File size: 5,480 Bytes
637209e | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 | {
"architectures": [
"MSPVisualForCTC"
],
"auto_map": {
"AutoConfig": "configuration_msp_visual.MSPVisualConfig",
"AutoModelForCTC": "modeling_msp_visual.MSPVisualForCTC"
},
"bos_token_id": 1,
"ctc_loss_reduction": "mean",
"ctc_zero_infinity": true,
"dtype": "float32",
"eos_token_id": 2,
"final_dropout": 0.1,
"model_type": "msp_visual",
"pad_token_id": 0,
"transformers_version": "5.10.2",
"use_cache": false,
"visual_config": {
"_name_or_path": "MahmoodAnaam/avhubert_encoder_large_noise_pt_noise_ft_433h-V0",
"a_upsample_ratio": 1,
"activation_dropout": 0.0,
"activation_function": "relu",
"adapter_attn_dim": null,
"adapter_kernel_size": 3,
"adapter_stride": 2,
"add_adapter": false,
"adim": 1024,
"aheads": 12,
"apply_spec_augment": true,
"architectures": [
"AVHubertModel"
],
"attention_dropout": 0.1,
"audio_dropout": 0.5,
"audio_feat_dim": 104,
"auto_map": {
"AutoConfig": "configuration_avhubert.AVHubertConfig",
"AutoModel": "modeling_avhubert.AVHubertModel"
},
"bos_token_id": 1,
"classifier_proj_size": 256,
"cnn_module_kernel": 31,
"codevector_dim": 256,
"contrastive_logits_temperature": 0.1,
"conv_bias": false,
"conv_channels": 1024,
"conv_dim": [
512,
512,
512,
512,
512,
512,
512
],
"conv_kernel": [
10,
3,
3,
3,
3,
2,
2
],
"conv_kernel_sizes": [
5,
5
],
"conv_stride": [
5,
2,
2,
2,
2,
2,
2
],
"ctc_loss_reduction": "sum",
"ctc_type": "builtin",
"ctc_zero_infinity": false,
"d_model": 1024,
"ddim": 1024,
"decoder_attention_heads": 8,
"decoder_ffn_dim": 4096,
"decoder_layerdrop": 0.0,
"decoder_layers": 9,
"decoder_start_token_id": 2,
"dheads": 16,
"diversity_loss_weight": 0.1,
"dlayers": 6,
"do_stable_layer_norm": false,
"dropout": 0.1,
"dropout_features": 0.1,
"dropout_input": 0.1,
"dropout_rate": 0.1,
"dtype": "float32",
"dunits": 3072,
"elayers": 12,
"encoder_attention_heads": 16,
"encoder_embed_dim": 1024,
"encoder_ffn_dim": 2048,
"encoder_layerdrop": 0.0,
"encoder_layers": 12,
"eos_token_id": 2,
"eunits": 3072,
"feat_extract_activation": "gelu",
"feat_extract_norm": "group",
"feat_proj_dropout": 0.1,
"feat_quantizer_dropout": 0.0,
"feature_grad_mult": 0.1,
"final_dim": 256,
"final_dropout": 0.0,
"freeze_feat_extract_train": true,
"fusion_hdim": 8192,
"fusion_norm": "batchnorm",
"hidden_act": "gelu",
"hidden_dropout": 0.1,
"hidden_size": 1024,
"init_std": 0.02,
"initializer_range": 0.02,
"input_channels": 1,
"input_feat_per_channel": 80,
"intermediate_size": 4096,
"label_rate": 25,
"layer_norm_eps": 1e-05,
"layerdrop": 0.0,
"logit_temp": 0.1,
"lsm_weight": 0.1,
"macaron_style": true,
"mask_channel_length": 10,
"mask_channel_min_space": 1,
"mask_channel_other": 0.0,
"mask_channel_prob": 0.0,
"mask_channel_selection": "static",
"mask_feature_length": 10,
"mask_feature_min_masks": 0,
"mask_feature_prob": 0.0,
"mask_length_audio": 10,
"mask_length_image": 5,
"mask_min_space": 1,
"mask_other": 0.0,
"mask_prob_audio": 0.8,
"mask_prob_image": 0.3,
"mask_selection": "static",
"mask_time_length": 10,
"mask_time_min_masks": 2,
"mask_time_min_space": 1,
"mask_time_other": 0.0,
"mask_time_prob": 0.0,
"mask_time_selection": "static",
"masking_type": "input",
"max_source_positions": 6000,
"max_target_positions": 2048,
"modality": "av",
"modality_dropout": 0.5,
"modality_fuse": "concat",
"model_type": "avhubert",
"mtlalpha": 0.1,
"no_mask_channel_overlap": false,
"no_mask_overlap": false,
"no_mask_time_overlap": false,
"num_adapter_layers": 3,
"num_attention_heads": 16,
"num_classes": 2004,
"num_codevector_groups": 2,
"num_codevectors_per_group": 320,
"num_conv_layers": 2,
"num_conv_pos_embedding_groups": 16,
"num_conv_pos_embeddings": 128,
"num_dictionaries": 1,
"num_feat_extract_layers": 7,
"num_hidden_layers": 24,
"num_negatives": 100,
"odim": 5049,
"output_hidden_size": 1024,
"pad_token_id": 1,
"proj_codevector_dim": 256,
"rel_pos_type": "latest",
"relu_type": "swish",
"resnet_relu_type": "prelu",
"resnet_weights": null,
"sample_rate": 25,
"scale_embedding": null,
"selection_type": "same_seq",
"sim_type": "cosine",
"skip_masked": false,
"skip_nomask": false,
"sub_encoder_layers": 0,
"target_glu": false,
"tdnn_dilation": [
1,
2,
3,
1,
1
],
"tdnn_dim": [
512,
512,
512,
512,
1500
],
"tdnn_kernel": [
5,
3,
3,
1,
1
],
"transformer_attn_dropout_rate": 0.1,
"transformer_encoder_attn_layer_type": "rel_mha",
"transformer_input_layer": "conv3d",
"transformer_length_normalized_loss": false,
"untie_final_proj": true,
"use_cache": true,
"use_cnn_module": true,
"use_weighted_layer_sum": false,
"vocab_size": 1000,
"xvector_output_dim": 512,
"zero_triu": false
},
"vocab_size": 32
}
|