RQ-VAE (NYC)

Generated at: 2025-11-04T18:55:06.576622+00:00Z

Dataset

  • Name: NYC
  • POIs: 5135
  • Input dim: 1407
  • Feature dims: {'category': 209, 'region': 91, 'temporal': 24, 'collaborative': 1083}

Model

  • VAE hidden dims: [128, 256, 512]
  • Codebooks: 3 x 32 x 64
  • Commitment weight: 0.25
  • Loss weights: {'reconstruction': 0.0, 'quantization': 1.0, 'utilization': 0.25, 'compactness': 0.25}

Training

  • Batch size: 128
  • Epochs: 3000
  • LR: 0.001
Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support