cris-cmd/michi-audio-affect
187
1{2 "activation_dropout": 0.0,3 "adapter_attn_dim": null,4 "adapter_kernel_size": 3,5 "adapter_stride": 2,6 "add_adapter": false,7 "apply_spec_augment": true,8 "architectures": [9 "Wav2Vec2ForSequenceClassification"10 ],11 "attention_dropout": 0.1,12 "bos_token_id": 1,13 "classifier_proj_size": 768,14 "codevector_dim": 256,15 "contrastive_logits_temperature": 0.1,16 "conv_bias": false,17 "conv_dim": [18 512,19 512,20 512,21 512,22 512,23 512,24 51225 ],26 "conv_kernel": [27 10,28 3,29 3,30 3,31 3,32 2,33 234 ],35 "conv_stride": [36 5,37 2,38 2,39 2,40 2,41 2,42 243 ],44 "ctc_loss_reduction": "sum",45 "ctc_zero_infinity": false,46 "diversity_loss_weight": 0.1,47 "do_stable_layer_norm": false,48 "dtype": "float32",49 "eos_token_id": 2,50 "feat_extract_activation": "gelu",51 "feat_extract_norm": "group",52 "feat_proj_dropout": 0.1,53 "feat_quantizer_dropout": 0.0,54 "final_dropout": 0.0,55 "freeze_feat_extract_train": true,56 "hidden_act": "gelu",57 "hidden_dropout": 0.1,58 "hidden_size": 768,59 "id2label": {60 "0": "ang",61 "1": "dis",62 "2": "fea",63 "3": "hap",64 "4": "neu",65 "5": "sad"66 },67 "initializer_range": 0.02,68 "intermediate_size": 3072,69 "label2id": {70 "ang": 0,71 "dis": 1,72 "fea": 2,73 "hap": 3,74 "neu": 4,75 "sad": 576 },77 "layer_norm_eps": 1e-05,78 "layerdrop": 0.0,79 "mask_channel_length": 10,80 "mask_channel_min_space": 1,81 "mask_channel_other": 0.0,82 "mask_channel_prob": 0.0,83 "mask_channel_selection": "static",84 "mask_feature_length": 10,85 "mask_feature_min_masks": 0,86 "mask_feature_prob": 0.0,87 "mask_time_length": 10,88 "mask_time_min_masks": 2,89 "mask_time_min_space": 1,90 "mask_time_other": 0.0,91 "mask_time_prob": 0.05,92 "mask_time_selection": "static",93 "model_type": "wav2vec2",94 "no_mask_channel_overlap": false,95 "no_mask_time_overlap": false,96 "num_adapter_layers": 3,97 "num_attention_heads": 12,98 "num_codevector_groups": 2,99 "num_codevectors_per_group": 320,100 "num_conv_pos_embedding_groups": 16,101 "num_conv_pos_embeddings": 128,102 "num_feat_extract_layers": 7,103 "num_hidden_layers": 12,104 "num_negatives": 100,105 "output_hidden_size": 768,106 "pad_token_id": 0,107 "proj_codevector_dim": 256,108 "tdnn_dilation": [109 1,110 2,111 3,112 1,113 1114 ],115 "tdnn_dim": [116 512,117 512,118 512,119 512,120 1500121 ],122 "tdnn_kernel": [123 5,124 3,125 3,126 1,127 1128 ],129 "transformers_version": "4.57.6",130 "use_weighted_layer_sum": true,131 "vocab_size": 32,132 "xvector_output_dim": 512133}134 