Commit
•
f43dd0c
1
Parent(s):
2922451
add config files
Browse files- config.json +109 -0
- preprocessor_config.json +9 -0
- rng_state.pth +3 -0
- scaler.pt +3 -0
- scheduler.pt +3 -0
- trainer_state.json +226 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,109 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "facebook/wav2vec2-large-lv60",
|
| 3 |
+
"activation_dropout": 0.1,
|
| 4 |
+
"adapter_kernel_size": 3,
|
| 5 |
+
"adapter_stride": 2,
|
| 6 |
+
"add_adapter": false,
|
| 7 |
+
"apply_spec_augment": true,
|
| 8 |
+
"architectures": [
|
| 9 |
+
"Wav2Vec2ForCTC"
|
| 10 |
+
],
|
| 11 |
+
"attention_dropout": 0.1,
|
| 12 |
+
"bos_token_id": 1,
|
| 13 |
+
"classifier_proj_size": 256,
|
| 14 |
+
"codevector_dim": 768,
|
| 15 |
+
"contrastive_logits_temperature": 0.1,
|
| 16 |
+
"conv_bias": true,
|
| 17 |
+
"conv_dim": [
|
| 18 |
+
512,
|
| 19 |
+
512,
|
| 20 |
+
512,
|
| 21 |
+
512,
|
| 22 |
+
512,
|
| 23 |
+
512,
|
| 24 |
+
512
|
| 25 |
+
],
|
| 26 |
+
"conv_kernel": [
|
| 27 |
+
10,
|
| 28 |
+
3,
|
| 29 |
+
3,
|
| 30 |
+
3,
|
| 31 |
+
3,
|
| 32 |
+
2,
|
| 33 |
+
2
|
| 34 |
+
],
|
| 35 |
+
"conv_stride": [
|
| 36 |
+
5,
|
| 37 |
+
2,
|
| 38 |
+
2,
|
| 39 |
+
2,
|
| 40 |
+
2,
|
| 41 |
+
2,
|
| 42 |
+
2
|
| 43 |
+
],
|
| 44 |
+
"ctc_loss_reduction": "mean",
|
| 45 |
+
"ctc_zero_infinity": false,
|
| 46 |
+
"diversity_loss_weight": 0.1,
|
| 47 |
+
"do_stable_layer_norm": true,
|
| 48 |
+
"eos_token_id": 2,
|
| 49 |
+
"feat_extract_activation": "gelu",
|
| 50 |
+
"feat_extract_dropout": 0.0,
|
| 51 |
+
"feat_extract_norm": "layer",
|
| 52 |
+
"feat_proj_dropout": 0.1,
|
| 53 |
+
"feat_quantizer_dropout": 0.0,
|
| 54 |
+
"final_dropout": 0.1,
|
| 55 |
+
"gradient_checkpointing": false,
|
| 56 |
+
"hidden_act": "gelu",
|
| 57 |
+
"hidden_dropout": 0.1,
|
| 58 |
+
"hidden_dropout_prob": 0.1,
|
| 59 |
+
"hidden_size": 1024,
|
| 60 |
+
"initializer_range": 0.02,
|
| 61 |
+
"intermediate_size": 4096,
|
| 62 |
+
"layer_norm_eps": 1e-05,
|
| 63 |
+
"layerdrop": 0.0,
|
| 64 |
+
"mask_feature_length": 10,
|
| 65 |
+
"mask_feature_min_masks": 0,
|
| 66 |
+
"mask_feature_prob": 0.0,
|
| 67 |
+
"mask_time_length": 10,
|
| 68 |
+
"mask_time_min_masks": 2,
|
| 69 |
+
"mask_time_prob": 0.05,
|
| 70 |
+
"model_type": "wav2vec2",
|
| 71 |
+
"num_adapter_layers": 3,
|
| 72 |
+
"num_attention_heads": 16,
|
| 73 |
+
"num_codevector_groups": 2,
|
| 74 |
+
"num_codevectors_per_group": 320,
|
| 75 |
+
"num_conv_pos_embedding_groups": 16,
|
| 76 |
+
"num_conv_pos_embeddings": 128,
|
| 77 |
+
"num_feat_extract_layers": 7,
|
| 78 |
+
"num_hidden_layers": 24,
|
| 79 |
+
"num_negatives": 100,
|
| 80 |
+
"output_hidden_size": 1024,
|
| 81 |
+
"pad_token_id": 29,
|
| 82 |
+
"proj_codevector_dim": 768,
|
| 83 |
+
"tdnn_dilation": [
|
| 84 |
+
1,
|
| 85 |
+
2,
|
| 86 |
+
3,
|
| 87 |
+
1,
|
| 88 |
+
1
|
| 89 |
+
],
|
| 90 |
+
"tdnn_dim": [
|
| 91 |
+
512,
|
| 92 |
+
512,
|
| 93 |
+
512,
|
| 94 |
+
512,
|
| 95 |
+
1500
|
| 96 |
+
],
|
| 97 |
+
"tdnn_kernel": [
|
| 98 |
+
5,
|
| 99 |
+
3,
|
| 100 |
+
3,
|
| 101 |
+
1,
|
| 102 |
+
1
|
| 103 |
+
],
|
| 104 |
+
"torch_dtype": "float32",
|
| 105 |
+
"transformers_version": "4.24.0",
|
| 106 |
+
"use_weighted_layer_sum": false,
|
| 107 |
+
"vocab_size": 32,
|
| 108 |
+
"xvector_output_dim": 512
|
| 109 |
+
}
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
| 4 |
+
"feature_size": 1,
|
| 5 |
+
"padding_side": "right",
|
| 6 |
+
"padding_value": 0.0,
|
| 7 |
+
"return_attention_mask": true,
|
| 8 |
+
"sampling_rate": 16000
|
| 9 |
+
}
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9f6b6a681c875a506e76aae537b7dff4b97c26bd9fee398027f733029d374668
|
| 3 |
+
size 14639
|
scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:298dad932c21e05d75d3da5a679e913eff6d3ab77165a881afe2baf33c70cdcb
|
| 3 |
+
size 557
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:93e09d5c51cc6495794492c0eeece2bdfc76e91c16615ab0b5a5a8ef29dd110b
|
| 3 |
+
size 627
|
trainer_state.json
ADDED
|
@@ -0,0 +1,226 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 28.11244979919679,
|
| 5 |
+
"global_step": 7000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 2.01,
|
| 12 |
+
"learning_rate": 4.97e-05,
|
| 13 |
+
"loss": 4.8083,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 2.01,
|
| 18 |
+
"eval_loss": 2.9298839569091797,
|
| 19 |
+
"eval_runtime": 72.494,
|
| 20 |
+
"eval_samples_per_second": 23.174,
|
| 21 |
+
"eval_steps_per_second": 2.897,
|
| 22 |
+
"eval_wer": 1.0,
|
| 23 |
+
"step": 500
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 4.02,
|
| 27 |
+
"learning_rate": 9.970000000000001e-05,
|
| 28 |
+
"loss": 1.5432,
|
| 29 |
+
"step": 1000
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 4.02,
|
| 33 |
+
"eval_loss": 0.3672347962856293,
|
| 34 |
+
"eval_runtime": 72.6811,
|
| 35 |
+
"eval_samples_per_second": 23.115,
|
| 36 |
+
"eval_steps_per_second": 2.889,
|
| 37 |
+
"eval_wer": 0.41520226035421404,
|
| 38 |
+
"step": 1000
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 6.02,
|
| 42 |
+
"learning_rate": 9.231839258114375e-05,
|
| 43 |
+
"loss": 0.305,
|
| 44 |
+
"step": 1500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 6.02,
|
| 48 |
+
"eval_loss": 0.2641327381134033,
|
| 49 |
+
"eval_runtime": 74.7155,
|
| 50 |
+
"eval_samples_per_second": 22.485,
|
| 51 |
+
"eval_steps_per_second": 2.811,
|
| 52 |
+
"eval_wer": 0.32127351664254705,
|
| 53 |
+
"step": 1500
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 8.03,
|
| 57 |
+
"learning_rate": 8.459041731066461e-05,
|
| 58 |
+
"loss": 0.1614,
|
| 59 |
+
"step": 2000
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 8.03,
|
| 63 |
+
"eval_loss": 0.2683059871196747,
|
| 64 |
+
"eval_runtime": 73.5962,
|
| 65 |
+
"eval_samples_per_second": 22.827,
|
| 66 |
+
"eval_steps_per_second": 2.853,
|
| 67 |
+
"eval_wer": 0.2949486596375164,
|
| 68 |
+
"step": 2000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 10.04,
|
| 72 |
+
"learning_rate": 7.686244204018547e-05,
|
| 73 |
+
"loss": 0.1096,
|
| 74 |
+
"step": 2500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 10.04,
|
| 78 |
+
"eval_loss": 0.25842466950416565,
|
| 79 |
+
"eval_runtime": 72.9953,
|
| 80 |
+
"eval_samples_per_second": 23.015,
|
| 81 |
+
"eval_steps_per_second": 2.877,
|
| 82 |
+
"eval_wer": 0.2761353455998897,
|
| 83 |
+
"step": 2500
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 12.05,
|
| 87 |
+
"learning_rate": 6.913446676970633e-05,
|
| 88 |
+
"loss": 0.0849,
|
| 89 |
+
"step": 3000
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 12.05,
|
| 93 |
+
"eval_loss": 0.2633412778377533,
|
| 94 |
+
"eval_runtime": 73.117,
|
| 95 |
+
"eval_samples_per_second": 22.977,
|
| 96 |
+
"eval_steps_per_second": 2.872,
|
| 97 |
+
"eval_wer": 0.26834814967955345,
|
| 98 |
+
"step": 3000
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 14.06,
|
| 102 |
+
"learning_rate": 6.14064914992272e-05,
|
| 103 |
+
"loss": 0.0666,
|
| 104 |
+
"step": 3500
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 14.06,
|
| 108 |
+
"eval_loss": 0.2584041357040405,
|
| 109 |
+
"eval_runtime": 72.8822,
|
| 110 |
+
"eval_samples_per_second": 23.051,
|
| 111 |
+
"eval_steps_per_second": 2.881,
|
| 112 |
+
"eval_wer": 0.2597339949004204,
|
| 113 |
+
"step": 3500
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 16.06,
|
| 117 |
+
"learning_rate": 5.3678516228748066e-05,
|
| 118 |
+
"loss": 0.0558,
|
| 119 |
+
"step": 4000
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 16.06,
|
| 123 |
+
"eval_loss": 0.25760215520858765,
|
| 124 |
+
"eval_runtime": 72.4376,
|
| 125 |
+
"eval_samples_per_second": 23.192,
|
| 126 |
+
"eval_steps_per_second": 2.899,
|
| 127 |
+
"eval_wer": 0.24939700916546068,
|
| 128 |
+
"step": 4000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 18.07,
|
| 132 |
+
"learning_rate": 4.5950540958268935e-05,
|
| 133 |
+
"loss": 0.0455,
|
| 134 |
+
"step": 4500
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 18.07,
|
| 138 |
+
"eval_loss": 0.2717011272907257,
|
| 139 |
+
"eval_runtime": 73.3321,
|
| 140 |
+
"eval_samples_per_second": 22.909,
|
| 141 |
+
"eval_steps_per_second": 2.864,
|
| 142 |
+
"eval_wer": 0.2531183240300462,
|
| 143 |
+
"step": 4500
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 20.08,
|
| 147 |
+
"learning_rate": 3.8222565687789804e-05,
|
| 148 |
+
"loss": 0.0422,
|
| 149 |
+
"step": 5000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 20.08,
|
| 153 |
+
"eval_loss": 0.29290544986724854,
|
| 154 |
+
"eval_runtime": 74.3543,
|
| 155 |
+
"eval_samples_per_second": 22.595,
|
| 156 |
+
"eval_steps_per_second": 2.824,
|
| 157 |
+
"eval_wer": 0.2513265798359865,
|
| 158 |
+
"step": 5000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 22.09,
|
| 162 |
+
"learning_rate": 3.0494590417310663e-05,
|
| 163 |
+
"loss": 0.0383,
|
| 164 |
+
"step": 5500
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 22.09,
|
| 168 |
+
"eval_loss": 0.27808326482772827,
|
| 169 |
+
"eval_runtime": 73.2045,
|
| 170 |
+
"eval_samples_per_second": 22.949,
|
| 171 |
+
"eval_steps_per_second": 2.869,
|
| 172 |
+
"eval_wer": 0.2480187444007994,
|
| 173 |
+
"step": 5500
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 24.1,
|
| 177 |
+
"learning_rate": 2.2766615146831533e-05,
|
| 178 |
+
"loss": 0.033,
|
| 179 |
+
"step": 6000
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 24.1,
|
| 183 |
+
"eval_loss": 0.2956600785255432,
|
| 184 |
+
"eval_runtime": 73.3873,
|
| 185 |
+
"eval_samples_per_second": 22.892,
|
| 186 |
+
"eval_steps_per_second": 2.862,
|
| 187 |
+
"eval_wer": 0.2503617945007236,
|
| 188 |
+
"step": 6000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 26.1,
|
| 192 |
+
"learning_rate": 1.5038639876352395e-05,
|
| 193 |
+
"loss": 0.0314,
|
| 194 |
+
"step": 6500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 26.1,
|
| 198 |
+
"eval_loss": 0.2865420877933502,
|
| 199 |
+
"eval_runtime": 73.2994,
|
| 200 |
+
"eval_samples_per_second": 22.92,
|
| 201 |
+
"eval_steps_per_second": 2.865,
|
| 202 |
+
"eval_wer": 0.24863896354489698,
|
| 203 |
+
"step": 6500
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 28.11,
|
| 207 |
+
"learning_rate": 7.310664605873262e-06,
|
| 208 |
+
"loss": 0.0308,
|
| 209 |
+
"step": 7000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 28.11,
|
| 213 |
+
"eval_loss": 0.29938989877700806,
|
| 214 |
+
"eval_runtime": 73.3049,
|
| 215 |
+
"eval_samples_per_second": 22.918,
|
| 216 |
+
"eval_steps_per_second": 2.865,
|
| 217 |
+
"eval_wer": 0.24732961201846876,
|
| 218 |
+
"step": 7000
|
| 219 |
+
}
|
| 220 |
+
],
|
| 221 |
+
"max_steps": 7470,
|
| 222 |
+
"num_train_epochs": 30,
|
| 223 |
+
"total_flos": 9.6467224304247e+18,
|
| 224 |
+
"trial_name": null,
|
| 225 |
+
"trial_params": null
|
| 226 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9951cfdb87834e9a623000e7c8338bb803c098d7ce35aa85419efee4dcb349c1
|
| 3 |
+
size 3387
|