Uploading trained model
Browse files- added_tokens.json +1 -0
- checkpoint-550/config.json +117 -0
- checkpoint-550/optimizer.pt +3 -0
- checkpoint-550/preprocessor_config.json +10 -0
- checkpoint-550/pytorch_model.bin +3 -0
- checkpoint-550/rng_state.pth +3 -0
- checkpoint-550/scaler.pt +3 -0
- checkpoint-550/scheduler.pt +3 -0
- checkpoint-550/trainer_state.json +472 -0
- checkpoint-550/training_args.bin +3 -0
- checkpoint-99/config.json +117 -0
- checkpoint-99/optimizer.pt +3 -0
- checkpoint-99/preprocessor_config.json +10 -0
- checkpoint-99/pytorch_model.bin +3 -0
- checkpoint-99/rng_state.pth +3 -0
- checkpoint-99/scaler.pt +3 -0
- checkpoint-99/scheduler.pt +3 -0
- checkpoint-99/trainer_state.json +97 -0
- checkpoint-99/training_args.bin +3 -0
- config.json +117 -0
- preprocessor_config.json +11 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +1 -0
- tokenizer_config.json +1 -0
- training_args.bin +3 -0
- vocab.json +1 -0
added_tokens.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"<s>": 43, "</s>": 44}
|
checkpoint-550/config.json
ADDED
|
@@ -0,0 +1,117 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "slplab/wav2vec2-xls-r-300m_phone-mfa_korean",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"adapter_kernel_size": 3,
|
| 5 |
+
"adapter_stride": 2,
|
| 6 |
+
"add_adapter": false,
|
| 7 |
+
"apply_spec_augment": true,
|
| 8 |
+
"architectures": [
|
| 9 |
+
"Wav2Vec2ForSpeechClassification"
|
| 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": 0.0,
|
| 55 |
+
"finetuning_task": "wav2vec2_clf",
|
| 56 |
+
"gradient_checkpointing": false,
|
| 57 |
+
"hidden_act": "gelu",
|
| 58 |
+
"hidden_dropout": 0.1,
|
| 59 |
+
"hidden_size": 1024,
|
| 60 |
+
"id2label": {
|
| 61 |
+
"0": "LABEL_0"
|
| 62 |
+
},
|
| 63 |
+
"initializer_range": 0.02,
|
| 64 |
+
"intermediate_size": 4096,
|
| 65 |
+
"label2id": {
|
| 66 |
+
"LABEL_0": 0
|
| 67 |
+
},
|
| 68 |
+
"layer_norm_eps": 1e-05,
|
| 69 |
+
"layerdrop": 0.1,
|
| 70 |
+
"mask_feature_length": 10,
|
| 71 |
+
"mask_feature_min_masks": 0,
|
| 72 |
+
"mask_feature_prob": 0.0,
|
| 73 |
+
"mask_time_length": 10,
|
| 74 |
+
"mask_time_min_masks": 2,
|
| 75 |
+
"mask_time_prob": 0.075,
|
| 76 |
+
"model_type": "wav2vec2",
|
| 77 |
+
"num_adapter_layers": 3,
|
| 78 |
+
"num_attention_heads": 16,
|
| 79 |
+
"num_codevector_groups": 2,
|
| 80 |
+
"num_codevectors_per_group": 320,
|
| 81 |
+
"num_conv_pos_embedding_groups": 16,
|
| 82 |
+
"num_conv_pos_embeddings": 128,
|
| 83 |
+
"num_feat_extract_layers": 7,
|
| 84 |
+
"num_hidden_layers": 24,
|
| 85 |
+
"num_negatives": 100,
|
| 86 |
+
"output_hidden_size": 1024,
|
| 87 |
+
"pad_token_id": 42,
|
| 88 |
+
"pooling_mode": "mean",
|
| 89 |
+
"problem_type": "regression",
|
| 90 |
+
"proj_codevector_dim": 768,
|
| 91 |
+
"tdnn_dilation": [
|
| 92 |
+
1,
|
| 93 |
+
2,
|
| 94 |
+
3,
|
| 95 |
+
1,
|
| 96 |
+
1
|
| 97 |
+
],
|
| 98 |
+
"tdnn_dim": [
|
| 99 |
+
512,
|
| 100 |
+
512,
|
| 101 |
+
512,
|
| 102 |
+
512,
|
| 103 |
+
1500
|
| 104 |
+
],
|
| 105 |
+
"tdnn_kernel": [
|
| 106 |
+
5,
|
| 107 |
+
3,
|
| 108 |
+
3,
|
| 109 |
+
1,
|
| 110 |
+
1
|
| 111 |
+
],
|
| 112 |
+
"torch_dtype": "float32",
|
| 113 |
+
"transformers_version": "4.19.3",
|
| 114 |
+
"use_weighted_layer_sum": false,
|
| 115 |
+
"vocab_size": 45,
|
| 116 |
+
"xvector_output_dim": 512
|
| 117 |
+
}
|
checkpoint-550/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0b06cc682f07e4dcf4ee4801f5cdb587ec58468fd89b8aa459aab642ffa305bc
|
| 3 |
+
size 2498465161
|
checkpoint-550/preprocessor_config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cache_dir": "/data2/excalibur12/.cache/huggingface/datasets",
|
| 3 |
+
"do_normalize": true,
|
| 4 |
+
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
| 5 |
+
"feature_size": 1,
|
| 6 |
+
"padding_side": "right",
|
| 7 |
+
"padding_value": 0.0,
|
| 8 |
+
"return_attention_mask": true,
|
| 9 |
+
"sampling_rate": 16000
|
| 10 |
+
}
|
checkpoint-550/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fc61e7cd85802dd3a58764549781397be6d31714c1a41b59e9327535093e2194
|
| 3 |
+
size 1266101869
|
checkpoint-550/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:46015840c7dc0498fdae4bd8dce131fc8fdb7beb4db6d28e7f511ed2c19d088f
|
| 3 |
+
size 14567
|
checkpoint-550/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1c33378e25f610e335ddba9ca3bc1885deef22bd7de80474020b185e11026b99
|
| 3 |
+
size 559
|
checkpoint-550/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:37e8d607b82b4eb73fe403c69aa9e37348793e30ec147c856a28d72e5b4d7d9b
|
| 3 |
+
size 623
|
checkpoint-550/trainer_state.json
ADDED
|
@@ -0,0 +1,472 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.30745795369148254,
|
| 3 |
+
"best_model_checkpoint": "asd_pronunciation_w2v_xlsr-reg/checkpoint-99",
|
| 4 |
+
"epoch": 49.977777777777774,
|
| 5 |
+
"global_step": 550,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.98,
|
| 12 |
+
"eval_loss": 0.8269791603088379,
|
| 13 |
+
"eval_mse": 0.8274638056755066,
|
| 14 |
+
"eval_runtime": 67.3404,
|
| 15 |
+
"eval_samples_per_second": 31.497,
|
| 16 |
+
"eval_steps_per_second": 0.995,
|
| 17 |
+
"step": 11
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 1.98,
|
| 21 |
+
"eval_loss": 0.4631172716617584,
|
| 22 |
+
"eval_mse": 0.46300947666168213,
|
| 23 |
+
"eval_runtime": 67.2163,
|
| 24 |
+
"eval_samples_per_second": 31.555,
|
| 25 |
+
"eval_steps_per_second": 0.997,
|
| 26 |
+
"step": 22
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 2.98,
|
| 30 |
+
"eval_loss": 0.351409912109375,
|
| 31 |
+
"eval_mse": 0.3515050709247589,
|
| 32 |
+
"eval_runtime": 67.5563,
|
| 33 |
+
"eval_samples_per_second": 31.396,
|
| 34 |
+
"eval_steps_per_second": 0.992,
|
| 35 |
+
"step": 33
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 3.98,
|
| 39 |
+
"eval_loss": 0.3585782051086426,
|
| 40 |
+
"eval_mse": 0.35852691531181335,
|
| 41 |
+
"eval_runtime": 69.7976,
|
| 42 |
+
"eval_samples_per_second": 30.388,
|
| 43 |
+
"eval_steps_per_second": 0.96,
|
| 44 |
+
"step": 44
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 4.98,
|
| 48 |
+
"eval_loss": 0.4412296712398529,
|
| 49 |
+
"eval_mse": 0.4412075877189636,
|
| 50 |
+
"eval_runtime": 79.1138,
|
| 51 |
+
"eval_samples_per_second": 26.809,
|
| 52 |
+
"eval_steps_per_second": 0.847,
|
| 53 |
+
"step": 55
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 5.98,
|
| 57 |
+
"eval_loss": 0.430486798286438,
|
| 58 |
+
"eval_mse": 0.43049314618110657,
|
| 59 |
+
"eval_runtime": 72.5547,
|
| 60 |
+
"eval_samples_per_second": 29.233,
|
| 61 |
+
"eval_steps_per_second": 0.923,
|
| 62 |
+
"step": 66
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 6.98,
|
| 66 |
+
"eval_loss": 0.42916765809059143,
|
| 67 |
+
"eval_mse": 0.4291659891605377,
|
| 68 |
+
"eval_runtime": 95.5528,
|
| 69 |
+
"eval_samples_per_second": 22.197,
|
| 70 |
+
"eval_steps_per_second": 0.701,
|
| 71 |
+
"step": 77
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 7.98,
|
| 75 |
+
"eval_loss": 0.4343053996562958,
|
| 76 |
+
"eval_mse": 0.43428245186805725,
|
| 77 |
+
"eval_runtime": 66.5007,
|
| 78 |
+
"eval_samples_per_second": 31.894,
|
| 79 |
+
"eval_steps_per_second": 1.008,
|
| 80 |
+
"step": 88
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 8.98,
|
| 84 |
+
"eval_loss": 0.30745795369148254,
|
| 85 |
+
"eval_mse": 0.3074318766593933,
|
| 86 |
+
"eval_runtime": 66.2359,
|
| 87 |
+
"eval_samples_per_second": 32.022,
|
| 88 |
+
"eval_steps_per_second": 1.012,
|
| 89 |
+
"step": 99
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 9.98,
|
| 93 |
+
"eval_loss": 0.42939332127571106,
|
| 94 |
+
"eval_mse": 0.4293690323829651,
|
| 95 |
+
"eval_runtime": 66.3865,
|
| 96 |
+
"eval_samples_per_second": 31.949,
|
| 97 |
+
"eval_steps_per_second": 1.009,
|
| 98 |
+
"step": 110
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 10.98,
|
| 102 |
+
"eval_loss": 0.3727492094039917,
|
| 103 |
+
"eval_mse": 0.3727482855319977,
|
| 104 |
+
"eval_runtime": 66.8857,
|
| 105 |
+
"eval_samples_per_second": 31.711,
|
| 106 |
+
"eval_steps_per_second": 1.002,
|
| 107 |
+
"step": 121
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 11.98,
|
| 111 |
+
"eval_loss": 0.42278197407722473,
|
| 112 |
+
"eval_mse": 0.42277276515960693,
|
| 113 |
+
"eval_runtime": 65.5197,
|
| 114 |
+
"eval_samples_per_second": 32.372,
|
| 115 |
+
"eval_steps_per_second": 1.023,
|
| 116 |
+
"step": 132
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 12.98,
|
| 120 |
+
"eval_loss": 0.4014572203159332,
|
| 121 |
+
"eval_mse": 0.40141549706459045,
|
| 122 |
+
"eval_runtime": 65.6627,
|
| 123 |
+
"eval_samples_per_second": 32.301,
|
| 124 |
+
"eval_steps_per_second": 1.02,
|
| 125 |
+
"step": 143
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 13.98,
|
| 129 |
+
"eval_loss": 0.37818050384521484,
|
| 130 |
+
"eval_mse": 0.37818947434425354,
|
| 131 |
+
"eval_runtime": 66.0031,
|
| 132 |
+
"eval_samples_per_second": 32.135,
|
| 133 |
+
"eval_steps_per_second": 1.015,
|
| 134 |
+
"step": 154
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 14.98,
|
| 138 |
+
"eval_loss": 0.4358225166797638,
|
| 139 |
+
"eval_mse": 0.43578916788101196,
|
| 140 |
+
"eval_runtime": 65.7571,
|
| 141 |
+
"eval_samples_per_second": 32.255,
|
| 142 |
+
"eval_steps_per_second": 1.019,
|
| 143 |
+
"step": 165
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 15.98,
|
| 147 |
+
"eval_loss": 0.36039263010025024,
|
| 148 |
+
"eval_mse": 0.3603877127170563,
|
| 149 |
+
"eval_runtime": 67.6323,
|
| 150 |
+
"eval_samples_per_second": 31.361,
|
| 151 |
+
"eval_steps_per_second": 0.991,
|
| 152 |
+
"step": 176
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 16.98,
|
| 156 |
+
"eval_loss": 0.40446653962135315,
|
| 157 |
+
"eval_mse": 0.40447959303855896,
|
| 158 |
+
"eval_runtime": 66.1021,
|
| 159 |
+
"eval_samples_per_second": 32.087,
|
| 160 |
+
"eval_steps_per_second": 1.014,
|
| 161 |
+
"step": 187
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 17.98,
|
| 165 |
+
"eval_loss": 0.3638509213924408,
|
| 166 |
+
"eval_mse": 0.3638397455215454,
|
| 167 |
+
"eval_runtime": 66.1765,
|
| 168 |
+
"eval_samples_per_second": 32.051,
|
| 169 |
+
"eval_steps_per_second": 1.012,
|
| 170 |
+
"step": 198
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 18.98,
|
| 174 |
+
"eval_loss": 0.4001483619213104,
|
| 175 |
+
"eval_mse": 0.400104284286499,
|
| 176 |
+
"eval_runtime": 66.5235,
|
| 177 |
+
"eval_samples_per_second": 31.883,
|
| 178 |
+
"eval_steps_per_second": 1.007,
|
| 179 |
+
"step": 209
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 19.98,
|
| 183 |
+
"eval_loss": 0.3836788535118103,
|
| 184 |
+
"eval_mse": 0.38362112641334534,
|
| 185 |
+
"eval_runtime": 65.5397,
|
| 186 |
+
"eval_samples_per_second": 32.362,
|
| 187 |
+
"eval_steps_per_second": 1.022,
|
| 188 |
+
"step": 220
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 20.98,
|
| 192 |
+
"eval_loss": 0.3813478350639343,
|
| 193 |
+
"eval_mse": 0.3813202679157257,
|
| 194 |
+
"eval_runtime": 67.0847,
|
| 195 |
+
"eval_samples_per_second": 31.617,
|
| 196 |
+
"eval_steps_per_second": 0.999,
|
| 197 |
+
"step": 231
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 21.98,
|
| 201 |
+
"eval_loss": 0.3911483883857727,
|
| 202 |
+
"eval_mse": 0.39113306999206543,
|
| 203 |
+
"eval_runtime": 66.4329,
|
| 204 |
+
"eval_samples_per_second": 31.927,
|
| 205 |
+
"eval_steps_per_second": 1.009,
|
| 206 |
+
"step": 242
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 22.98,
|
| 210 |
+
"eval_loss": 0.43994417786598206,
|
| 211 |
+
"eval_mse": 0.4399244487285614,
|
| 212 |
+
"eval_runtime": 66.1358,
|
| 213 |
+
"eval_samples_per_second": 32.07,
|
| 214 |
+
"eval_steps_per_second": 1.013,
|
| 215 |
+
"step": 253
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 23.98,
|
| 219 |
+
"eval_loss": 0.4228975772857666,
|
| 220 |
+
"eval_mse": 0.42288002371788025,
|
| 221 |
+
"eval_runtime": 66.6907,
|
| 222 |
+
"eval_samples_per_second": 31.804,
|
| 223 |
+
"eval_steps_per_second": 1.005,
|
| 224 |
+
"step": 264
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 24.98,
|
| 228 |
+
"eval_loss": 0.4192221462726593,
|
| 229 |
+
"eval_mse": 0.4191807508468628,
|
| 230 |
+
"eval_runtime": 65.8241,
|
| 231 |
+
"eval_samples_per_second": 32.222,
|
| 232 |
+
"eval_steps_per_second": 1.018,
|
| 233 |
+
"step": 275
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 25.98,
|
| 237 |
+
"eval_loss": 0.40975797176361084,
|
| 238 |
+
"eval_mse": 0.4097472131252289,
|
| 239 |
+
"eval_runtime": 67.5766,
|
| 240 |
+
"eval_samples_per_second": 31.387,
|
| 241 |
+
"eval_steps_per_second": 0.991,
|
| 242 |
+
"step": 286
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 26.98,
|
| 246 |
+
"eval_loss": 0.3760901093482971,
|
| 247 |
+
"eval_mse": 0.37607377767562866,
|
| 248 |
+
"eval_runtime": 65.9157,
|
| 249 |
+
"eval_samples_per_second": 32.177,
|
| 250 |
+
"eval_steps_per_second": 1.016,
|
| 251 |
+
"step": 297
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 27.98,
|
| 255 |
+
"eval_loss": 0.389096200466156,
|
| 256 |
+
"eval_mse": 0.3890584111213684,
|
| 257 |
+
"eval_runtime": 65.956,
|
| 258 |
+
"eval_samples_per_second": 32.158,
|
| 259 |
+
"eval_steps_per_second": 1.016,
|
| 260 |
+
"step": 308
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 28.98,
|
| 264 |
+
"eval_loss": 0.4243176579475403,
|
| 265 |
+
"eval_mse": 0.4242975115776062,
|
| 266 |
+
"eval_runtime": 66.4794,
|
| 267 |
+
"eval_samples_per_second": 31.905,
|
| 268 |
+
"eval_steps_per_second": 1.008,
|
| 269 |
+
"step": 319
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 29.98,
|
| 273 |
+
"eval_loss": 0.4235914349555969,
|
| 274 |
+
"eval_mse": 0.4235744774341583,
|
| 275 |
+
"eval_runtime": 66.2277,
|
| 276 |
+
"eval_samples_per_second": 32.026,
|
| 277 |
+
"eval_steps_per_second": 1.012,
|
| 278 |
+
"step": 330
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 30.98,
|
| 282 |
+
"eval_loss": 0.4235081076622009,
|
| 283 |
+
"eval_mse": 0.42347782850265503,
|
| 284 |
+
"eval_runtime": 66.6381,
|
| 285 |
+
"eval_samples_per_second": 31.829,
|
| 286 |
+
"eval_steps_per_second": 1.005,
|
| 287 |
+
"step": 341
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 31.98,
|
| 291 |
+
"eval_loss": 0.4236636757850647,
|
| 292 |
+
"eval_mse": 0.4236546456813812,
|
| 293 |
+
"eval_runtime": 65.6032,
|
| 294 |
+
"eval_samples_per_second": 32.331,
|
| 295 |
+
"eval_steps_per_second": 1.021,
|
| 296 |
+
"step": 352
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 32.98,
|
| 300 |
+
"eval_loss": 0.4269878566265106,
|
| 301 |
+
"eval_mse": 0.4269687533378601,
|
| 302 |
+
"eval_runtime": 65.9665,
|
| 303 |
+
"eval_samples_per_second": 32.153,
|
| 304 |
+
"eval_steps_per_second": 1.016,
|
| 305 |
+
"step": 363
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 33.98,
|
| 309 |
+
"eval_loss": 0.39142194390296936,
|
| 310 |
+
"eval_mse": 0.3913804888725281,
|
| 311 |
+
"eval_runtime": 66.6217,
|
| 312 |
+
"eval_samples_per_second": 31.836,
|
| 313 |
+
"eval_steps_per_second": 1.006,
|
| 314 |
+
"step": 374
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 34.98,
|
| 318 |
+
"eval_loss": 0.3899790942668915,
|
| 319 |
+
"eval_mse": 0.38995301723480225,
|
| 320 |
+
"eval_runtime": 65.771,
|
| 321 |
+
"eval_samples_per_second": 32.248,
|
| 322 |
+
"eval_steps_per_second": 1.019,
|
| 323 |
+
"step": 385
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 35.98,
|
| 327 |
+
"eval_loss": 0.4031297564506531,
|
| 328 |
+
"eval_mse": 0.4031302034854889,
|
| 329 |
+
"eval_runtime": 68.033,
|
| 330 |
+
"eval_samples_per_second": 31.176,
|
| 331 |
+
"eval_steps_per_second": 0.985,
|
| 332 |
+
"step": 396
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 36.98,
|
| 336 |
+
"eval_loss": 0.373826265335083,
|
| 337 |
+
"eval_mse": 0.3738201856613159,
|
| 338 |
+
"eval_runtime": 66.1644,
|
| 339 |
+
"eval_samples_per_second": 32.057,
|
| 340 |
+
"eval_steps_per_second": 1.013,
|
| 341 |
+
"step": 407
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 37.98,
|
| 345 |
+
"eval_loss": 0.37409740686416626,
|
| 346 |
+
"eval_mse": 0.37407544255256653,
|
| 347 |
+
"eval_runtime": 65.7007,
|
| 348 |
+
"eval_samples_per_second": 32.283,
|
| 349 |
+
"eval_steps_per_second": 1.02,
|
| 350 |
+
"step": 418
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 38.98,
|
| 354 |
+
"eval_loss": 0.41098639369010925,
|
| 355 |
+
"eval_mse": 0.4109634459018707,
|
| 356 |
+
"eval_runtime": 65.8334,
|
| 357 |
+
"eval_samples_per_second": 32.218,
|
| 358 |
+
"eval_steps_per_second": 1.018,
|
| 359 |
+
"step": 429
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 39.98,
|
| 363 |
+
"eval_loss": 0.38580140471458435,
|
| 364 |
+
"eval_mse": 0.3857785165309906,
|
| 365 |
+
"eval_runtime": 65.912,
|
| 366 |
+
"eval_samples_per_second": 32.179,
|
| 367 |
+
"eval_steps_per_second": 1.017,
|
| 368 |
+
"step": 440
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 40.98,
|
| 372 |
+
"eval_loss": 0.40168315172195435,
|
| 373 |
+
"eval_mse": 0.4016563296318054,
|
| 374 |
+
"eval_runtime": 67.322,
|
| 375 |
+
"eval_samples_per_second": 31.505,
|
| 376 |
+
"eval_steps_per_second": 0.995,
|
| 377 |
+
"step": 451
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 41.98,
|
| 381 |
+
"eval_loss": 0.3875749111175537,
|
| 382 |
+
"eval_mse": 0.3875587284564972,
|
| 383 |
+
"eval_runtime": 65.9445,
|
| 384 |
+
"eval_samples_per_second": 32.163,
|
| 385 |
+
"eval_steps_per_second": 1.016,
|
| 386 |
+
"step": 462
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 42.98,
|
| 390 |
+
"eval_loss": 0.401607483625412,
|
| 391 |
+
"eval_mse": 0.4015834629535675,
|
| 392 |
+
"eval_runtime": 66.235,
|
| 393 |
+
"eval_samples_per_second": 32.022,
|
| 394 |
+
"eval_steps_per_second": 1.012,
|
| 395 |
+
"step": 473
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 43.98,
|
| 399 |
+
"eval_loss": 0.3939042389392853,
|
| 400 |
+
"eval_mse": 0.3938945233821869,
|
| 401 |
+
"eval_runtime": 66.0054,
|
| 402 |
+
"eval_samples_per_second": 32.134,
|
| 403 |
+
"eval_steps_per_second": 1.015,
|
| 404 |
+
"step": 484
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 44.98,
|
| 408 |
+
"eval_loss": 0.40803390741348267,
|
| 409 |
+
"eval_mse": 0.40802931785583496,
|
| 410 |
+
"eval_runtime": 66.0842,
|
| 411 |
+
"eval_samples_per_second": 32.095,
|
| 412 |
+
"eval_steps_per_second": 1.014,
|
| 413 |
+
"step": 495
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 45.44,
|
| 417 |
+
"learning_rate": 3.03030303030303e-05,
|
| 418 |
+
"loss": 0.4034,
|
| 419 |
+
"step": 500
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 45.98,
|
| 423 |
+
"eval_loss": 0.38819119334220886,
|
| 424 |
+
"eval_mse": 0.38818415999412537,
|
| 425 |
+
"eval_runtime": 67.7262,
|
| 426 |
+
"eval_samples_per_second": 31.317,
|
| 427 |
+
"eval_steps_per_second": 0.989,
|
| 428 |
+
"step": 506
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 46.98,
|
| 432 |
+
"eval_loss": 0.4062108099460602,
|
| 433 |
+
"eval_mse": 0.4061962068080902,
|
| 434 |
+
"eval_runtime": 63.6037,
|
| 435 |
+
"eval_samples_per_second": 33.347,
|
| 436 |
+
"eval_steps_per_second": 1.053,
|
| 437 |
+
"step": 517
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 47.98,
|
| 441 |
+
"eval_loss": 0.38834279775619507,
|
| 442 |
+
"eval_mse": 0.38832658529281616,
|
| 443 |
+
"eval_runtime": 64.2256,
|
| 444 |
+
"eval_samples_per_second": 33.024,
|
| 445 |
+
"eval_steps_per_second": 1.043,
|
| 446 |
+
"step": 528
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 48.98,
|
| 450 |
+
"eval_loss": 0.3969601094722748,
|
| 451 |
+
"eval_mse": 0.3969435691833496,
|
| 452 |
+
"eval_runtime": 63.6409,
|
| 453 |
+
"eval_samples_per_second": 33.328,
|
| 454 |
+
"eval_steps_per_second": 1.053,
|
| 455 |
+
"step": 539
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 49.98,
|
| 459 |
+
"eval_loss": 0.3963707685470581,
|
| 460 |
+
"eval_mse": 0.3963526785373688,
|
| 461 |
+
"eval_runtime": 63.4841,
|
| 462 |
+
"eval_samples_per_second": 33.41,
|
| 463 |
+
"eval_steps_per_second": 1.055,
|
| 464 |
+
"step": 550
|
| 465 |
+
}
|
| 466 |
+
],
|
| 467 |
+
"max_steps": 550,
|
| 468 |
+
"num_train_epochs": 50,
|
| 469 |
+
"total_flos": 1.603121844978697e+20,
|
| 470 |
+
"trial_name": null,
|
| 471 |
+
"trial_params": null
|
| 472 |
+
}
|
checkpoint-550/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d6d18a7837d69e5d09a40f73faf5071bd1c07927dac0c1c18d211646e1533cbb
|
| 3 |
+
size 3247
|
checkpoint-99/config.json
ADDED
|
@@ -0,0 +1,117 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "slplab/wav2vec2-xls-r-300m_phone-mfa_korean",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"adapter_kernel_size": 3,
|
| 5 |
+
"adapter_stride": 2,
|
| 6 |
+
"add_adapter": false,
|
| 7 |
+
"apply_spec_augment": true,
|
| 8 |
+
"architectures": [
|
| 9 |
+
"Wav2Vec2ForSpeechClassification"
|
| 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": 0.0,
|
| 55 |
+
"finetuning_task": "wav2vec2_clf",
|
| 56 |
+
"gradient_checkpointing": false,
|
| 57 |
+
"hidden_act": "gelu",
|
| 58 |
+
"hidden_dropout": 0.1,
|
| 59 |
+
"hidden_size": 1024,
|
| 60 |
+
"id2label": {
|
| 61 |
+
"0": "LABEL_0"
|
| 62 |
+
},
|
| 63 |
+
"initializer_range": 0.02,
|
| 64 |
+
"intermediate_size": 4096,
|
| 65 |
+
"label2id": {
|
| 66 |
+
"LABEL_0": 0
|
| 67 |
+
},
|
| 68 |
+
"layer_norm_eps": 1e-05,
|
| 69 |
+
"layerdrop": 0.1,
|
| 70 |
+
"mask_feature_length": 10,
|
| 71 |
+
"mask_feature_min_masks": 0,
|
| 72 |
+
"mask_feature_prob": 0.0,
|
| 73 |
+
"mask_time_length": 10,
|
| 74 |
+
"mask_time_min_masks": 2,
|
| 75 |
+
"mask_time_prob": 0.075,
|
| 76 |
+
"model_type": "wav2vec2",
|
| 77 |
+
"num_adapter_layers": 3,
|
| 78 |
+
"num_attention_heads": 16,
|
| 79 |
+
"num_codevector_groups": 2,
|
| 80 |
+
"num_codevectors_per_group": 320,
|
| 81 |
+
"num_conv_pos_embedding_groups": 16,
|
| 82 |
+
"num_conv_pos_embeddings": 128,
|
| 83 |
+
"num_feat_extract_layers": 7,
|
| 84 |
+
"num_hidden_layers": 24,
|
| 85 |
+
"num_negatives": 100,
|
| 86 |
+
"output_hidden_size": 1024,
|
| 87 |
+
"pad_token_id": 42,
|
| 88 |
+
"pooling_mode": "mean",
|
| 89 |
+
"problem_type": "regression",
|
| 90 |
+
"proj_codevector_dim": 768,
|
| 91 |
+
"tdnn_dilation": [
|
| 92 |
+
1,
|
| 93 |
+
2,
|
| 94 |
+
3,
|
| 95 |
+
1,
|
| 96 |
+
1
|
| 97 |
+
],
|
| 98 |
+
"tdnn_dim": [
|
| 99 |
+
512,
|
| 100 |
+
512,
|
| 101 |
+
512,
|
| 102 |
+
512,
|
| 103 |
+
1500
|
| 104 |
+
],
|
| 105 |
+
"tdnn_kernel": [
|
| 106 |
+
5,
|
| 107 |
+
3,
|
| 108 |
+
3,
|
| 109 |
+
1,
|
| 110 |
+
1
|
| 111 |
+
],
|
| 112 |
+
"torch_dtype": "float32",
|
| 113 |
+
"transformers_version": "4.19.3",
|
| 114 |
+
"use_weighted_layer_sum": false,
|
| 115 |
+
"vocab_size": 45,
|
| 116 |
+
"xvector_output_dim": 512
|
| 117 |
+
}
|
checkpoint-99/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e45b75190c0075591cb83659bea8786524011d16b66bc51030ee67788dcfce10
|
| 3 |
+
size 2498464777
|
checkpoint-99/preprocessor_config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cache_dir": "/data2/excalibur12/.cache/huggingface/datasets",
|
| 3 |
+
"do_normalize": true,
|
| 4 |
+
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
| 5 |
+
"feature_size": 1,
|
| 6 |
+
"padding_side": "right",
|
| 7 |
+
"padding_value": 0.0,
|
| 8 |
+
"return_attention_mask": true,
|
| 9 |
+
"sampling_rate": 16000
|
| 10 |
+
}
|
checkpoint-99/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0e47fa51983f1f4f6c591d3b181d5a8bb4d4d7b57224ffbc4f2dae26521d1b84
|
| 3 |
+
size 1266101869
|
checkpoint-99/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e53ff83bc5a0f7e3f45273d2de1bf2192e1300b343841603a2b096df6ad8c127
|
| 3 |
+
size 14567
|
checkpoint-99/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:78a083ceb88b83ff7ed1f0adf6d62580b8ceb7d9c17ad4684f511424e289c436
|
| 3 |
+
size 559
|
checkpoint-99/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0de4bd42e78a35b7fa3b4618172ba102f4f2a9be54789adf5790bde4d76e585e
|
| 3 |
+
size 623
|
checkpoint-99/trainer_state.json
ADDED
|
@@ -0,0 +1,97 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.30745795369148254,
|
| 3 |
+
"best_model_checkpoint": "asd_pronunciation_w2v_xlsr-reg/checkpoint-99",
|
| 4 |
+
"epoch": 8.977777777777778,
|
| 5 |
+
"global_step": 99,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.98,
|
| 12 |
+
"eval_loss": 0.8269791603088379,
|
| 13 |
+
"eval_mse": 0.8274638056755066,
|
| 14 |
+
"eval_runtime": 67.3404,
|
| 15 |
+
"eval_samples_per_second": 31.497,
|
| 16 |
+
"eval_steps_per_second": 0.995,
|
| 17 |
+
"step": 11
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 1.98,
|
| 21 |
+
"eval_loss": 0.4631172716617584,
|
| 22 |
+
"eval_mse": 0.46300947666168213,
|
| 23 |
+
"eval_runtime": 67.2163,
|
| 24 |
+
"eval_samples_per_second": 31.555,
|
| 25 |
+
"eval_steps_per_second": 0.997,
|
| 26 |
+
"step": 22
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 2.98,
|
| 30 |
+
"eval_loss": 0.351409912109375,
|
| 31 |
+
"eval_mse": 0.3515050709247589,
|
| 32 |
+
"eval_runtime": 67.5563,
|
| 33 |
+
"eval_samples_per_second": 31.396,
|
| 34 |
+
"eval_steps_per_second": 0.992,
|
| 35 |
+
"step": 33
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 3.98,
|
| 39 |
+
"eval_loss": 0.3585782051086426,
|
| 40 |
+
"eval_mse": 0.35852691531181335,
|
| 41 |
+
"eval_runtime": 69.7976,
|
| 42 |
+
"eval_samples_per_second": 30.388,
|
| 43 |
+
"eval_steps_per_second": 0.96,
|
| 44 |
+
"step": 44
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 4.98,
|
| 48 |
+
"eval_loss": 0.4412296712398529,
|
| 49 |
+
"eval_mse": 0.4412075877189636,
|
| 50 |
+
"eval_runtime": 79.1138,
|
| 51 |
+
"eval_samples_per_second": 26.809,
|
| 52 |
+
"eval_steps_per_second": 0.847,
|
| 53 |
+
"step": 55
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 5.98,
|
| 57 |
+
"eval_loss": 0.430486798286438,
|
| 58 |
+
"eval_mse": 0.43049314618110657,
|
| 59 |
+
"eval_runtime": 72.5547,
|
| 60 |
+
"eval_samples_per_second": 29.233,
|
| 61 |
+
"eval_steps_per_second": 0.923,
|
| 62 |
+
"step": 66
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 6.98,
|
| 66 |
+
"eval_loss": 0.42916765809059143,
|
| 67 |
+
"eval_mse": 0.4291659891605377,
|
| 68 |
+
"eval_runtime": 95.5528,
|
| 69 |
+
"eval_samples_per_second": 22.197,
|
| 70 |
+
"eval_steps_per_second": 0.701,
|
| 71 |
+
"step": 77
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 7.98,
|
| 75 |
+
"eval_loss": 0.4343053996562958,
|
| 76 |
+
"eval_mse": 0.43428245186805725,
|
| 77 |
+
"eval_runtime": 66.5007,
|
| 78 |
+
"eval_samples_per_second": 31.894,
|
| 79 |
+
"eval_steps_per_second": 1.008,
|
| 80 |
+
"step": 88
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 8.98,
|
| 84 |
+
"eval_loss": 0.30745795369148254,
|
| 85 |
+
"eval_mse": 0.3074318766593933,
|
| 86 |
+
"eval_runtime": 66.2359,
|
| 87 |
+
"eval_samples_per_second": 32.022,
|
| 88 |
+
"eval_steps_per_second": 1.012,
|
| 89 |
+
"step": 99
|
| 90 |
+
}
|
| 91 |
+
],
|
| 92 |
+
"max_steps": 550,
|
| 93 |
+
"num_train_epochs": 50,
|
| 94 |
+
"total_flos": 2.8846380565942047e+19,
|
| 95 |
+
"trial_name": null,
|
| 96 |
+
"trial_params": null
|
| 97 |
+
}
|
checkpoint-99/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d6d18a7837d69e5d09a40f73faf5071bd1c07927dac0c1c18d211646e1533cbb
|
| 3 |
+
size 3247
|
config.json
ADDED
|
@@ -0,0 +1,117 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "slplab/wav2vec2-xls-r-300m_phone-mfa_korean",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"adapter_kernel_size": 3,
|
| 5 |
+
"adapter_stride": 2,
|
| 6 |
+
"add_adapter": false,
|
| 7 |
+
"apply_spec_augment": true,
|
| 8 |
+
"architectures": [
|
| 9 |
+
"Wav2Vec2ForSpeechClassification"
|
| 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": 0.0,
|
| 55 |
+
"finetuning_task": "wav2vec2_clf",
|
| 56 |
+
"gradient_checkpointing": false,
|
| 57 |
+
"hidden_act": "gelu",
|
| 58 |
+
"hidden_dropout": 0.1,
|
| 59 |
+
"hidden_size": 1024,
|
| 60 |
+
"id2label": {
|
| 61 |
+
"0": "LABEL_0"
|
| 62 |
+
},
|
| 63 |
+
"initializer_range": 0.02,
|
| 64 |
+
"intermediate_size": 4096,
|
| 65 |
+
"label2id": {
|
| 66 |
+
"LABEL_0": 0
|
| 67 |
+
},
|
| 68 |
+
"layer_norm_eps": 1e-05,
|
| 69 |
+
"layerdrop": 0.1,
|
| 70 |
+
"mask_feature_length": 10,
|
| 71 |
+
"mask_feature_min_masks": 0,
|
| 72 |
+
"mask_feature_prob": 0.0,
|
| 73 |
+
"mask_time_length": 10,
|
| 74 |
+
"mask_time_min_masks": 2,
|
| 75 |
+
"mask_time_prob": 0.075,
|
| 76 |
+
"model_type": "wav2vec2",
|
| 77 |
+
"num_adapter_layers": 3,
|
| 78 |
+
"num_attention_heads": 16,
|
| 79 |
+
"num_codevector_groups": 2,
|
| 80 |
+
"num_codevectors_per_group": 320,
|
| 81 |
+
"num_conv_pos_embedding_groups": 16,
|
| 82 |
+
"num_conv_pos_embeddings": 128,
|
| 83 |
+
"num_feat_extract_layers": 7,
|
| 84 |
+
"num_hidden_layers": 24,
|
| 85 |
+
"num_negatives": 100,
|
| 86 |
+
"output_hidden_size": 1024,
|
| 87 |
+
"pad_token_id": 42,
|
| 88 |
+
"pooling_mode": "mean",
|
| 89 |
+
"problem_type": "regression",
|
| 90 |
+
"proj_codevector_dim": 768,
|
| 91 |
+
"tdnn_dilation": [
|
| 92 |
+
1,
|
| 93 |
+
2,
|
| 94 |
+
3,
|
| 95 |
+
1,
|
| 96 |
+
1
|
| 97 |
+
],
|
| 98 |
+
"tdnn_dim": [
|
| 99 |
+
512,
|
| 100 |
+
512,
|
| 101 |
+
512,
|
| 102 |
+
512,
|
| 103 |
+
1500
|
| 104 |
+
],
|
| 105 |
+
"tdnn_kernel": [
|
| 106 |
+
5,
|
| 107 |
+
3,
|
| 108 |
+
3,
|
| 109 |
+
1,
|
| 110 |
+
1
|
| 111 |
+
],
|
| 112 |
+
"torch_dtype": "float32",
|
| 113 |
+
"transformers_version": "4.19.3",
|
| 114 |
+
"use_weighted_layer_sum": false,
|
| 115 |
+
"vocab_size": 45,
|
| 116 |
+
"xvector_output_dim": 512
|
| 117 |
+
}
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cache_dir": "/data2/excalibur12/.cache/huggingface/datasets",
|
| 3 |
+
"do_normalize": true,
|
| 4 |
+
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
| 5 |
+
"feature_size": 1,
|
| 6 |
+
"padding_side": "right",
|
| 7 |
+
"padding_value": 0.0,
|
| 8 |
+
"processor_class": "Wav2Vec2Processor",
|
| 9 |
+
"return_attention_mask": true,
|
| 10 |
+
"sampling_rate": 16000
|
| 11 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0e47fa51983f1f4f6c591d3b181d5a8bb4d4d7b57224ffbc4f2dae26521d1b84
|
| 3 |
+
size 1266101869
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "[UNK]", "pad_token": "[PAD]", "additional_special_tokens": [{"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}]}
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"unk_token": "[UNK]", "bos_token": "<s>", "eos_token": "</s>", "pad_token": "[PAD]", "word_delimiter_token": null, "phone_delimiter_token": "|", "do_phonemize": false, "phonemizer_lang": "en-us", "phonemizer_backend": "espeak", "name_or_path": "slplab/wav2vec2-xls-r-300m_phone-mfa_korean", "special_tokens_map_file": null, "tokenizer_class": "Wav2Vec2PhonemeCTCTokenizer", "processor_class": "Wav2Vec2Processor"}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d6d18a7837d69e5d09a40f73faf5071bd1c07927dac0c1c18d211646e1533cbb
|
| 3 |
+
size 3247
|
vocab.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"A": 0, "B": 1, "BB": 2, "CHh": 3, "D": 4, "DD": 5, "E": 6, "EO": 7, "EU": 8, "G": 9, "GG": 10, "H": 11, "I": 12, "J": 13, "JJ": 14, "Kh": 15, "L": 16, "M": 17, "N": 18, "NG": 19, "O": 20, "Ph": 21, "R": 22, "S": 23, "SS": 24, "Th": 25, "U": 26, "[PAD]": 42, "[UNK]": 41, "euI": 27, "iA": 28, "iE": 29, "iEO": 30, "iO": 31, "iU": 32, "k": 33, "oA": 34, "oE": 35, "p": 36, "t": 37, "uEO": 38, "uI": 39, "|": 40}
|