AtmosIQ's picture
Upload folder using huggingface_hub
7215ff1 verified
|
Raw History Blame Contribute Delete
1.1 kB
# AtmosIQ — Finest Models (v3 final)
**Checkpoint:** `atmosiq_v3_lp_final.pt` (289,740 params) — SSL encoder (frozen) + head-only
supervised stage (TP-7 val_mse selection). Encoder: `encoder_v3.pt`.
**M2 spatial-held-out R² (h0, ungauged cells, vs M1 XGBoost floor):**
pm2_5 +0.303 (xgb 0.30, BEAT) | pm10 +0.363 (0.35, BEAT) | o3 +0.190 (0.13, BEAT)
co +0.083 (0.06, BEAT) | nh3 +0.118 (0.06, BEAT) | so2 -0.008 (xgb -0.05) | no2 +0.029 (0.07, below)
**Component evidence (not a per-node MLP):** graph_gain pm10 +0.041 / o3 +0.085;
alpha_self 0.45/0.63; ORPHAN eval (test node's own satellite+static masked, neighbors only):
pm2_5 +0.113, pm10 +0.211 — the ungauged-cell claim, demonstrated.
**Protocol:** TP-1 full pruned-graph training (no node sampling), structural spatial holdout,
seed-pinned (digit-level reproducible). λ_phys=0 (covariance proxy refuted by EXP-C, see
logs/eval_lp_nophys.txt). Known limits: no2 weakest; σ/ν calibration pending U-1.
**Reproduce:** scripts/ + tensors/ repo folder + norm.npz + refs.json →
`train_v3.py --stage all` then `evaluate_v3.py`.