Muhammad Agung Hambali
commited on
Commit
•
fcadd8a
1
Parent(s):
d7a0091
Commit
Browse files- config.json +76 -0
- optimizer.pt +3 -0
- preprocessor_config.json +8 -0
- pytorch_model.bin +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +1 -0
- tokenizer_config.json +1 -0
- trainer_state.json +364 -0
- training_args.bin +3 -0
- vocab.json +1 -0
config.json
ADDED
|
@@ -0,0 +1,76 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "facebook/wav2vec2-large-xlsr-53",
|
| 3 |
+
"activation_dropout": 0.055,
|
| 4 |
+
"apply_spec_augment": true,
|
| 5 |
+
"architectures": [
|
| 6 |
+
"Wav2Vec2ForCTC"
|
| 7 |
+
],
|
| 8 |
+
"attention_dropout": 0.094,
|
| 9 |
+
"bos_token_id": 1,
|
| 10 |
+
"conv_bias": true,
|
| 11 |
+
"conv_dim": [
|
| 12 |
+
512,
|
| 13 |
+
512,
|
| 14 |
+
512,
|
| 15 |
+
512,
|
| 16 |
+
512,
|
| 17 |
+
512,
|
| 18 |
+
512
|
| 19 |
+
],
|
| 20 |
+
"conv_kernel": [
|
| 21 |
+
10,
|
| 22 |
+
3,
|
| 23 |
+
3,
|
| 24 |
+
3,
|
| 25 |
+
3,
|
| 26 |
+
2,
|
| 27 |
+
2
|
| 28 |
+
],
|
| 29 |
+
"conv_stride": [
|
| 30 |
+
5,
|
| 31 |
+
2,
|
| 32 |
+
2,
|
| 33 |
+
2,
|
| 34 |
+
2,
|
| 35 |
+
2,
|
| 36 |
+
2
|
| 37 |
+
],
|
| 38 |
+
"ctc_loss_reduction": "mean",
|
| 39 |
+
"ctc_zero_infinity": false,
|
| 40 |
+
"do_stable_layer_norm": true,
|
| 41 |
+
"eos_token_id": 2,
|
| 42 |
+
"feat_extract_activation": "gelu",
|
| 43 |
+
"feat_extract_dropout": 0.0,
|
| 44 |
+
"feat_extract_norm": "layer",
|
| 45 |
+
"feat_proj_dropout": 0.04,
|
| 46 |
+
"final_dropout": 0.0,
|
| 47 |
+
"gradient_checkpointing": true,
|
| 48 |
+
"hidden_act": "gelu",
|
| 49 |
+
"hidden_dropout": 0.047,
|
| 50 |
+
"hidden_size": 1024,
|
| 51 |
+
"initializer_range": 0.02,
|
| 52 |
+
"intermediate_size": 4096,
|
| 53 |
+
"layer_norm_eps": 1e-05,
|
| 54 |
+
"layerdrop": 0.1,
|
| 55 |
+
"mask_channel_length": 10,
|
| 56 |
+
"mask_channel_min_space": 1,
|
| 57 |
+
"mask_channel_other": 0.0,
|
| 58 |
+
"mask_channel_prob": 0.0,
|
| 59 |
+
"mask_channel_selection": "static",
|
| 60 |
+
"mask_feature_length": 10,
|
| 61 |
+
"mask_feature_prob": 0.0,
|
| 62 |
+
"mask_time_length": 10,
|
| 63 |
+
"mask_time_min_space": 1,
|
| 64 |
+
"mask_time_other": 0.0,
|
| 65 |
+
"mask_time_prob": 0.4,
|
| 66 |
+
"mask_time_selection": "static",
|
| 67 |
+
"model_type": "wav2vec2",
|
| 68 |
+
"num_attention_heads": 16,
|
| 69 |
+
"num_conv_pos_embedding_groups": 16,
|
| 70 |
+
"num_conv_pos_embeddings": 128,
|
| 71 |
+
"num_feat_extract_layers": 7,
|
| 72 |
+
"num_hidden_layers": 24,
|
| 73 |
+
"pad_token_id": 36,
|
| 74 |
+
"transformers_version": "4.5.0.dev0",
|
| 75 |
+
"vocab_size": 37
|
| 76 |
+
}
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f81a6d827f36ff160b01894d8fba9e568d1b1a5f4b3683f9c10c374ec0caa0c1
|
| 3 |
+
size 2490380551
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"feature_size": 1,
|
| 4 |
+
"padding_side": "right",
|
| 5 |
+
"padding_value": 0.0,
|
| 6 |
+
"return_attention_mask": true,
|
| 7 |
+
"sampling_rate": 16000
|
| 8 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:77efb9ebce7119724b2f50470d43818968e2995389e8c40da30ca1a2aa35938d
|
| 3 |
+
size 1262085527
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a31ffece1d5b746b3cd21adea0f69f1f47e7e31286d49b1ee37e9881e62d6cac
|
| 3 |
+
size 623
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
| 1 |
+
{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "[UNK]", "pad_token": "[PAD]"}
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
| 1 |
+
{"unk_token": "[UNK]", "bos_token": "<s>", "eos_token": "</s>", "pad_token": "[PAD]", "do_lower_case": false, "word_delimiter_token": "|"}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,364 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 58.8235294117647,
|
| 5 |
+
"global_step": 9000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 1.96,
|
| 12 |
+
"eval_loss": 2.9419331550598145,
|
| 13 |
+
"eval_runtime": 165.8041,
|
| 14 |
+
"eval_samples_per_second": 11.122,
|
| 15 |
+
"eval_wer": 1.0,
|
| 16 |
+
"step": 300
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 3.27,
|
| 20 |
+
"learning_rate": 9.774774774774775e-05,
|
| 21 |
+
"loss": 5.732,
|
| 22 |
+
"step": 500
|
| 23 |
+
},
|
| 24 |
+
{
|
| 25 |
+
"epoch": 3.92,
|
| 26 |
+
"eval_loss": 2.8035287857055664,
|
| 27 |
+
"eval_runtime": 163.2568,
|
| 28 |
+
"eval_samples_per_second": 11.295,
|
| 29 |
+
"eval_wer": 1.0,
|
| 30 |
+
"step": 600
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 5.88,
|
| 34 |
+
"eval_loss": 0.5059587955474854,
|
| 35 |
+
"eval_runtime": 166.512,
|
| 36 |
+
"eval_samples_per_second": 11.074,
|
| 37 |
+
"eval_wer": 0.646839455528622,
|
| 38 |
+
"step": 900
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 6.54,
|
| 42 |
+
"learning_rate": 9.211711711711712e-05,
|
| 43 |
+
"loss": 1.7477,
|
| 44 |
+
"step": 1000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 7.84,
|
| 48 |
+
"eval_loss": 0.31517693400382996,
|
| 49 |
+
"eval_runtime": 168.535,
|
| 50 |
+
"eval_samples_per_second": 10.941,
|
| 51 |
+
"eval_wer": 0.49533842998321836,
|
| 52 |
+
"step": 1200
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 9.8,
|
| 56 |
+
"learning_rate": 8.64864864864865e-05,
|
| 57 |
+
"loss": 0.4124,
|
| 58 |
+
"step": 1500
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 9.8,
|
| 62 |
+
"eval_loss": 0.30525174736976624,
|
| 63 |
+
"eval_runtime": 170.5049,
|
| 64 |
+
"eval_samples_per_second": 10.815,
|
| 65 |
+
"eval_wer": 0.4454596308036547,
|
| 66 |
+
"step": 1500
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 11.76,
|
| 70 |
+
"eval_loss": 0.27497437596321106,
|
| 71 |
+
"eval_runtime": 171.8183,
|
| 72 |
+
"eval_samples_per_second": 10.732,
|
| 73 |
+
"eval_wer": 0.4267201193361924,
|
| 74 |
+
"step": 1800
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 13.07,
|
| 78 |
+
"learning_rate": 8.085585585585586e-05,
|
| 79 |
+
"loss": 0.2831,
|
| 80 |
+
"step": 2000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 13.73,
|
| 84 |
+
"eval_loss": 0.28638383746147156,
|
| 85 |
+
"eval_runtime": 174.9018,
|
| 86 |
+
"eval_samples_per_second": 10.543,
|
| 87 |
+
"eval_wer": 0.4133880290881969,
|
| 88 |
+
"step": 2100
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 15.69,
|
| 92 |
+
"eval_loss": 0.27328914403915405,
|
| 93 |
+
"eval_runtime": 174.9487,
|
| 94 |
+
"eval_samples_per_second": 10.54,
|
| 95 |
+
"eval_wer": 0.40583628566101065,
|
| 96 |
+
"step": 2400
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 16.34,
|
| 100 |
+
"learning_rate": 7.522522522522523e-05,
|
| 101 |
+
"loss": 0.2231,
|
| 102 |
+
"step": 2500
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 17.65,
|
| 106 |
+
"eval_loss": 0.29071566462516785,
|
| 107 |
+
"eval_runtime": 177.3068,
|
| 108 |
+
"eval_samples_per_second": 10.4,
|
| 109 |
+
"eval_wer": 0.3988439306358382,
|
| 110 |
+
"step": 2700
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 19.61,
|
| 114 |
+
"learning_rate": 6.95945945945946e-05,
|
| 115 |
+
"loss": 0.1885,
|
| 116 |
+
"step": 3000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 19.61,
|
| 120 |
+
"eval_loss": 0.2765745520591736,
|
| 121 |
+
"eval_runtime": 177.2476,
|
| 122 |
+
"eval_samples_per_second": 10.404,
|
| 123 |
+
"eval_wer": 0.39912362483684505,
|
| 124 |
+
"step": 3000
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 21.57,
|
| 128 |
+
"eval_loss": 0.28941160440444946,
|
| 129 |
+
"eval_runtime": 180.7189,
|
| 130 |
+
"eval_samples_per_second": 10.204,
|
| 131 |
+
"eval_wer": 0.38355398098079435,
|
| 132 |
+
"step": 3300
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 22.88,
|
| 136 |
+
"learning_rate": 6.396396396396397e-05,
|
| 137 |
+
"loss": 0.1657,
|
| 138 |
+
"step": 3500
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 23.53,
|
| 142 |
+
"eval_loss": 0.27626267075538635,
|
| 143 |
+
"eval_runtime": 182.3178,
|
| 144 |
+
"eval_samples_per_second": 10.114,
|
| 145 |
+
"eval_wer": 0.38094350177139663,
|
| 146 |
+
"step": 3600
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 25.49,
|
| 150 |
+
"eval_loss": 0.28250962495803833,
|
| 151 |
+
"eval_runtime": 183.703,
|
| 152 |
+
"eval_samples_per_second": 10.038,
|
| 153 |
+
"eval_wer": 0.3820622785754242,
|
| 154 |
+
"step": 3900
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 26.14,
|
| 158 |
+
"learning_rate": 5.833333333333334e-05,
|
| 159 |
+
"loss": 0.1449,
|
| 160 |
+
"step": 4000
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 27.45,
|
| 164 |
+
"eval_loss": 0.27114051580429077,
|
| 165 |
+
"eval_runtime": 183.7677,
|
| 166 |
+
"eval_samples_per_second": 10.034,
|
| 167 |
+
"eval_wer": 0.38140965877307476,
|
| 168 |
+
"step": 4200
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 29.41,
|
| 172 |
+
"learning_rate": 5.27027027027027e-05,
|
| 173 |
+
"loss": 0.1344,
|
| 174 |
+
"step": 4500
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 29.41,
|
| 178 |
+
"eval_loss": 0.3000315725803375,
|
| 179 |
+
"eval_runtime": 185.7074,
|
| 180 |
+
"eval_samples_per_second": 9.93,
|
| 181 |
+
"eval_wer": 0.374230840947231,
|
| 182 |
+
"step": 4500
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 31.37,
|
| 186 |
+
"eval_loss": 0.2914937734603882,
|
| 187 |
+
"eval_runtime": 169.8087,
|
| 188 |
+
"eval_samples_per_second": 10.859,
|
| 189 |
+
"eval_wer": 0.3726459071415253,
|
| 190 |
+
"step": 4800
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 32.68,
|
| 194 |
+
"learning_rate": 4.707207207207208e-05,
|
| 195 |
+
"loss": 0.1229,
|
| 196 |
+
"step": 5000
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 33.33,
|
| 200 |
+
"eval_loss": 0.2908037304878235,
|
| 201 |
+
"eval_runtime": 189.3466,
|
| 202 |
+
"eval_samples_per_second": 9.739,
|
| 203 |
+
"eval_wer": 0.3699421965317919,
|
| 204 |
+
"step": 5100
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 35.29,
|
| 208 |
+
"eval_loss": 0.3009437620639801,
|
| 209 |
+
"eval_runtime": 170.798,
|
| 210 |
+
"eval_samples_per_second": 10.796,
|
| 211 |
+
"eval_wer": 0.3701286593324632,
|
| 212 |
+
"step": 5400
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 35.95,
|
| 216 |
+
"learning_rate": 4.1441441441441444e-05,
|
| 217 |
+
"loss": 0.1133,
|
| 218 |
+
"step": 5500
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 37.25,
|
| 222 |
+
"eval_loss": 0.3189816176891327,
|
| 223 |
+
"eval_runtime": 171.2661,
|
| 224 |
+
"eval_samples_per_second": 10.767,
|
| 225 |
+
"eval_wer": 0.36947603953011376,
|
| 226 |
+
"step": 5700
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 39.22,
|
| 230 |
+
"learning_rate": 3.581081081081081e-05,
|
| 231 |
+
"loss": 0.11,
|
| 232 |
+
"step": 6000
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 39.22,
|
| 236 |
+
"eval_loss": 0.30211004614830017,
|
| 237 |
+
"eval_runtime": 190.742,
|
| 238 |
+
"eval_samples_per_second": 9.668,
|
| 239 |
+
"eval_wer": 0.36267014730561253,
|
| 240 |
+
"step": 6000
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 41.18,
|
| 244 |
+
"eval_loss": 0.30648261308670044,
|
| 245 |
+
"eval_runtime": 173.8977,
|
| 246 |
+
"eval_samples_per_second": 10.604,
|
| 247 |
+
"eval_wer": 0.36677232892038036,
|
| 248 |
+
"step": 6300
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 42.48,
|
| 252 |
+
"learning_rate": 3.0180180180180183e-05,
|
| 253 |
+
"loss": 0.1006,
|
| 254 |
+
"step": 6500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 43.14,
|
| 258 |
+
"eval_loss": 0.30688807368278503,
|
| 259 |
+
"eval_runtime": 195.1525,
|
| 260 |
+
"eval_samples_per_second": 9.449,
|
| 261 |
+
"eval_wer": 0.36826403132575053,
|
| 262 |
+
"step": 6600
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 45.1,
|
| 266 |
+
"eval_loss": 0.31185880303382874,
|
| 267 |
+
"eval_runtime": 176.2725,
|
| 268 |
+
"eval_samples_per_second": 10.461,
|
| 269 |
+
"eval_wer": 0.36164460190192055,
|
| 270 |
+
"step": 6900
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 45.75,
|
| 274 |
+
"learning_rate": 2.454954954954955e-05,
|
| 275 |
+
"loss": 0.0998,
|
| 276 |
+
"step": 7000
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 47.06,
|
| 280 |
+
"eval_loss": 0.31436818838119507,
|
| 281 |
+
"eval_runtime": 195.8727,
|
| 282 |
+
"eval_samples_per_second": 9.414,
|
| 283 |
+
"eval_wer": 0.36108521349990674,
|
| 284 |
+
"step": 7200
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 49.02,
|
| 288 |
+
"learning_rate": 1.891891891891892e-05,
|
| 289 |
+
"loss": 0.0952,
|
| 290 |
+
"step": 7500
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 49.02,
|
| 294 |
+
"eval_loss": 0.31186020374298096,
|
| 295 |
+
"eval_runtime": 201.6879,
|
| 296 |
+
"eval_samples_per_second": 9.143,
|
| 297 |
+
"eval_wer": 0.3613649077009137,
|
| 298 |
+
"step": 7500
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 50.98,
|
| 302 |
+
"eval_loss": 0.3063485026359558,
|
| 303 |
+
"eval_runtime": 177.6307,
|
| 304 |
+
"eval_samples_per_second": 10.381,
|
| 305 |
+
"eval_wer": 0.35959351109453663,
|
| 306 |
+
"step": 7800
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 52.29,
|
| 310 |
+
"learning_rate": 1.3288288288288289e-05,
|
| 311 |
+
"loss": 0.0898,
|
| 312 |
+
"step": 8000
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 52.94,
|
| 316 |
+
"eval_loss": 0.32244834303855896,
|
| 317 |
+
"eval_runtime": 179.3678,
|
| 318 |
+
"eval_samples_per_second": 10.281,
|
| 319 |
+
"eval_wer": 0.360525825097893,
|
| 320 |
+
"step": 8100
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 54.9,
|
| 324 |
+
"eval_loss": 0.32377830147743225,
|
| 325 |
+
"eval_runtime": 179.7421,
|
| 326 |
+
"eval_samples_per_second": 10.259,
|
| 327 |
+
"eval_wer": 0.360525825097893,
|
| 328 |
+
"step": 8400
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 55.56,
|
| 332 |
+
"learning_rate": 7.657657657657658e-06,
|
| 333 |
+
"loss": 0.0891,
|
| 334 |
+
"step": 8500
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 56.86,
|
| 338 |
+
"eval_loss": 0.3199419677257538,
|
| 339 |
+
"eval_runtime": 178.805,
|
| 340 |
+
"eval_samples_per_second": 10.313,
|
| 341 |
+
"eval_wer": 0.3609919820995711,
|
| 342 |
+
"step": 8700
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 58.82,
|
| 346 |
+
"learning_rate": 2.0270270270270273e-06,
|
| 347 |
+
"loss": 0.0877,
|
| 348 |
+
"step": 9000
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 58.82,
|
| 352 |
+
"eval_loss": 0.3155083954334259,
|
| 353 |
+
"eval_runtime": 180.9925,
|
| 354 |
+
"eval_samples_per_second": 10.188,
|
| 355 |
+
"eval_wer": 0.35922058549319413,
|
| 356 |
+
"step": 9000
|
| 357 |
+
}
|
| 358 |
+
],
|
| 359 |
+
"max_steps": 9180,
|
| 360 |
+
"num_train_epochs": 60,
|
| 361 |
+
"total_flos": 6.238876246016606e+19,
|
| 362 |
+
"trial_name": null,
|
| 363 |
+
"trial_params": null
|
| 364 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:85def95de9b09edef95f86214537378b9d6ac7188834833e0eb400d64bd26951
|
| 3 |
+
size 2351
|
vocab.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
| 1 |
+
{"n": 0, "k": 1, "m": 2, "z": 3, "v": 4, "∩": 5, "f": 6, "p": 8, "h": 9, "，": 10, "i": 11, "î": 12, "！": 13, "a": 14, "r": 15, "?": 16, "t": 17, "ε": 18, "b": 19, "w": 20, "c": 21, "s": 22, "l": 23, ",": 24, "x": 25, "d": 26, "ⁿ": 27, "u": 28, "e": 29, "o": 30, "j": 31, "'": 32, "y": 33, "g": 34, "|": 7, "[UNK]": 35, "[PAD]": 36}
|