From 647468123eba86a360cb09116c74690433a75f4d Mon Sep 17 00:00:00 2001 From: mbsantiago Date: Thu, 26 Jun 2025 10:04:00 -0600 Subject: [PATCH] Update model config in example config --- config.yaml | 44 +++++++++++++++++++++++++++++++++++--------- 1 file changed, 35 insertions(+), 9 deletions(-) diff --git a/config.yaml b/config.yaml index 178851b..c0021f7 100644 --- a/config.yaml +++ b/config.yaml @@ -79,12 +79,36 @@ labels: sigma: 3 model: - name: Net2DFast input_height: 128 - encoder_channels: [1, 32, 64, 128] - bottleneck_channels: 256 - decoder_channels: [256, 64, 32, 32] + in_channels: 1 out_channels: 32 + encoder: + layers: + - block_type: FreqCoordConvDown + out_channels: 32 + - block_type: FreqCoordConvDown + out_channels: 64 + - block_type: LayerGroup + layers: + - block_type: FreqCoordConvDown + out_channels: 128 + - block_type: ConvBlock + out_channels: 256 + bottleneck: + channels: 256 + self_attention: true + decoder: + layers: + - block_type: FreqCoordConvUp + out_channels: 64 + - block_type: FreqCoordConvUp + out_channels: 32 + - block_type: LayerGroup + layers: + - block_type: FreqCoordConvUp + out_channels: 32 + - block_type: ConvBlock + out_channels: 32 train: batch_size: 8 @@ -104,29 +128,31 @@ train: optimizer: learning_rate: 0.001 t_max: 100 + logger: + logger_type: dvclive augmentations: steps: - - augmentation_type: mix + - augmentation_type: mix_audio probability: 0.2 min_weight: 0.3 max_weight: 0.7 - - agumentation_type: echo + - augmentation_type: add_echo probability: 0.2 max_delay: 0.005 min_weight: 0.0 max_weight: 1.0 - - augmentation_type: volume + - augmentation_type: scale_volume probability: 0.2 min_scaling: 0.0 max_scaling: 2.0 - augmentation_type: warp probability: 0.2 delta: 0.04 - - augmentation_type: time_mask + - augmentation_type: mask_time probability: 0.2 max_perc: 0.05 max_masks: 3 - - augmentation_type: frequency_mask + - augmentation_type: mask_freq probability: 0.2 max_perc: 0.10 max_masks: 3