VAE trained on Banking 77 Open Intent Classification Dataset

Architecture

• input_dim: 768
• hidden_dim: 256
• latent_dim: 64

Encoder

Decoder

Metrics

1. Training set: VAE Loss
   • 50% reconstruction loss between original input vs reconstructed output
   • 50% KL divergence between Latent Z vs standard normal distribution
2. Validation set: 100% reconstruction loss -> used to find the best model (with the lowest reconstruction loss)