Kalhara JA← All Works
KJA·2026 · Case study
SHOPLYTICS
Phase-adaptive donor-guided forecasting on the VISUELLE 2.0 SO-fore benchmark
The brief
A final-year research project on demand forecasting for new fashion products, combining a phase-adaptive similarity engine with donor-guided LSTM transfer to reach WAPE 90.82 on the official VISUELLE 2.0 benchmark.
Approach
Combined a phase-adaptive multi-metric similarity engine with donor-guided LSTM transfer fine-tuning and a BCa-bootstrap negative-transfer gate, evaluated end-to-end on the official VISUELLE 2.0 SO-fore benchmark.
Outcome
Reached WAPE 90.82 — a statistically significant 2.17-point improvement over the strongest baseline, rising to +9.54 WAPE on the low-volume SME-relevant tier — reproducible end-to-end with 330 automated tests.
Key features
Phase-adaptive progressive similarity engine (Contribution 1)
The active set of similarity metrics adapts to the observation window length — Phase A uses metadata + image; Phase B adds demand correlation; Phase C adds seasonality alignment + volatility matching. Heterogeneous metrics fused by parameter-free rank averaging.
Donor-guided LSTM transfer fine-tuning (Contribution 2)
Weekly-native LSTM pretrained on 96,166 (product, store) pairs; per-target frozen-base head fine-tune on the top-K=20 donors selected by Contribution 1. Paired bootstrap shows the LSTM significantly improves over the donor-weighted aggregator (skill +0.013, 95% CI [+0.010, +0.016]).
BCa-bootstrap negative-transfer gate (Contribution 3)
Per-product runtime decision rule based on the lower bound of a 95% bias-corrected-and-accelerated bootstrap CI on the skill score s = 1 − WAPE_transfer / WAPE_baseline. Decision boundary at zero, no tuned threshold. Diebold-Mariano test retained as an independent diagnostic.
Reproduced Skenderi 2022 Tier 1+2 baselines
Naive, SES, kNN time-series (k=11), and kNN+image (CLIP and ResNet-101) reproductions on the official VISUELLE 2.0 SO-fore_2-10 split, all within published tolerance. Both image variants reproduce the documented qualitative finding that motivates Contribution 1.
Statistically rigorous evaluation
Paired bootstrap CIs (1000 resamples) on every WAPE / MAE / skill cell. Population-level skill score replaces a per-pair gate at T_obs=14d, where rolling-origin CV cannot produce enough folds for a meaningful BCa CI.
Per-tier breakdown surfacing SME-relevant value
Tercile analysis on observed-mean weekly volume reveals the biggest win lands on low-volume pairs (n=2,337): WAPE 87.23 vs kNN-ts 96.77, a +9.54 WAPE improvement — the cold-start regime SHOPLYTICS targets.
End-to-end reproducibility from a clean checkout
scripts/run_official_evaluation.sh reproduces the entire baseline table; scripts/module1/run_official_pipeline.py reproduces the headline method. Frozen test-set manifest with SHA256 hashes; pretrain manifests record holdout exclusion; 330 automated tests.
Team
Kalhara J. A. K.Researcher, Engineer, and Author
Gallery