Discidius/Speech-Emotion-Classification

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": 256,14  "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  "eos_token_id": 2,49  "feat_extract_activation": "gelu",50  "feat_extract_norm": "group",51  "feat_proj_dropout": 0.1,52  "feat_quantizer_dropout": 0.0,53  "final_dropout": 0.0,54  "freeze_feat_extract_train": true,55  "hidden_act": "gelu",56  "hidden_dropout": 0.1,57  "hidden_size": 768,58  "id2label": {59    "0": "ANG",60    "1": "CAL",61    "2": "DIS",62    "3": "FEA",63    "4": "HAP",64    "5": "NEU",65    "6": "SAD",66    "7": "SUR"67  },68  "initializer_range": 0.02,69  "intermediate_size": 3072,70  "label2id": {71    "ANG": 0,72    "CAL": 1,73    "DIS": 2,74    "FEA": 3,75    "HAP": 4,76    "NEU": 5,77    "SAD": 6,78    "SUR": 779  },80  "layer_norm_eps": 1e-05,81  "layerdrop": 0.0,82  "mask_channel_length": 10,83  "mask_channel_min_space": 1,84  "mask_channel_other": 0.0,85  "mask_channel_prob": 0.0,86  "mask_channel_selection": "static",87  "mask_feature_length": 10,88  "mask_feature_min_masks": 0,89  "mask_feature_prob": 0.0,90  "mask_time_length": 10,91  "mask_time_min_masks": 2,92  "mask_time_min_space": 1,93  "mask_time_other": 0.0,94  "mask_time_prob": 0.05,95  "mask_time_selection": "static",96  "model_type": "wav2vec2",97  "no_mask_channel_overlap": false,98  "no_mask_time_overlap": false,99  "num_adapter_layers": 3,100  "num_attention_heads": 12,101  "num_codevector_groups": 2,102  "num_codevectors_per_group": 320,103  "num_conv_pos_embedding_groups": 16,104  "num_conv_pos_embeddings": 128,105  "num_feat_extract_layers": 7,106  "num_hidden_layers": 12,107  "num_negatives": 100,108  "output_hidden_size": 768,109  "pad_token_id": 0,110  "proj_codevector_dim": 256,111  "tdnn_dilation": [112    1,113    2,114    3,115    1,116    1117  ],118  "tdnn_dim": [119    512,120    512,121    512,122    512,123    1500124  ],125  "tdnn_kernel": [126    5,127    3,128    3,129    1,130    1131  ],132  "torch_dtype": "float32",133  "transformers_version": "4.50.0",134  "use_weighted_layer_sum": false,135  "vocab_size": 32,136  "xvector_output_dim": 512137}138