0 unverified report
potosnakw/neuralforecast mentioned on GitHubpytorchApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Deep Learning for Time Series Forecasting: Tutorial and Literature Survey (arXiv:2004.10240; that paper's own run record: 7 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 0 violated, 7 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. report
realdanielbyrne/N-BEATS-Lightning mentioned on GitHubpytorchMITSyntology: 14 samples, 11 ran (0 instrument-failed) · 3 unverified report
servicenow/n-beats mentioned on GitHubpytorchNOASSERTIONSyntology: not harvested report
unit8co/darts mentioned on GitHubpytorchApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Long-term Forecasting with TiDE: Time-series Dense Encoder (arXiv:2304.08424; that paper's own run record: 2 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified); nothing checks that they implement this paper's method. report
weifantt/depts mentioned on GitHubpytorchSyntology: not harvested report
etna-team/etna pytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Time-Series Representation Learning via Temporal and Contextual Contrasting (arXiv:2106.14112; that paper's own run record: 2 ran (of which 1 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. report
leejoonhun/fan-beats pytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Feature-aligned N-BEATS with Sinkhorn divergence (arXiv:2305.15196; that paper's own run record: 6 ran (of which 6 constructed an object rather than computing a result; 6 with no instrument failure: 0 honoured, 0 violated, 6 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified; every one of the 6 samples that ran constructed an object rather than computing a result); nothing checks that they implement this paper's method. report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

39 samples harvested: 34 ran (of which 0 constructed an object rather than computing a result; 29 with no instrument failure: 5 honoured, 0 violated, 24 with no contract checked; 5 where Syntology's instrument failed) · 5 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

5ran · honoured contract
5ran · our draft was wrong
24ran
5unverified

Licence: 3 of the 39 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 6 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

Chasm4359/ProTS community (archive-listed) 10 samples, 8 ran (0 instrument-failed) · 2 unverified
add_modified_external_features feature/feature_external_selection.py ran BSD-3-Clause (permissive) · 146b2387bd3f249b · report fixture: signature data:DataFrame/0/?, external_features:DataFrame/0/? · returned DataFrame · level recorded 2026-09-24
center_quantile_weighting evaluation/weighting.py ran BSD-3-Clause (permissive) · edba5939feafb629 · report fixture: signature quantiles:list/1/float · returned Series · level recorded 2026-09-24
lag_target feature/time_lag.py ran BSD-3-Clause (permissive) · 75e380a62ebd9b70 · report fixture: signature data:DataFrame/0/?, target:str/0/str, time_lags:list/1/int · returned DataFrame · level recorded 2026-09-24
left_tail_quantile_weighting evaluation/weighting.py ran BSD-3-Clause (permissive) · a8edd7770dc36a56 · report fixture: signature quantiles:list/1/float · returned Series · level recorded 2026-09-24
remove_lag_interval feature/time_lag.py ran BSD-3-Clause (permissive) · 1728441ddecfd61c · report fixture: signature data:DataFrame/0/?, horizon:int/0/int, target_lags:list/1/int, external_lags_by_name:dict/0/? · returned DataFrame · level recorded 2026-09-24
remove_original_external_columns feature/feature_external_selection.py ran BSD-3-Clause (permissive) · 8562140261c1d51a · report fixture: signature data:DataFrame/0/?, original_external_feature_names:list/1/str · returned DataFrame · level recorded 2026-09-24
rename_lag_series feature/time_lag.py ran BSD-3-Clause (permissive) · 19f916807ae1c6d6 · report fixture: signature column:Series/1/float, lag:int/0/int · returned Series · level recorded 2026-09-24
uniform_quantile_weighting evaluation/weighting.py ran BSD-3-Clause (permissive) · e1f3643c7325edfe · report fixture: signature quantiles:list/1/float · returned Series · level recorded 2026-09-24
add_datetime_features feature/time_categorical.py unverified BSD-3-Clause (permissive) · 63a5cbe82b3db20f · report
add_datetime_features_with_lag feature/time_categorical.py unverified BSD-3-Clause (permissive) · 87a5512b788503ab · report
MLD3/Deep-Residual-Time-Series-Forecasting community (archive-listed) 3 samples, 3 ran (3 instrument-failed) · 0 unverified
mse in 78 papers
census: 5 implementations from 6 papers, 4 distinct outputs across 3 buckets, one shared input per bucket
drtf.py ran · our draft was wrong fingerprinted no licence file found · pointer only · f0085a43c76debf6 · report fixture: signature output:list/1/Tensor, target:Tensor/2/float · returned Tensor · level recorded 2026-09-22
mse_one drtf.py ran · our draft was wrong fingerprinted no licence file found · pointer only · 5a4c78d16ecd7cf6 · report fixture: signature output:Tensor/2/float, target:Tensor/2/float · returned Tensor · level recorded 2026-09-22
msedoubs drtf.py ran · our draft was wrong no licence file found · pointer only · 32065b0b0a4d53b7 · report fixture: signature output:list/1/Tensor, target:list/1/Tensor · returned Tensor · level recorded 2026-09-22
Nixtla/neuralforecast community (archive-listed) 3 samples, 3 ran (0 instrument-failed) · 0 unverified
generate_changepoint_basis neuralforecast/models/nbeats.py ran · honoured contract fingerprinted Apache-2.0 (permissive) · a12334ffc582c940 · report fixture: signature length:int/0/int, n_basis:int/0/int · returned ndarray · level recorded 2026-09-22
generate_legendre_basis neuralforecast/models/nbeats.py ran · honoured contract fingerprinted Apache-2.0 (permissive) · 635c614da20b298a · report fixture: signature length:int/0/int, n_basis:int/0/int · returned ndarray · level recorded 2026-09-22
generate_polynomial_basis neuralforecast/models/nbeats.py ran · honoured contract fingerprinted Apache-2.0 (permissive) · e6ae83dce808c9fb · report fixture: signature length:int/0/int, n_basis:int/0/int · returned ndarray · level recorded 2026-09-22
flaviagiammarino/nbeats-tensorflow community (archive-listed) 6 samples, 6 ran (0 instrument-failed) · 0 unverified
cast_target_to_array nbeats_tensorflow/utils.py ran MIT (permissive) · 8bbd336c54d6b7cd · report fixture: signature y:list/1/float · returned ndarray · level recorded 2026-09-24
get_time_indices nbeats_tensorflow/utils.py ran MIT (permissive) · 1215fe3b290892b4 · report fixture: signature t:int/0/int, H:int/0/int · returned tuple · level recorded 2026-09-24
get_training_sequences nbeats_tensorflow/utils.py ran MIT (permissive) · 331a1c7dbefcc4b7 · report fixture: signature y:ndarray/1/float, t:int/0/int, H:int/0/int · returned tuple · level recorded 2026-09-24
seasonality_model nbeats_tensorflow/blocks.py ran MIT (permissive) · cab77a83d7c17130 · report fixture: signature theta:EagerTensor/2/float, t_:EagerTensor/1/float, p:int/0/int · returned EagerTensor · level recorded 2026-09-24
trend_block nbeats_tensorflow/blocks.py ran MIT (permissive) · 37f9452716df75df · report fixture: signature h:EagerTensor/2/float, p:int/0/int, t_b:EagerTensor/1/float, t_f:EagerTensor/1/float, share_theta:bool/0/bool · returned tuple · level recorded 2026-09-24
trend_model nbeats_tensorflow/blocks.py ran MIT (permissive) · 7f2c7f15af1134dc · report fixture: signature theta:EagerTensor/2/float, t_:EagerTensor/1/float, p:int/0/int · returned EagerTensor · level recorded 2026-09-24
philipperemy/n-beats community (archive-listed) 3 samples, 3 ran (2 instrument-failed) · 0 unverified
seasonality_model nbeats_pytorch/model.py ran · our draft was wrong MIT (permissive) · a80543442281cf38 · report fixture: signature thetas:Tensor/2/float, t:ndarray/1/float, device:device/0/? · returned Tensor · level recorded 2026-09-22
squeeze_last_dim nbeats_pytorch/model.py ran · honoured contract fingerprinted MIT (permissive) · 6b82ed8d296402a1 · report fixture: signature tensor:Tensor/3/float · returned Tensor · level recorded 2026-09-22
trend_model nbeats_pytorch/model.py ran · our draft was wrong MIT (permissive) · 48c2e955f3036f30 · report fixture: signature thetas:Tensor/2/float, t:ndarray/1/float, device:device/0/? · returned Tensor · level recorded 2026-09-22
realdanielbyrne/N-BEATS-Lightning community (archive-listed) 14 samples, 11 ran (0 instrument-failed) · 3 unverified
apply_protocol_training_fallbacks experiments/run_from_yaml.py ran MIT (permissive) · 5bb518a4df00ea2f · report fixture: signature training:dict/0/?, protocol:dict/0/?, explicit_training:dict/0/? · returned dict · level recorded 2026-09-24
build_training_plan experiments/run_ensemble_experiments.py ran MIT (permissive) · fe6e935bdca93fca · report fixture: signature ensemble_proposals:dict/0/?, periods:list/1/str, n_runs:int/0/int, multipliers:list/1/int, main_predictions_dir:str/0/str · returned tuple · level recorded 2026-09-24
count_parameters in 278 papers experiments/run_milk_sumloss_convergence.py ran · honoured contract MIT (permissive) · f6b944f50d3f15ae · report fixture: signature model:callable/0/? · returned int · level recorded, date not recorded
deep_merge in 9 papers experiments/run_from_yaml.py ran MIT (permissive) · e3a4e525486f1ca0 · report fixture: signature base:dict/0/?, override:dict/0/? · returned dict · level recorded 2026-09-24
find_existing_5h_prediction experiments/run_ensemble_experiments.py ran MIT (permissive) · 7cb61355d3f91b1a · report fixture: signature config_name:str/0/str, pass_name:str/0/str, period:str/0/str, run_idx:int/0/int, predictions_dir:str/0/str · returned str · level recorded 2026-09-24
generate_config in 5 papers experiments/generate_basis_dim_sweep.py ran MIT (permissive) · f056b22e9df589de · report fixture: signature base_config:dict/0/?, basis_dim:int/0/int · returned dict · level recorded 2026-09-24
generate_lg_vae_configs experiments/run_lg_vae_study.py ran MIT (permissive) · f3b131fc93467b98 · report fixture: signature dataset_name:str/0/str, round_num:int/0/int, promoted_configs:NoneType/0/? · returned dict · level recorded 2026-09-24
get_ensemble_npz_name experiments/run_ensemble_experiments.py ran MIT (permissive) · cfb69e4e7f450e79 · report fixture: signature pass_name:str/0/str, config_name:str/0/str, period:str/0/str, mult:int/0/int, run_idx:int/0/int · returned str · level recorded 2026-09-24
parse_pool_freq experiments/analyze_pooling_search.py ran MIT (permissive) · 04f0884e55ae1f54 · report fixture: signature name:str/0/str · returned tuple · level recorded 2026-09-24
rank_and_promote experiments/run_ae_trend_study.py ran MIT (permissive) · 486664f3dc245249 · report fixture: signature csv_path:str/0/str, round_num:int/0/int, keep_fraction:float/0/float, meta_forecaster:NoneType/0/?, top_k_override:NoneType/0/? · returned list · level recorded 2026-09-24
result_exists experiments/run_milk_sumloss_convergence.py ran MIT (permissive) · cbfbf22ba63c7382 · report fixture: signature csv_path:str/0/str, config_name:str/0/str, run_idx:int/0/int · returned bool · level recorded 2026-09-24
load_yaml_config in 18 papers experiments/run_from_yaml.py unverified MIT (permissive) · 1f8a0c5844dcb4e9 · report
rank_and_promote experiments/run_lg_vae_study.py unverified MIT (permissive) · 5b546d8444f809bb · report
write_yaml experiments/generate_comprehensive_sweep.py unverified MIT (permissive) · 01e92d90f57fa4f6 · report

Tasks

Time SeriesTime Series AnalysisTime Series ForecastingTime-Series Few-Shot Learning with Heterogeneous ChannelsUnivariate Time Series Forecasting

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections