Apppublic

RabbitRUI/ruispace

sourceHugging Facemitupdated 3y agoView on Hugging Face

0likes

auido2pose.yaml50 linesDownload Raw Back to config

1DATASET:2  TRAIN_FILE_LIST: /apdcephfs_cq2/share_1290939/wenxuazhang/code/audio2pose_unet_noAudio/dataset/train_33.txt3  EVAL_FILE_LIST: /apdcephfs_cq2/share_1290939/wenxuazhang/code/audio2pose_unet_noAudio/dataset/val.txt4  TRAIN_BATCH_SIZE: 645  EVAL_BATCH_SIZE: 16  EXP: True7  EXP_DIM: 648  FRAME_LEN: 329  COEFF_LEN: 7310  NUM_CLASSES: 4611  AUDIO_ROOT_PATH: /apdcephfs_cq2/share_1290939/wenxuazhang/voxceleb1/wav12  COEFF_ROOT_PATH: /apdcephfs_cq2/share_1290939/shadowcun/datasets/VoxCeleb/v1/imdb13  DEBUG: True14  15 16MODEL:17  AUDIOENCODER:18    LEAKY_RELU: True19    NORM: 'IN'20  DISCRIMINATOR:21    LEAKY_RELU: False22    INPUT_CHANNELS: 623  CVAE:24    AUDIO_EMB_IN_SIZE: 51225    AUDIO_EMB_OUT_SIZE: 626    SEQ_LEN: 3227    LATENT_SIZE: 6428    ENCODER_LAYER_SIZES: [192, 128]29    DECODER_LAYER_SIZES: [128, 192]30    31 32TRAIN:33  MAX_EPOCH: 15034  GENERATOR:35    LR: 1.0e-436  DISCRIMINATOR:37    LR: 1.0e-438  LOSS:39    LAMBDA_REG: 140    LAMBDA_LANDMARKS: 041    LAMBDA_VERTICES: 042    LAMBDA_GAN_MOTION: 0.743    LAMBDA_GAN_COEFF: 044    LAMBDA_KL: 145 46TAG:47  NAME: cvae_UNET_useAudio_usewav2lipAudioEncoder48 49 50