Ethanfel 8e3ab999f0 fix: load VAE state dict with strict=False ...

vae.ckpt is a full training checkpoint containing discriminator, STFT
loss modules, and EMA wrappers that are absent from the inference
AudioAutoencoder. strict=False ignores these training-only keys while
still loading all encoder/decoder/bottleneck weights correctly.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

2026-03-27 19:51:51 +01:00

..

__init__.py

feat: project scaffolding with shared utils and node registration

2026-03-27 16:59:21 +01:00

feature_extractor.py

feat: auto-install pip venv for feature extraction on first use

2026-03-27 19:27:27 +01:00

feature_loader.py

feat: PrismAudioFeatureLoader node for pre-computed .npz files

2026-03-27 18:04:32 +01:00

model_loader.py

fix: load VAE state dict with strict=False

2026-03-27 19:51:51 +01:00

sampler.py

feat: PrismAudioSampler node with correct metadata format and peak normalization

2026-03-27 18:07:33 +01:00

text_only.py

feat: PrismAudioTextOnly node with correct T5-Gemma encoding

2026-03-27 18:09:11 +01:00

utils.py

feat: project scaffolding with shared utils and node registration

2026-03-27 16:59:21 +01:00