Salesfev / AutoGluon ↗Rossmann Store Daily & Weekly Sales
Benchmarking weekly revenue forecasting across 1,115 German Rossmann drugstore locations over a 13-week quarter-ahead planning horizon. Enables automated inventory replenishment, staffing allocation, and promotional impact quantification across heterogeneous store formats.
Business Objective: Rossmann Store Daily & Weekly Sales
01Business Context
Retail pharmacy and drugstore operations require medium-term quarter-ahead (13-week) demand forecasts to synchronize warehouse logistics, supplier purchase orders, and regional store staffing. Demand fluctuates substantially based on recurring bi-weekly promotions, state-level public holidays, school vacation schedules, and localized competitor openings.
02Analytical Objective
Generate accurate weekly net store sales forecasts across 1,115 retail locations for the upcoming 13 weeks (horizon H=13) leveraging historical sales, foot traffic, promotional schedules, and static store metadata.
03Economic & Decision Impact
Under-forecasting leads to stockouts of high-velocity fast-moving consumer goods (FMCG), lost turnover, and acute under-staffing during promotional surges. Over-forecasting ties up working capital in perishable inventory, increases holding costs, and drives unnecessary distribution center transfers.
Classical univariate baselines such as Seasonal Naive and AutoARIMA struggle to capture cross-store promotional lifts and holiday shifts, serving primarily as baseline sanity checks. Gradient boosted decision trees (LightGBM) with engineered lag and rolling window features historically set strong benchmarks on this dataset, while zero-shot time series foundation models (Mistral's Timestral, TimesFM, MOIRAI) demonstrate competitive zero-shot cross-series transfer without extensive store-level feature engineering.
Target & Feature Column Definitions
Exhaustive business definitions, measurement units, roles, and target variables across all 18 columns.
| Col | Variable / Feature Name | Role | Type | Unit / Scale | Description & Business Meaning |
|---|---|---|---|---|---|
| C | SalesWeekly Sales | TARGET | NUM | EUR | Total gross weekly sales revenue generated by the store. |
| A | idSeries Identifier | SERIES ID | CAT | — | Unique identifier assigned to each individual retail store time series. |
| B | timestampWeek Timestamp | TIME | TIME | — | The weekly aggregation end-date timestamp (Sunday) marking the observation period. |
| D | OpenOpen Days Ratio | Feature | NUM | ratio | Fraction of days within the week that the store was open for customer business (0.0 to 1.0). |
| E | PromoPromotion Days Ratio | Feature | NUM | ratio | Fraction of operating days during the week featuring an active promotional discount campaign. |
| F | SchoolHolidaySchool Holiday Days | Feature | NUM | count | Number of days in the week affected by regional public school closures and vacations. |
| G | StateHolidayState Holiday Days | Feature | NUM | count | Number of days in the week designated as official public state holidays. |
| H | CustomersCustomer Footfall | Feature | NUM | count | Total observed weekly customer footfall; available historically but unknown in the future horizon. |
| I | AssortmentAssortment Level | Feature | CAT | — | Product assortment depth tier offered by the location (a = basic, b = extra, c = extended). |
| J | CompetitionDistanceCompetition Distance | Feature | NUM | m | Radial distance in meters to the nearest direct competitor drugstore store front. |
| K | CompetitionOpenSinceMonthCompetition Opening Month | Feature | NUM | month | Calendar month (1-12) during which the closest competitor began local operations. |
| L | CompetitionOpenSinceYearCompetition Opening Year | Feature | NUM | year | Calendar year in which the closest competitor store opened. |
| M | Promo2Continuous Promo Active | Feature | NUM | — | Binary indicator specifying whether the store participates in the continuous loyalty coupon initiative Promo2. |
| N | Promo2SinceWeekPromo2 Start Week | Feature | NUM | week | Calendar week index in which the store initially enrolled into the continuous Promo2 program. |
| O | Promo2SinceYearPromo2 Start Year | Feature | NUM | year | Calendar year in which the store enrolled into the continuous Promo2 program. |
| P | PromoIntervalPromo2 Schedule Cycle | Feature | CAT | — | Months in which recurring Promo2 coupon booklets are mailed (e.g. 'Jan,Apr,Jul,Oct'). |
| Q | StoreStore Number | Feature | NUM | — | Numerical business identifier identifying each physical store branch. |
| R | StoreTypeStore Format Type | Feature | CAT | — | Categorical retail store classification format (a, b, c, d) indicating floor plan layout and retail style. |
Interactive Data Table
Explore rows, feature values, and target labels for Rossmann Store Daily & Weekly Sales.
Foundation Models vs Traditional ML & Baselines (41 Methods)
Empirical evaluation directly from the AutoGluon fev benchmark suite evaluating zero-shot foundation models against deep learning, gradient-boosted trees, and statistical baselines across 100 evaluation tasks.
| Rank | Method / Model | Family | Organization | SQL Win Rate?Average win rate (%) against all competing models across the benchmark tasks under Scaled Quantile Loss. | SQL Skill?Normalized skill score relative to the Seasonal Naive baseline (0% = baseline, higher is better). | MASE Win Rate?Average win rate (%) under Mean Absolute Scaled Error. | Latency?Median end-to-end execution time per 100 time series in seconds. | Link |
|---|---|---|---|---|---|---|---|---|
| #1 | Timestral-209M | Foundation Model | Mistral AI | 87.7% | +50.4% | 86.0% | 0.51s | Repo ↗ |
| #2 | Timestral-123M (Carla Engine) | Foundation Model | Mistral AI | 86.8% | +50.2% | 85.0% | 0.32s | Repo ↗ |
| #3 | TimeRSI | Closed-Source API | GaTech & SamayLabs | 84.8% | +51.2% | 83.8% | 2512s | Repo ↗ |
| #4 | t0-gamma | Foundation Model | The Forecasting Company | 84.4% | +49.4% | 81.8% | 4.38s | Repo ↗ |
| #5 | Timestral-28M | Foundation Model | Mistral AI | 79.2% | +48.4% | 77.0% | 0.11s | Repo ↗ |
| #6 | timesfm-3 | Foundation Model | 78.5% | +49.0% | 77.0% | 3.41s | Repo ↗ | |
| #7 | ChakraTS | Baseline | Community | 78.2% | +48.1% | 79.5% | 14.22s | Repo ↗ |
| #8 | Lingjiang2.0 | Baseline | Community | 73.9% | +47.0% | 71.3% | 8.89s | Repo ↗ |
| #9 | Chronos-2 | Foundation Model | AWS | 70.4% | +47.3% | 68.4% | 0.84s | Repo ↗ |
| #10 | timeh2o-v1-tiny | Foundation Model | H2O.ai | 69.9% | +45.5% | 67.0% | 0.19s | Repo ↗ |
| #11 | Timestral-5M | Foundation Model | Mistral AI | 69.0% | +46.2% | 67.7% | 40ms | Repo ↗ |
| #12 | Toto-2.0-2.5B | Baseline | Community | 67.5% | +44.4% | 65.7% | 5.58s | Repo ↗ |
| #13 | t0-beta | Foundation Model | The Forecasting Company | 67.4% | +46.6% | 64.5% | 2.29s | Repo ↗ |
| #14 | Toto-2.0-1B | Baseline | Community | 66.4% | +44.4% | 65.5% | 2.48s | Repo ↗ |
| #15 | TiRex-2 | Foundation Model | NX-AI | 65.8% | +45.5% | 63.4% | 0.27s | Repo ↗ |
| #16 | Toto-2.0-313m | Baseline | Community | 65.1% | +44.2% | 63.5% | 1.06s | Repo ↗ |
| #17 | TiRex | Foundation Model | NX-AI | 58.0% | +42.6% | 54.1% | 0.24s | Repo ↗ |
| #18 | Toto-2.0-22m | Baseline | Community | 57.6% | +42.9% | 56.0% | 0.25s | Repo ↗ |
| #19 | TimesFM-2.5 | Baseline | Community | 54.6% | +42.2% | 53.5% | 1.87s | Repo ↗ |
| #20 | TabPFN-TS-3 | Foundation Model | Prior Labs | 52.4% | +43.1% | 51.3% | 235s | Repo ↗ |
| #21 | TS-ICL | Foundation Model | EDF-Lab | 51.2% | +43.1% | 49.8% | 2.49s | Repo ↗ |
| #22 | FlowState | Foundation Model | IBM | 49.3% | +39.9% | 49.9% | 2.29s | Repo ↗ |
| #23 | citras-fm | Foundation Model | Hitachi | 48.8% | +41.2% | 46.9% | 0.60s | Repo ↗ |
| #24 | Toto-1.0 | Baseline | Community | 47.5% | +40.8% | 45.9% | 22.35s | Repo ↗ |
| #25 | TabPFN-TS | Foundation Model | Prior Labs | 47.3% | +41.5% | 45.4% | 109s | Repo ↗ |
| #26 | Toto-2.0-4m | Baseline | Community | 46.2% | +40.7% | 43.6% | 0.20s | Repo ↗ |
| #27 | Moirai-2.0 | Baseline | Community | 41.9% | +39.3% | 40.5% | 0.35s | Repo ↗ |
| #28 | Chronos-Bolt | Foundation Model | AWS | 41.4% | +38.9% | 39.3% | 0.26s | Repo ↗ |
| #29 | TFT | Deep Learning | — | 31.7% | +32.2% | 29.2% | 829s | Repo ↗ |
| #30 | Stat. Ensemble | Baseline | Community | 27.5% | +20.7% | 30.6% | 147s | Repo ↗ |
| #31 | Sundial-Base | Foundation Model | Tsinghua University | 27.4% | +33.4% | 33.6% | 8.01s | Repo ↗ |
| #32 | PatchTST | Deep Learning | — | 26.8% | +30.1% | 25.1% | 664s | Repo ↗ |
| #33 | DeepAR | Deep Learning | — | 26.1% | +29.1% | 23.6% | 1113s | Repo ↗ |
| #34 | AutoARIMA | Statistical Baseline | — | 23.3% | +21.3% | 21.5% | 20.14s | Repo ↗ |
| #35 | AutoETS | Statistical Baseline | — | 20.3% | -26.8% | 22.5% | 3.48s | Repo ↗ |
| #36 | CatBoost | Tree Ensemble | — | 19.1% | +23.0% | 35.1% | 29.48s | Repo ↗ |
| #37 | LightGBM | Tree Ensemble | — | 17.8% | +21.0% | 33.1% | 3.03s | Repo ↗ |
| #38 | AutoTheta | Statistical Baseline | — | 15.6% | +5.5% | 19.8% | 3.28s | Repo ↗ |
| #39 | Seasonal Naive | Baseline | Community | 10.3% | 0.0% | 11.8% | 0.48s | Repo ↗ |
| #40 | Naive | Statistical Baseline | — | 6.7% | -45.4% | 10.9% | 0.47s | Repo ↗ |
| #41 | Drift | Statistical Baseline | — | 6.1% | -45.8% | 10.0% | 0.45s | Repo ↗ |
Variable Schema & Summary Distributions
Observed numerical ranges, category cardinalities, missing rates, and sample values across 148,295 rows.
| Col | Variable Name | Type | Missing | Distinct | Summary Stats / Distribution | Sample Values |
|---|---|---|---|---|---|---|
| C | SalesTARGET | NUM | 0 | 52,530 | [0, 195930] μ=39110.0 σ=17813.9 | 329522597833071 |
| A | id | CAT | 0 | 1,115 | — | 111 |
| B | timestamp | CAT | 0 | 133 | — | 2013-01-13T00:00:002013-01-20T00:00:002013-01-27T00:00:00 |
| D | Open | NUM | 0 | 8 | [0, 1] μ=0.8 σ=0.2 | 0.85714286565780640.85714286565780640.8571428656578064 |
| E | Promo | NUM | 0 | 3 | [0, 0.7142857313156128] μ=0.4 σ=0.4 | 0.714285731315612800.7142857313156128 |
| F | SchoolHoliday | NUM | 0 | 8 | [0, 7] μ=1.2 σ=2.1 | 500 |
| G | StateHoliday | NUM | 0 | 3 | [0, 2] μ=0.2 σ=0.4 | 000 |
| H | Customers | NUM | 0 | 11,625 | [0, 30188] μ=4291.6 σ=2440.1 | 391834173862 |
| I | Assortment | CAT | 0 | 3 | — | aaa |
| J | CompetitionDistance | NUM | 399 (0.00269%) | 654 | [20, 75860] μ=5404.9 σ=7659.8 | 127012701270 |
| K | CompetitionOpenSinceMonth | NUM | 47,082 (0.31749%) | 12 | [1, 12] μ=7.2 σ=3.2 | 999 |
| L | CompetitionOpenSinceYear | NUM | 47,082 (0.31749%) | 23 | [1900, 2015] μ=2008.7 σ=6.2 | 200820082008 |
| M | Promo2 | NUM | 0 | 2 | [0, 1] μ=0.5 σ=0.5 | 000 |
| N | Promo2SinceWeek | NUM | 72,352 (0.48789%) | 24 | [1, 50] μ=23.6 σ=14.1 | 404040 |
| O | Promo2SinceYear | NUM | 72,352 (0.48789%) | 7 | [2009, 2015] μ=2011.8 σ=1.7 | 201420142014 |
| P | PromoInterval | CAT | 72,352 (0.48789%) | 3 | — | Jan,Apr,Jul,OctJan,Apr,Jul,OctJan,Apr,Jul,Oct |
| Q | Store | NUM | 0 | 1,115 | [1, 1115] μ=558 σ=321.9 | 111 |
| R | StoreType | CAT | 0 | 4 | — | ccc |
Frequently Asked Questions: Rossmann Store Daily & Weekly Sales
Common questions regarding the Rossmann Store Daily & Weekly Sales dataset, machine learning task formulations, and in-browser tabular inference.
What is the forecasting horizon for Rossmann Store Daily & Weekly Sales?
The benchmark evaluates a 13-week (one full quarter) forward horizon, reflecting the standard operational replenishment and financial planning cycle of multi-branch retail networks.
Can I run zero-server forecasts on this dataset in Google Sheets?
Yes, Carla enables zero-server in-context time series forecasting directly in spreadsheets using Mistral's Timestral foundation model.
Which models perform best on Rossmann Store Daily & Weekly Sales?
Gradient boosted lag models (LightGBM) and modern zero-shot foundation models (such as Mistral's Timestral and TimesFM) achieve superior accuracy by effectively utilizing promotion schedules and cross-series structural store similarities.
Test TabICLv2 on Rossmann Store Daily & Weekly Sales Yourself
Open the pre-loaded Google Sheet and let Carla configure the target and task for local, zero-cloud tabular machine learning.
Launch Carla
Click once to open the spreadsheet and Carla side panel together.
Review the Setup
Carla selects the dataset target and task from this page automatically.
Evaluate & Predict
Run predictions and compute metrics with zero server uploads.
Provenance & Attribution
Rossmann Store Sales Prediction Challenge. Kaggle, 2015. https://www.kaggle.com/c/rossmann-store-sales