RMDig/snowGAN-core

snowGAN — core backbone (v0.1.0)

WGAN-GP trained on the Rocky Mountain Snowpack dataset, single-modality (core), depth=1, 1024x1024 resolution. Published from the snowGAN project as a research artifact.

Not for safety decisions. Nothing here predicts avalanche danger, slope stability or snowpack breakability. Do not use it for any backcountry decision.

How to use

The canonical consumer-side path uses snowgan.weights.fetch to download + cache, then snowgan.models.Discriminator to rebuild and load:

from snowgan.weights import fetch
from snowgan.config import build
from snowgan.models.discriminator import Discriminator

# Fetch the release (cached locally after first call).
path = fetch("RMDig/snowGAN-core", "v0.1.0")

# Reconstruct the model from its sidecar config, then load weights.
cfg = build(str(path / "discriminator_config.json"))
disc = Discriminator(cfg)
disc.model.build((None, cfg.depth, cfg.resolution[0], cfg.resolution[1], cfg.channels))
disc.model.load_weights(str(path / "discriminator.weights.h5"))

# Tap the named features layer — the cross-repo contract with downstream consumers.
features = disc.model.get_layer("features")
print("backbone features:", features.output.shape)  # (None, 1048576)

Requires pip install snowgan[hub] (pulls in huggingface_hub). Without the [hub] extra, fetch() raises a clean ImportError naming the missing dep.

This release's generator_config.json predates the generator's architecture fields. With snowgan 1.0.0, rebuild the generator with gen_upsampler="transpose" and gen_convs_per_resolution=1; the 1.0.0 defaults describe a different generator and the weights will not load.

Intended use

Research artifact from snowGAN's GAN-training experiments. The discriminator's features layer and the generator are published so the results can be reproduced and tested. Publishing them does not mean they are known to be useful downstream.

Transfer value is unmeasured. Nobody has measured whether this core backbone's features help any task, or whether they beat a randomly initialised copy of the same network. The one such comparison so far, on the companion magnified-profile backbone, favoured random init (see Quality evidence). Compare against that baseline before relying on these features.

Architecture

Field	Value
Modality	core (depth=1)
Resolution	1024x1024
Channels	3
Latent dim	100
Generator filter counts	[1024, 512, 256, 128, 64]
Discriminator filter counts	[64, 128, 256, 512, 1024]
Conv kernel / stride	[3, 3] / [2, 2]
Backbone (Flatten "features") dim	1048576
Final activation	tanh

The discriminator's Conv3D layers use ksize=(1, kH, kW), so the depth axis is broadcast — kernels themselves are depth-agnostic. This is the contract that lets downstream consumers compose multiple single-modality backbones into a depth=N model (e.g. profile + core stacked at depth=2 for paired-modality transfer).

Training

Trained with WGAN-GP loss (Wasserstein + gradient penalty, λ_gp=10.0) on the core samples of rmdig/rocky_mountain_snowpack. At release: fade_step=130000.

Stabilizer	Setting
Spectral norm	False
Differentiable augmentation	True
Adaptive augmentation (ADA) target	0.6
Adaptive disc/gen step ratio	False
EMA decay (generator shadow)	0.999
Multi-scale discriminator	True
Gradient clip (global norm)	1.0
LR decay schedule	cosine (lr_min=1e-07)
FID eval interval	5000 steps (no valid FID; see Quality evidence)

Training data and splits

• 13 (site, column, core) groups, all from sites 0-2, one operator. The dataset's sites 3-6 were added on 2026-08-17, after this release, and are unseen by it.
• The sidecars persist a trained_pool / validation_pool / test_pool split (10 / 1 / 2 groups). These pools are not held out from this model. The training loop at release ignored them, so the GAN trained on all 13 groups (snowGAN UPGRADES #54).
• The GAN streamed the dataset's row-level train split as it stood at release: 391 of the 495 images of this modality, covering all 13 groups. The remaining images come from the same groups, so they are not independent held-out data. Sites 3-6 are the only data unseen at the group level.

Quality evidence

• No valid FID. The in-training FID used 64 samples against 2048-dimensional Inception features, a rank-deficient and sample-size-biased estimate. In this run it also compared core samples against magnified-profile reals. The fid_interval setting above produced no usable number. No KID or memorisation check has been run on this release yet.
• The logged critic loss left the 1-Lipschitz range early. A 1-Lipschitz critic on inputs in [-1, 1] at 1024x1024x3 has |Wasserstein estimate| at most 2sqrt(10241024*3) ≈ 3,547. Single steps exceeded that from step ~120; the logged |loss| stayed above it persistently from about step 14-15k (median 3,599 over steps 13.5k-18.5k). From 80k to the end the mean |loss| was 28,903 and the peak 156,321 (230,377 over the whole run), as a roughly symmetric ± oscillation rather than a critic that was winning.
  • Caveat: the logged value is main-critic Wasserstein + λ·GP + 0.5·(low-res critic Wasserstein). The low-res critic had no gradient penalty and no spectral norm, so nothing bounded it. Values past the ceiling show that at least one of the two critics exceeded its bound; the log cannot say which.
• disc:gen update ratio 2:3 throughout (adaptive steps off).
• Downstream measurement (snowGradient, 2026-10). In a pre-registered modality-recognition probe (core vs profile vs magnified profile, leave-one-site-out over sites 3-6, at 256x256 with global average pooling), the pretrained magnified-profile backbone scored 0.868 balanced accuracy. A randomly initialised copy of the same network scored 0.987 / 0.982 (2 seeds) and was better on every held-out site. 24 of the pretrained backbone's 26 errors were core <-> magnified-profile confusions. This is one task, at a quarter of training resolution.
• The core backbone itself has not been compared against random init.

Limitations

• Trained on the core modality of rmdig/rocky_mountain_snowpack; generalization untested.
• The logged critic loss was outside the 1-Lipschitz range for most of training (see Quality evidence), so the score is not a usable Wasserstein estimate. Its features have not been shown to be informative.
• Single-modality backbone (depth=1); paired-modality features must be composed on the consumer side. See AvAIs load_backbones for the canonical two-backbone composition pattern.
• 13 groups from sites 0-2, one operator: a very small, narrow training set. There is no held-out data from those sites (see Training data and splits).

Release history

Two commits titled "Release v0.1.0" were made about 50 minutes apart on 2026-06-03. The first carried no model card (a 3-line README); the second added this card and corrected the MANIFEST's provenance. The v0.1.0 tag was created on the second commit, one second after it; there is no record of a tag on the first. All weight files are byte-identical across both commits (same LFS sha256), so either snapshot computes the same function.

Card corrected 2026-10: removed unmeasured claims about feature quality and transfer learning, and added the measurements above.

Files in this release

• discriminator.weights.h5 — main discriminator weights.
• discriminator_config.json — architecture sidecar; pass to snowgan.models.Discriminator(cfg).
• generator.weights.h5 + generator_config.json — generator weights and sidecar.
• generator_ema.weights.h5 — EMA shadow weights (only if EMA was enabled during training).
• generator_fade_endpoints.weights.h5 — progressive-fade toRGB endpoints (only if fade was used).
• discriminator_lowres.weights.h5 — multi-scale 256×256 critic (only if multiscale_disc was on).
• MANIFEST.md — full provenance dump (git SHA, every training flag, every artifact). Read this for debugging.
• README.md — this file.

License

Apache 2.0 — see the snowGAN repository for the full license text.

Cross-references

• Source code: github.com/dennys246/snowGAN
• Downstream project: github.com/dennys246/AvAI
• Companion modality backbones: RMDig/snowGAN-magnified-profile, RMDig/snowGAN-core
• Training dataset: rmdig/rocky_mountain_snowpack

Release notes

First core release. Trained ~127k steps without spectral_norm. The fade-step counter reads 130,000 but the loss log has 127,616 steps. The logged discriminator loss left the 1-Lipschitz range persistently from about step 14-15k (see Quality evidence).

Corrected 2026-10. Earlier text said the backbone "has learned meaningful features" and is "usable for transfer learning". Neither claim was measured, and the one downstream comparison against random init (on the magnified-profile backbone) went the other way. It also dated the divergence to step 80k.