Rebound

Navigation

Explore

Choose a platform module or vertical, then jump straight to the corresponding page.

Platform / Module 02

Rashomon Set
Manager

Any serious model team ends up with a set of models that all look equally good on validation. Pick one arbitrarily, and you've silently accepted a risk profile you never evaluated. The Rashomon Set Manager treats the ensemble as the unit of deployment — registered, profiled, health-checked, and continuously revalidated.

pools / wind_dispatch
Pool
wind_dispatch
Models
5
3 families
Diversity score
0.81
target ≥ 0.75
Blind spots
k > 36h
1 band flagged
transformer-v3
Transformer
0.043
Short
0.78
Mid
0.82
Long
0.61
Healthy
reservoir-net
Reservoir
0.045
Short
0.96
Mid
0.71
Long
0.38
Healthy
cnn-lstm
CNN-LSTM
0.044
Short
0.84
Mid
0.88
Long
0.52
Healthy
gbdt-baseline
GBDT
0.047
Short
0.71
Mid
0.68
Long
0.44
Profiling
llm-finetuned-b
LLM
0.046
Short
0.65
Mid
0.77
Long
0.69
Healthy
Reliability decay · per model
How each member of the set degrades with forecast horizon
k* = 14horizon k (steps) →reliability
transformer-v3reservoir-netcnn-lstmgbdt-baselinellm-finetuned-b
Architectural diversity
Coverage vs target
Short-horizonMid-horizonLong-horizonHigh chaosAdversarial

Dashed polygon: target coverage. Solid: current pool. Adversarial axis underperforming — consider adding a robust model.

Last revalidation: 2026-04-12 02:00 UTC · next in 18h

Your ensemble,
governed as one object.

Register models from PyTorch, TensorFlow, JAX, scikit-learn, ONNX Runtime, Hugging Face, SageMaker, Vertex AI, or Azure ML. The manager continuously profiles each one, labels its architectural family, and tracks how its reliability decays across forecast horizons.

  • Multi-framework registration via ONNX or native adapters
  • Automated reliability-decay profiling per model
  • Architecture family tagging (Transformer, Reservoir, CNN, LLM, GBDT…)
  • Diversity scoring and blind-spot band detection
  • Drift-triggered revalidation and scheduled health checks
  • Pool versioning with rollback
python · Rebound sdk
from Rebound import Pool

pool = Pool.create(
    name="wind_dispatch",
    risk="balancing_penalty",
)

pool.register("transformer-v3",  onnx="s3://models/tf_v3.onnx")
pool.register("reservoir-net",   torch="models/res_net.pt")
pool.register("cnn-lstm",        torch="models/cnn_lstm.pt")
pool.register("gbdt-baseline",   joblib="models/gbdt.pkl")
pool.register("llm-finetuned-b", hf="org/llm-ft-b")

report = pool.profile(
    dataset="wind_2025_q4",
    horizons=[1, 6, 12, 24, 36, 48],
)

report.diversity        # 0.81
report.blind_spots      # ["k > 36h"]
report.recommend        # "add: adversarially-robust TCN"

Already have 3+ models in production?

That is a Rashomon set. We will characterize it for you.