Upload 10 files
Browse files- config.json +117 -0
- optimizer.pt +3 -0
- preprocessor_config.json +9 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- tokenizer_config.json +13 -0
- trainer_state.json +520 -0
- training_args.bin +3 -0
- vocab.json +30 -0
config.json
ADDED
|
@@ -0,0 +1,117 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "facebook/wav2vec2-large-xlsr-53",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"adapter_attn_dim": null,
|
| 5 |
+
"adapter_kernel_size": 3,
|
| 6 |
+
"adapter_stride": 2,
|
| 7 |
+
"add_adapter": false,
|
| 8 |
+
"apply_spec_augment": true,
|
| 9 |
+
"architectures": [
|
| 10 |
+
"Wav2Vec2ForCTC"
|
| 11 |
+
],
|
| 12 |
+
"attention_dropout": 0.0,
|
| 13 |
+
"bos_token_id": 1,
|
| 14 |
+
"classifier_proj_size": 256,
|
| 15 |
+
"codevector_dim": 768,
|
| 16 |
+
"contrastive_logits_temperature": 0.1,
|
| 17 |
+
"conv_bias": true,
|
| 18 |
+
"conv_dim": [
|
| 19 |
+
512,
|
| 20 |
+
512,
|
| 21 |
+
512,
|
| 22 |
+
512,
|
| 23 |
+
512,
|
| 24 |
+
512,
|
| 25 |
+
512
|
| 26 |
+
],
|
| 27 |
+
"conv_kernel": [
|
| 28 |
+
10,
|
| 29 |
+
3,
|
| 30 |
+
3,
|
| 31 |
+
3,
|
| 32 |
+
3,
|
| 33 |
+
2,
|
| 34 |
+
2
|
| 35 |
+
],
|
| 36 |
+
"conv_stride": [
|
| 37 |
+
5,
|
| 38 |
+
2,
|
| 39 |
+
2,
|
| 40 |
+
2,
|
| 41 |
+
2,
|
| 42 |
+
2,
|
| 43 |
+
2
|
| 44 |
+
],
|
| 45 |
+
"ctc_loss_reduction": "mean",
|
| 46 |
+
"ctc_zero_infinity": true,
|
| 47 |
+
"diversity_loss_weight": 0.1,
|
| 48 |
+
"do_stable_layer_norm": true,
|
| 49 |
+
"eos_token_id": 2,
|
| 50 |
+
"feat_extract_activation": "gelu",
|
| 51 |
+
"feat_extract_dropout": 0.0,
|
| 52 |
+
"feat_extract_norm": "layer",
|
| 53 |
+
"feat_proj_dropout": 0.0,
|
| 54 |
+
"feat_quantizer_dropout": 0.0,
|
| 55 |
+
"final_dropout": 0.0,
|
| 56 |
+
"gradient_checkpointing": false,
|
| 57 |
+
"hidden_act": "gelu",
|
| 58 |
+
"hidden_dropout": 0.0,
|
| 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_channel_length": 10,
|
| 65 |
+
"mask_channel_min_space": 1,
|
| 66 |
+
"mask_channel_other": 0.0,
|
| 67 |
+
"mask_channel_prob": 0.0,
|
| 68 |
+
"mask_channel_selection": "static",
|
| 69 |
+
"mask_feature_length": 10,
|
| 70 |
+
"mask_feature_min_masks": 0,
|
| 71 |
+
"mask_feature_prob": 0.0,
|
| 72 |
+
"mask_time_length": 10,
|
| 73 |
+
"mask_time_min_masks": 2,
|
| 74 |
+
"mask_time_min_space": 1,
|
| 75 |
+
"mask_time_other": 0.0,
|
| 76 |
+
"mask_time_prob": 0.05,
|
| 77 |
+
"mask_time_selection": "static",
|
| 78 |
+
"model_type": "wav2vec2",
|
| 79 |
+
"num_adapter_layers": 3,
|
| 80 |
+
"num_attention_heads": 16,
|
| 81 |
+
"num_codevector_groups": 2,
|
| 82 |
+
"num_codevectors_per_group": 320,
|
| 83 |
+
"num_conv_pos_embedding_groups": 16,
|
| 84 |
+
"num_conv_pos_embeddings": 128,
|
| 85 |
+
"num_feat_extract_layers": 7,
|
| 86 |
+
"num_hidden_layers": 24,
|
| 87 |
+
"num_negatives": 100,
|
| 88 |
+
"output_hidden_size": 1024,
|
| 89 |
+
"pad_token_id": 30,
|
| 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": "4.31.0",
|
| 114 |
+
"use_weighted_layer_sum": false,
|
| 115 |
+
"vocab_size": 31,
|
| 116 |
+
"xvector_output_dim": 512
|
| 117 |
+
}
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:656d17329f3e590e7a50b86d2fba465268f1f52efe3e5f975aa6857e9ecdbd77
|
| 3 |
+
size 2524013434
|
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 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1736e79ab663ee98726b0e27dc404b35acb21c78968f5fe9cbc144ee45b8925f
|
| 3 |
+
size 1262029482
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fb11a5c36ed30169cf9f95a187e4eaff9d04ebc29c7a664bfc4b5895dc564c8b
|
| 3 |
+
size 14308
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3430397572aad2726ca253ff8c5e8a64c42924ae35ade1affda4ceebd201f6cf
|
| 3 |
+
size 1064
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<s>",
|
| 3 |
+
"clean_up_tokenization_spaces": true,
|
| 4 |
+
"do_lower_case": false,
|
| 5 |
+
"eos_token": "</s>",
|
| 6 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 7 |
+
"pad_token": "[PAD]",
|
| 8 |
+
"replace_word_delimiter_char": " ",
|
| 9 |
+
"target_lang": null,
|
| 10 |
+
"tokenizer_class": "Wav2Vec2CTCTokenizer",
|
| 11 |
+
"unk_token": "[UNK]",
|
| 12 |
+
"word_delimiter_token": "|"
|
| 13 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,520 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.15105250983113577,
|
| 3 |
+
"best_model_checkpoint": "xlsr_Paiwan/checkpoint-19740",
|
| 4 |
+
"epoch": 30.0,
|
| 5 |
+
"global_step": 19740,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.76,
|
| 12 |
+
"learning_rate": 0.0002924620060790273,
|
| 13 |
+
"loss": 2.0437,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 1.0,
|
| 18 |
+
"eval_loss": 0.43691062927246094,
|
| 19 |
+
"eval_runtime": 54.5716,
|
| 20 |
+
"eval_samples_per_second": 54.79,
|
| 21 |
+
"eval_steps_per_second": 6.853,
|
| 22 |
+
"eval_wer": 0.57050328806054,
|
| 23 |
+
"step": 658
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 1.52,
|
| 27 |
+
"learning_rate": 0.00028486322188449846,
|
| 28 |
+
"loss": 0.4605,
|
| 29 |
+
"step": 1000
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 2.0,
|
| 33 |
+
"eval_loss": 0.3218730688095093,
|
| 34 |
+
"eval_runtime": 53.5979,
|
| 35 |
+
"eval_samples_per_second": 55.786,
|
| 36 |
+
"eval_steps_per_second": 6.978,
|
| 37 |
+
"eval_wer": 0.424407653415287,
|
| 38 |
+
"step": 1316
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 2.28,
|
| 42 |
+
"learning_rate": 0.0002772644376899696,
|
| 43 |
+
"loss": 0.3486,
|
| 44 |
+
"step": 1500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 3.0,
|
| 48 |
+
"eval_loss": 0.2743721604347229,
|
| 49 |
+
"eval_runtime": 52.9623,
|
| 50 |
+
"eval_samples_per_second": 56.455,
|
| 51 |
+
"eval_steps_per_second": 7.062,
|
| 52 |
+
"eval_wer": 0.3111265325005783,
|
| 53 |
+
"step": 1974
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 3.04,
|
| 57 |
+
"learning_rate": 0.0002696656534954407,
|
| 58 |
+
"loss": 0.3005,
|
| 59 |
+
"step": 2000
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 3.8,
|
| 63 |
+
"learning_rate": 0.00026206686930091183,
|
| 64 |
+
"loss": 0.243,
|
| 65 |
+
"step": 2500
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 4.0,
|
| 69 |
+
"eval_loss": 0.2703934609889984,
|
| 70 |
+
"eval_runtime": 52.6582,
|
| 71 |
+
"eval_samples_per_second": 56.781,
|
| 72 |
+
"eval_steps_per_second": 7.102,
|
| 73 |
+
"eval_wer": 0.3132745117477942,
|
| 74 |
+
"step": 2632
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 4.56,
|
| 78 |
+
"learning_rate": 0.000254468085106383,
|
| 79 |
+
"loss": 0.2074,
|
| 80 |
+
"step": 3000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 5.0,
|
| 84 |
+
"eval_loss": 0.2540125250816345,
|
| 85 |
+
"eval_runtime": 51.8815,
|
| 86 |
+
"eval_samples_per_second": 57.631,
|
| 87 |
+
"eval_steps_per_second": 7.209,
|
| 88 |
+
"eval_wer": 0.29308350682396483,
|
| 89 |
+
"step": 3290
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 5.32,
|
| 93 |
+
"learning_rate": 0.0002468693009118541,
|
| 94 |
+
"loss": 0.1827,
|
| 95 |
+
"step": 3500
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 6.0,
|
| 99 |
+
"eval_loss": 0.24963043630123138,
|
| 100 |
+
"eval_runtime": 52.0611,
|
| 101 |
+
"eval_samples_per_second": 57.433,
|
| 102 |
+
"eval_steps_per_second": 7.184,
|
| 103 |
+
"eval_wer": 0.24047453818446185,
|
| 104 |
+
"step": 3948
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 6.08,
|
| 108 |
+
"learning_rate": 0.0002392705167173252,
|
| 109 |
+
"loss": 0.1649,
|
| 110 |
+
"step": 4000
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 6.84,
|
| 114 |
+
"learning_rate": 0.00023167173252279633,
|
| 115 |
+
"loss": 0.1371,
|
| 116 |
+
"step": 4500
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 7.0,
|
| 120 |
+
"eval_loss": 0.24666966497898102,
|
| 121 |
+
"eval_runtime": 52.0071,
|
| 122 |
+
"eval_samples_per_second": 57.492,
|
| 123 |
+
"eval_steps_per_second": 7.191,
|
| 124 |
+
"eval_wer": 0.23730213806549685,
|
| 125 |
+
"step": 4606
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 7.6,
|
| 129 |
+
"learning_rate": 0.00022407294832826744,
|
| 130 |
+
"loss": 0.1258,
|
| 131 |
+
"step": 5000
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 8.0,
|
| 135 |
+
"eval_loss": 0.25583532452583313,
|
| 136 |
+
"eval_runtime": 52.0669,
|
| 137 |
+
"eval_samples_per_second": 57.426,
|
| 138 |
+
"eval_steps_per_second": 7.183,
|
| 139 |
+
"eval_wer": 0.22309242919929942,
|
| 140 |
+
"step": 5264
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 8.36,
|
| 144 |
+
"learning_rate": 0.00021647416413373858,
|
| 145 |
+
"loss": 0.1078,
|
| 146 |
+
"step": 5500
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 9.0,
|
| 150 |
+
"eval_loss": 0.2600269019603729,
|
| 151 |
+
"eval_runtime": 52.0714,
|
| 152 |
+
"eval_samples_per_second": 57.421,
|
| 153 |
+
"eval_steps_per_second": 7.182,
|
| 154 |
+
"eval_wer": 0.213310862165824,
|
| 155 |
+
"step": 5922
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 9.12,
|
| 159 |
+
"learning_rate": 0.00020887537993920973,
|
| 160 |
+
"loss": 0.1038,
|
| 161 |
+
"step": 6000
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 9.88,
|
| 165 |
+
"learning_rate": 0.00020127659574468082,
|
| 166 |
+
"loss": 0.0906,
|
| 167 |
+
"step": 6500
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 10.0,
|
| 171 |
+
"eval_loss": 0.25083455443382263,
|
| 172 |
+
"eval_runtime": 52.0294,
|
| 173 |
+
"eval_samples_per_second": 57.467,
|
| 174 |
+
"eval_steps_per_second": 7.188,
|
| 175 |
+
"eval_wer": 0.20997323287399625,
|
| 176 |
+
"step": 6580
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 10.64,
|
| 180 |
+
"learning_rate": 0.00019367781155015196,
|
| 181 |
+
"loss": 0.0843,
|
| 182 |
+
"step": 7000
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 11.0,
|
| 186 |
+
"eval_loss": 0.2840941548347473,
|
| 187 |
+
"eval_runtime": 52.1825,
|
| 188 |
+
"eval_samples_per_second": 57.299,
|
| 189 |
+
"eval_steps_per_second": 7.167,
|
| 190 |
+
"eval_wer": 0.19976206999107762,
|
| 191 |
+
"step": 7238
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 11.4,
|
| 195 |
+
"learning_rate": 0.0001860790273556231,
|
| 196 |
+
"loss": 0.0765,
|
| 197 |
+
"step": 7500
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 12.0,
|
| 201 |
+
"eval_loss": 0.26433253288269043,
|
| 202 |
+
"eval_runtime": 52.2172,
|
| 203 |
+
"eval_samples_per_second": 57.261,
|
| 204 |
+
"eval_steps_per_second": 7.162,
|
| 205 |
+
"eval_wer": 0.19328508641485742,
|
| 206 |
+
"step": 7896
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 12.16,
|
| 210 |
+
"learning_rate": 0.00017848024316109422,
|
| 211 |
+
"loss": 0.0699,
|
| 212 |
+
"step": 8000
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 12.92,
|
| 216 |
+
"learning_rate": 0.00017088145896656533,
|
| 217 |
+
"loss": 0.0671,
|
| 218 |
+
"step": 8500
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 13.0,
|
| 222 |
+
"eval_loss": 0.2808719873428345,
|
| 223 |
+
"eval_runtime": 52.2538,
|
| 224 |
+
"eval_samples_per_second": 57.221,
|
| 225 |
+
"eval_steps_per_second": 7.157,
|
| 226 |
+
"eval_wer": 0.18816298205611182,
|
| 227 |
+
"step": 8554
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 13.68,
|
| 231 |
+
"learning_rate": 0.00016329787234042553,
|
| 232 |
+
"loss": 0.0589,
|
| 233 |
+
"step": 9000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 14.0,
|
| 237 |
+
"eval_loss": 0.2795076072216034,
|
| 238 |
+
"eval_runtime": 52.106,
|
| 239 |
+
"eval_samples_per_second": 57.383,
|
| 240 |
+
"eval_steps_per_second": 7.178,
|
| 241 |
+
"eval_wer": 0.18647764449291168,
|
| 242 |
+
"step": 9212
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 14.44,
|
| 246 |
+
"learning_rate": 0.00015569908814589664,
|
| 247 |
+
"loss": 0.0566,
|
| 248 |
+
"step": 9500
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 15.0,
|
| 252 |
+
"eval_loss": 0.27890855073928833,
|
| 253 |
+
"eval_runtime": 52.1531,
|
| 254 |
+
"eval_samples_per_second": 57.331,
|
| 255 |
+
"eval_steps_per_second": 7.171,
|
| 256 |
+
"eval_wer": 0.18158686097617396,
|
| 257 |
+
"step": 9870
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 15.2,
|
| 261 |
+
"learning_rate": 0.00014810030395136776,
|
| 262 |
+
"loss": 0.0514,
|
| 263 |
+
"step": 10000
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 15.96,
|
| 267 |
+
"learning_rate": 0.0001405015197568389,
|
| 268 |
+
"loss": 0.0507,
|
| 269 |
+
"step": 10500
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 16.0,
|
| 273 |
+
"eval_loss": 0.28275853395462036,
|
| 274 |
+
"eval_runtime": 52.1154,
|
| 275 |
+
"eval_samples_per_second": 57.373,
|
| 276 |
+
"eval_steps_per_second": 7.176,
|
| 277 |
+
"eval_wer": 0.17676216912858134,
|
| 278 |
+
"step": 10528
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 16.72,
|
| 282 |
+
"learning_rate": 0.00013291793313069907,
|
| 283 |
+
"loss": 0.0452,
|
| 284 |
+
"step": 11000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 17.0,
|
| 288 |
+
"eval_loss": 0.2956693470478058,
|
| 289 |
+
"eval_runtime": 56.2891,
|
| 290 |
+
"eval_samples_per_second": 53.119,
|
| 291 |
+
"eval_steps_per_second": 6.644,
|
| 292 |
+
"eval_wer": 0.1839661610653977,
|
| 293 |
+
"step": 11186
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 17.48,
|
| 297 |
+
"learning_rate": 0.0001253191489361702,
|
| 298 |
+
"loss": 0.0414,
|
| 299 |
+
"step": 11500
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 18.0,
|
| 303 |
+
"eval_loss": 0.272775262594223,
|
| 304 |
+
"eval_runtime": 52.2587,
|
| 305 |
+
"eval_samples_per_second": 57.215,
|
| 306 |
+
"eval_steps_per_second": 7.157,
|
| 307 |
+
"eval_wer": 0.17431677737021248,
|
| 308 |
+
"step": 11844
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 18.24,
|
| 312 |
+
"learning_rate": 0.00011772036474164133,
|
| 313 |
+
"loss": 0.0373,
|
| 314 |
+
"step": 12000
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 19.0,
|
| 318 |
+
"learning_rate": 0.00011012158054711246,
|
| 319 |
+
"loss": 0.0362,
|
| 320 |
+
"step": 12500
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 19.0,
|
| 324 |
+
"eval_loss": 0.2982702851295471,
|
| 325 |
+
"eval_runtime": 52.2158,
|
| 326 |
+
"eval_samples_per_second": 57.262,
|
| 327 |
+
"eval_steps_per_second": 7.163,
|
| 328 |
+
"eval_wer": 0.17220184395756916,
|
| 329 |
+
"step": 12502
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 19.76,
|
| 333 |
+
"learning_rate": 0.00010252279635258359,
|
| 334 |
+
"loss": 0.0348,
|
| 335 |
+
"step": 13000
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 20.0,
|
| 339 |
+
"eval_loss": 0.2988567054271698,
|
| 340 |
+
"eval_runtime": 52.245,
|
| 341 |
+
"eval_samples_per_second": 57.23,
|
| 342 |
+
"eval_steps_per_second": 7.159,
|
| 343 |
+
"eval_wer": 0.1638082019761409,
|
| 344 |
+
"step": 13160
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 20.52,
|
| 348 |
+
"learning_rate": 9.493920972644377e-05,
|
| 349 |
+
"loss": 0.0324,
|
| 350 |
+
"step": 13500
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 21.0,
|
| 354 |
+
"eval_loss": 0.3093124032020569,
|
| 355 |
+
"eval_runtime": 52.8274,
|
| 356 |
+
"eval_samples_per_second": 56.599,
|
| 357 |
+
"eval_steps_per_second": 7.08,
|
| 358 |
+
"eval_wer": 0.16509698952447044,
|
| 359 |
+
"step": 13818
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 21.28,
|
| 363 |
+
"learning_rate": 8.734042553191488e-05,
|
| 364 |
+
"loss": 0.0303,
|
| 365 |
+
"step": 14000
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 22.0,
|
| 369 |
+
"eval_loss": 0.3057407736778259,
|
| 370 |
+
"eval_runtime": 52.5941,
|
| 371 |
+
"eval_samples_per_second": 56.851,
|
| 372 |
+
"eval_steps_per_second": 7.111,
|
| 373 |
+
"eval_wer": 0.16489871451703514,
|
| 374 |
+
"step": 14476
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 22.04,
|
| 378 |
+
"learning_rate": 7.974164133738601e-05,
|
| 379 |
+
"loss": 0.0293,
|
| 380 |
+
"step": 14500
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 22.8,
|
| 384 |
+
"learning_rate": 7.214285714285714e-05,
|
| 385 |
+
"loss": 0.0267,
|
| 386 |
+
"step": 15000
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 23.0,
|
| 390 |
+
"eval_loss": 0.3179123103618622,
|
| 391 |
+
"eval_runtime": 52.2627,
|
| 392 |
+
"eval_samples_per_second": 57.211,
|
| 393 |
+
"eval_steps_per_second": 7.156,
|
| 394 |
+
"eval_wer": 0.1582895476025247,
|
| 395 |
+
"step": 15134
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 23.56,
|
| 399 |
+
"learning_rate": 6.455927051671732e-05,
|
| 400 |
+
"loss": 0.026,
|
| 401 |
+
"step": 15500
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 24.0,
|
| 405 |
+
"eval_loss": 0.3296581506729126,
|
| 406 |
+
"eval_runtime": 52.0664,
|
| 407 |
+
"eval_samples_per_second": 57.427,
|
| 408 |
+
"eval_steps_per_second": 7.183,
|
| 409 |
+
"eval_wer": 0.15799213509137172,
|
| 410 |
+
"step": 15792
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 24.32,
|
| 414 |
+
"learning_rate": 5.696048632218844e-05,
|
| 415 |
+
"loss": 0.0246,
|
| 416 |
+
"step": 16000
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 25.0,
|
| 420 |
+
"eval_loss": 0.3182116746902466,
|
| 421 |
+
"eval_runtime": 51.9316,
|
| 422 |
+
"eval_samples_per_second": 57.576,
|
| 423 |
+
"eval_steps_per_second": 7.202,
|
| 424 |
+
"eval_wer": 0.15541455999471265,
|
| 425 |
+
"step": 16450
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 25.08,
|
| 429 |
+
"learning_rate": 4.937689969604863e-05,
|
| 430 |
+
"loss": 0.0235,
|
| 431 |
+
"step": 16500
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 25.84,
|
| 435 |
+
"learning_rate": 4.179331306990881e-05,
|
| 436 |
+
"loss": 0.0215,
|
| 437 |
+
"step": 17000
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 26.0,
|
| 441 |
+
"eval_loss": 0.3294830620288849,
|
| 442 |
+
"eval_runtime": 52.2884,
|
| 443 |
+
"eval_samples_per_second": 57.183,
|
| 444 |
+
"eval_steps_per_second": 7.153,
|
| 445 |
+
"eval_wer": 0.15541455999471265,
|
| 446 |
+
"step": 17108
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 26.6,
|
| 450 |
+
"learning_rate": 3.419452887537994e-05,
|
| 451 |
+
"loss": 0.0199,
|
| 452 |
+
"step": 17500
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 27.0,
|
| 456 |
+
"eval_loss": 0.33707642555236816,
|
| 457 |
+
"eval_runtime": 51.7426,
|
| 458 |
+
"eval_samples_per_second": 57.786,
|
| 459 |
+
"eval_steps_per_second": 7.228,
|
| 460 |
+
"eval_wer": 0.1535309474240772,
|
| 461 |
+
"step": 17766
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 27.36,
|
| 465 |
+
"learning_rate": 2.6595744680851064e-05,
|
| 466 |
+
"loss": 0.019,
|
| 467 |
+
"step": 18000
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 28.0,
|
| 471 |
+
"eval_loss": 0.34009498357772827,
|
| 472 |
+
"eval_runtime": 51.768,
|
| 473 |
+
"eval_samples_per_second": 57.758,
|
| 474 |
+
"eval_steps_per_second": 7.225,
|
| 475 |
+
"eval_wer": 0.15174647235715938,
|
| 476 |
+
"step": 18424
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 28.12,
|
| 480 |
+
"learning_rate": 1.8996960486322186e-05,
|
| 481 |
+
"loss": 0.0185,
|
| 482 |
+
"step": 18500
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 28.88,
|
| 486 |
+
"learning_rate": 1.139817629179331e-05,
|
| 487 |
+
"loss": 0.0172,
|
| 488 |
+
"step": 19000
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 29.0,
|
| 492 |
+
"eval_loss": 0.34455737471580505,
|
| 493 |
+
"eval_runtime": 51.8182,
|
| 494 |
+
"eval_samples_per_second": 57.702,
|
| 495 |
+
"eval_steps_per_second": 7.218,
|
| 496 |
+
"eval_wer": 0.15108555566570833,
|
| 497 |
+
"step": 19082
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 29.64,
|
| 501 |
+
"learning_rate": 3.7993920972644374e-06,
|
| 502 |
+
"loss": 0.0166,
|
| 503 |
+
"step": 19500
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 30.0,
|
| 507 |
+
"eval_loss": 0.3466561436653137,
|
| 508 |
+
"eval_runtime": 51.8971,
|
| 509 |
+
"eval_samples_per_second": 57.614,
|
| 510 |
+
"eval_steps_per_second": 7.207,
|
| 511 |
+
"eval_wer": 0.15105250983113577,
|
| 512 |
+
"step": 19740
|
| 513 |
+
}
|
| 514 |
+
],
|
| 515 |
+
"max_steps": 19740,
|
| 516 |
+
"num_train_epochs": 30,
|
| 517 |
+
"total_flos": 7.422835365656386e+19,
|
| 518 |
+
"trial_name": null,
|
| 519 |
+
"trial_params": null
|
| 520 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ee5a82240ed6de36239549abd4fe5c6ad164eec81894a36b657a75bd87ea373f
|
| 3 |
+
size 4408
|
vocab.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"'": 1,
|
| 3 |
+
"a": 2,
|
| 4 |
+
"b": 3,
|
| 5 |
+
"c": 4,
|
| 6 |
+
"d": 5,
|
| 7 |
+
"e": 6,
|
| 8 |
+
"f": 7,
|
| 9 |
+
"g": 8,
|
| 10 |
+
"h": 9,
|
| 11 |
+
"i": 10,
|
| 12 |
+
"j": 11,
|
| 13 |
+
"k": 12,
|
| 14 |
+
"l": 13,
|
| 15 |
+
"m": 14,
|
| 16 |
+
"n": 15,
|
| 17 |
+
"p": 16,
|
| 18 |
+
"q": 17,
|
| 19 |
+
"r": 18,
|
| 20 |
+
"s": 19,
|
| 21 |
+
"t": 20,
|
| 22 |
+
"u": 21,
|
| 23 |
+
"v": 22,
|
| 24 |
+
"w": 23,
|
| 25 |
+
"x": 24,
|
| 26 |
+
"y": 25,
|
| 27 |
+
"z": 26,
|
| 28 |
+
"|": 0,
|
| 29 |
+
"…": 27
|
| 30 |
+
}
|