Downloads · 30 days
6
17% of all-time downloads
goethe_v3
tope1129/goethe_v3
goethe_v3 is a machine learning model from tope1129. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
Upgraded generator combining the three public top artifacts, reweighted against lagged cascade-eval-pool/snapshots (through 2026-07-22).
All-time downloads
35
Public
Repo size
—
Likes
0
Public
Repo makeup
Click a slice to open those files.
.py73.1 KB · 94%
At a glance
- Access
- Public
- Created
- Jul 24, 2026
- Updated
- Jul 24, 2026
- SHA
- bbeb6bb6
Model card
From the Hugging Face model README
cascade-poolaligned-v23
Upgraded generator combining the three public top artifacts, reweighted against lagged cascade-eval-pool/snapshots (through 2026-07-22).
Sources
| Rank | Hub artifact | Kept |
|---|---|---|
| 1 (king) | jan/cascade9 | Dynamics-heavy core, prefetch stream, spectral/long-memory/OU families |
| 2 | cascade-gen/cascade-v21 | Coupled calendar pairs, periods 15/60/240, integrated Student-t + SV, TSMixup, prefix pad, proportional sanitize |
| 3 | cascade-gen/cascade-v17-preparing1 | retail_demand, richer intermittent/pulse, prefix-calibrated artifacts, burn-in / Davies–Harte long memory |
Pool-driven upgrades (v23)
- Cadence mass ≈ empirical periods: 7, 96, 288, 60, 240, 144, 48, 24, 15
- Family mass ↑ nature sensors / spectral / long-memory / retail / counts; ↓ chaos
- Ultra-persistent nonnegative sensor archetype (streamflow/SST-like)
- Pageview-scale lognormal bursts in counts + heavier pulse minority
- Nonnegative lift on a majority of spectral GP rows
- Slightly higher couple rate, pad_prefix, and TSMixup for short pooled histories
Validate
python -m cascade.miner.cli verify ./_top_codes/cascade-v23-poolaligned --chain-toml chain.toml
Contract validity and CPU throughput do not establish forecasting quality. Run a production-faithful GPU A/B against the current king before deploying.
- Created
- Jul 24, 2026
- Updated
- Jul 24, 2026