Hugging Face's logo

yesj1234
/
xlsr_cycle0_ko

Automatic Speech Recognition
Transformers
PyTorch
wav2vec2
./sample_speech.py
Generated from Trainer
Inference Endpoints
Model card Files Community
yesj1234 commited on
Commit
8fcacba
โ€ข
1 Parent(s): 9b952c3

Upload folder using huggingface_hub

Browse files
Files changed (28) hide show
  1. README.md +69 -0
  2. added_tokens.json +4 -0
  3. all_results.json +14 -0
  4. checkpoint-2250/added_tokens.json +4 -0
  5. checkpoint-2250/config.json +117 -0
  6. checkpoint-2250/optimizer.pt +3 -0
  7. checkpoint-2250/preprocessor_config.json +10 -0
  8. checkpoint-2250/pytorch_model.bin +3 -0
  9. checkpoint-2250/rng_state_0.pth +3 -0
  10. checkpoint-2250/rng_state_1.pth +3 -0
  11. checkpoint-2250/rng_state_2.pth +3 -0
  12. checkpoint-2250/rng_state_3.pth +3 -0
  13. checkpoint-2250/scheduler.pt +3 -0
  14. checkpoint-2250/special_tokens_map.json +36 -0
  15. checkpoint-2250/tokenizer_config.json +17 -0
  16. checkpoint-2250/trainer_state.json +337 -0
  17. checkpoint-2250/training_args.bin +3 -0
  18. checkpoint-2250/vocab.json +759 -0
  19. config.json +117 -0
  20. eval_results.json +9 -0
  21. preprocessor_config.json +10 -0
  22. pytorch_model.bin +3 -0
  23. special_tokens_map.json +36 -0
  24. tokenizer_config.json +17 -0
  25. train_results.json +8 -0
  26. trainer_state.json +346 -0
  27. training_args.bin +3 -0
  28. vocab.json +759 -0
README.md ADDED
@@ -0,0 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ base_model: facebook/wav2vec2-large-xlsr-53
4
+ tags:
5
+ - automatic-speech-recognition
6
+ - ./sample_speech.py
7
+ - generated_from_trainer
8
+ datasets:
9
+ - sample_speech
10
+ model-index:
11
+ - name: ko-xlsr
12
+ results: []
13
+ ---
14
+
15
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
16
+ should probably proofread and complete it, then remove this comment. -->
17
+
18
+ # ko-xlsr
19
+
20
+ This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the ./SAMPLE_SPEECH.PY - NA dataset.
21
+ It achieves the following results on the evaluation set:
22
+ - Loss: 1.5649
23
+ - Cer: 0.3569
24
+
25
+ ## Model description
26
+
27
+ More information needed
28
+
29
+ ## Intended uses & limitations
30
+
31
+ More information needed
32
+
33
+ ## Training and evaluation data
34
+
35
+ More information needed
36
+
37
+ ## Training procedure
38
+
39
+ ### Training hyperparameters
40
+
41
+ The following hyperparameters were used during training:
42
+ - learning_rate: 0.0005
43
+ - train_batch_size: 4
44
+ - eval_batch_size: 4
45
+ - seed: 42
46
+ - distributed_type: multi-GPU
47
+ - num_devices: 4
48
+ - gradient_accumulation_steps: 2
49
+ - total_train_batch_size: 32
50
+ - total_eval_batch_size: 16
51
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
52
+ - lr_scheduler_type: linear
53
+ - lr_scheduler_warmup_steps: 300
54
+ - num_epochs: 50
55
+
56
+ ### Training results
57
+
58
+ | Training Loss | Epoch | Step | Validation Loss | Cer |
59
+ |:-------------:|:-----:|:----:|:---------------:|:------:|
60
+ | 2.5849 | 22.22 | 1000 | 2.5846 | 0.5985 |
61
+ | 0.7224 | 44.44 | 2000 | 1.5880 | 0.3664 |
62
+
63
+
64
+ ### Framework versions
65
+
66
+ - Transformers 4.33.1
67
+ - Pytorch 2.0.1+cu117
68
+ - Datasets 2.14.5
69
+ - Tokenizers 0.13.3
added_tokens.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "</s>": 758,
3
+ "<s>": 757
4
+ }
all_results.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 50.0,
3
+ "eval_cer": 0.35694765608576834,
4
+ "eval_loss": 1.5648640394210815,
5
+ "eval_runtime": 2.3295,
6
+ "eval_samples": 178,
7
+ "eval_samples_per_second": 76.41,
8
+ "eval_steps_per_second": 5.151,
9
+ "train_loss": 3.618984514872233,
10
+ "train_runtime": 1635.0894,
11
+ "train_samples": 1427,
12
+ "train_samples_per_second": 43.637,
13
+ "train_steps_per_second": 1.376
14
+ }
checkpoint-2250/added_tokens.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "</s>": 758,
3
+ "<s>": 757
4
+ }
checkpoint-2250/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": 0.1,
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": false,
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": 0.1,
54
+ "feat_quantizer_dropout": 0.0,
55
+ "final_dropout": 0.0,
56
+ "gradient_checkpointing": false,
57
+ "hidden_act": "gelu",
58
+ "hidden_dropout": 0.1,
59
+ "hidden_size": 1024,
60
+ "initializer_range": 0.02,
61
+ "intermediate_size": 4096,
62
+ "layer_norm_eps": 1e-05,
63
+ "layerdrop": 0.1,
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": 64,
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": 0.3,
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": 756,
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": "4.33.1",
114
+ "use_weighted_layer_sum": false,
115
+ "vocab_size": 759,
116
+ "xvector_output_dim": 512
117
+ }
checkpoint-2250/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fce9939803523ca2fe7fecfbfec004c9002f12e3035588bb4762b16640d7059a
3
+ size 2496383105
checkpoint-2250/preprocessor_config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "feature_extractor_type": "Wav2Vec2FeatureExtractor",
4
+ "feature_size": 1,
5
+ "padding_side": "right",
6
+ "padding_value": 0,
7
+ "processor_class": "Wav2Vec2Processor",
8
+ "return_attention_mask": true,
9
+ "sampling_rate": 16000
10
+ }
checkpoint-2250/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:97c86e323d40ef74a32ad8abfac0ae57d2d2f3a5d9d6c9e1cf59fa08fee1a0d1
3
+ size 1265013741
checkpoint-2250/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f7c0989ad5c09270c9304f1c644a1430b5e674e438a95cdca0e1ee52f4edf69b
3
+ size 17719
checkpoint-2250/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:80e6e88fc6da924d5c0e15d0ffee84126dd123d38019452301d4356e6ec04fb0
3
+ size 17719
checkpoint-2250/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f336d8e416c03538956386304870acd0e806f528b25f3b4f0d61b065ee02e726
3
+ size 17655
checkpoint-2250/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:612d3a8128bd854ae687c9266850cc6d70abbc99891df1c71ecfc80ab37cee9e
3
+ size 17655
checkpoint-2250/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:842ac40ff3956e2a114ea4619968ab3039307127cc5d8c24851b2830d4d5c4c3
3
+ size 627
checkpoint-2250/special_tokens_map.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": true,
7
+ "rstrip": false,
8
+ "single_word": false
9
+ },
10
+ {
11
+ "content": "</s>",
12
+ "lstrip": false,
13
+ "normalized": true,
14
+ "rstrip": false,
15
+ "single_word": false
16
+ },
17
+ {
18
+ "content": "<s>",
19
+ "lstrip": false,
20
+ "normalized": true,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ },
24
+ {
25
+ "content": "</s>",
26
+ "lstrip": false,
27
+ "normalized": true,
28
+ "rstrip": false,
29
+ "single_word": false
30
+ }
31
+ ],
32
+ "bos_token": "<s>",
33
+ "eos_token": "</s>",
34
+ "pad_token": "[PAD]",
35
+ "unk_token": "[UNK]"
36
+ }
checkpoint-2250/tokenizer_config.json ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "clean_up_tokenization_spaces": true,
4
+ "config": null,
5
+ "do_lower_case": false,
6
+ "eos_token": "</s>",
7
+ "model_max_length": 1000000000000000019884624838656,
8
+ "pad_token": "[PAD]",
9
+ "processor_class": "Wav2Vec2Processor",
10
+ "replace_word_delimiter_char": " ",
11
+ "target_lang": null,
12
+ "tokenizer_class": "Wav2Vec2CTCTokenizer",
13
+ "tokenizer_type": "wav2vec2",
14
+ "trust_remote_code": false,
15
+ "unk_token": "[UNK]",
16
+ "word_delimiter_token": "|"
17
+ }
checkpoint-2250/trainer_state.json ADDED
@@ -0,0 +1,337 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 50.0,
5
+ "eval_steps": 1000,
6
+ "global_step": 2250,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 1.0,
13
+ "learning_rate": 6.500000000000001e-05,
14
+ "loss": 42.9058,
15
+ "step": 45
16
+ },
17
+ {
18
+ "epoch": 2.0,
19
+ "learning_rate": 0.00014000000000000001,
20
+ "loss": 21.7572,
21
+ "step": 90
22
+ },
23
+ {
24
+ "epoch": 3.0,
25
+ "learning_rate": 0.000215,
26
+ "loss": 6.6528,
27
+ "step": 135
28
+ },
29
+ {
30
+ "epoch": 4.0,
31
+ "learning_rate": 0.00029,
32
+ "loss": 4.7079,
33
+ "step": 180
34
+ },
35
+ {
36
+ "epoch": 5.0,
37
+ "learning_rate": 0.000365,
38
+ "loss": 4.6824,
39
+ "step": 225
40
+ },
41
+ {
42
+ "epoch": 6.0,
43
+ "learning_rate": 0.00044,
44
+ "loss": 4.6861,
45
+ "step": 270
46
+ },
47
+ {
48
+ "epoch": 7.0,
49
+ "learning_rate": 0.0004976923076923077,
50
+ "loss": 4.6781,
51
+ "step": 315
52
+ },
53
+ {
54
+ "epoch": 8.0,
55
+ "learning_rate": 0.0004861538461538462,
56
+ "loss": 4.657,
57
+ "step": 360
58
+ },
59
+ {
60
+ "epoch": 9.0,
61
+ "learning_rate": 0.00047461538461538463,
62
+ "loss": 4.6171,
63
+ "step": 405
64
+ },
65
+ {
66
+ "epoch": 10.0,
67
+ "learning_rate": 0.00046307692307692304,
68
+ "loss": 4.5707,
69
+ "step": 450
70
+ },
71
+ {
72
+ "epoch": 11.0,
73
+ "learning_rate": 0.00045153846153846156,
74
+ "loss": 4.4978,
75
+ "step": 495
76
+ },
77
+ {
78
+ "epoch": 12.0,
79
+ "learning_rate": 0.00044,
80
+ "loss": 4.4697,
81
+ "step": 540
82
+ },
83
+ {
84
+ "epoch": 13.0,
85
+ "learning_rate": 0.0004284615384615385,
86
+ "loss": 4.4129,
87
+ "step": 585
88
+ },
89
+ {
90
+ "epoch": 14.0,
91
+ "learning_rate": 0.0004169230769230769,
92
+ "loss": 4.3792,
93
+ "step": 630
94
+ },
95
+ {
96
+ "epoch": 15.0,
97
+ "learning_rate": 0.0004053846153846154,
98
+ "loss": 4.3352,
99
+ "step": 675
100
+ },
101
+ {
102
+ "epoch": 16.0,
103
+ "learning_rate": 0.00039384615384615386,
104
+ "loss": 4.3007,
105
+ "step": 720
106
+ },
107
+ {
108
+ "epoch": 17.0,
109
+ "learning_rate": 0.0003823076923076923,
110
+ "loss": 4.2279,
111
+ "step": 765
112
+ },
113
+ {
114
+ "epoch": 18.0,
115
+ "learning_rate": 0.0003707692307692308,
116
+ "loss": 4.1306,
117
+ "step": 810
118
+ },
119
+ {
120
+ "epoch": 19.0,
121
+ "learning_rate": 0.00035923076923076924,
122
+ "loss": 3.9656,
123
+ "step": 855
124
+ },
125
+ {
126
+ "epoch": 20.0,
127
+ "learning_rate": 0.0003476923076923077,
128
+ "loss": 3.62,
129
+ "step": 900
130
+ },
131
+ {
132
+ "epoch": 21.0,
133
+ "learning_rate": 0.00033615384615384616,
134
+ "loss": 3.0615,
135
+ "step": 945
136
+ },
137
+ {
138
+ "epoch": 22.0,
139
+ "learning_rate": 0.0003246153846153846,
140
+ "loss": 2.5849,
141
+ "step": 990
142
+ },
143
+ {
144
+ "epoch": 22.22,
145
+ "eval_cer": 0.5984864410342653,
146
+ "eval_loss": 2.5845649242401123,
147
+ "eval_runtime": 2.3789,
148
+ "eval_samples_per_second": 74.826,
149
+ "eval_steps_per_second": 5.044,
150
+ "step": 1000
151
+ },
152
+ {
153
+ "epoch": 23.0,
154
+ "learning_rate": 0.0003130769230769231,
155
+ "loss": 2.1788,
156
+ "step": 1035
157
+ },
158
+ {
159
+ "epoch": 24.0,
160
+ "learning_rate": 0.00030153846153846154,
161
+ "loss": 1.9325,
162
+ "step": 1080
163
+ },
164
+ {
165
+ "epoch": 25.0,
166
+ "learning_rate": 0.00029,
167
+ "loss": 1.7492,
168
+ "step": 1125
169
+ },
170
+ {
171
+ "epoch": 26.0,
172
+ "learning_rate": 0.00027846153846153846,
173
+ "loss": 1.6023,
174
+ "step": 1170
175
+ },
176
+ {
177
+ "epoch": 27.0,
178
+ "learning_rate": 0.0002669230769230769,
179
+ "loss": 1.4718,
180
+ "step": 1215
181
+ },
182
+ {
183
+ "epoch": 28.0,
184
+ "learning_rate": 0.0002553846153846154,
185
+ "loss": 1.3751,
186
+ "step": 1260
187
+ },
188
+ {
189
+ "epoch": 29.0,
190
+ "learning_rate": 0.00024384615384615384,
191
+ "loss": 1.2758,
192
+ "step": 1305
193
+ },
194
+ {
195
+ "epoch": 30.0,
196
+ "learning_rate": 0.0002323076923076923,
197
+ "loss": 1.2088,
198
+ "step": 1350
199
+ },
200
+ {
201
+ "epoch": 31.0,
202
+ "learning_rate": 0.00022076923076923076,
203
+ "loss": 1.1447,
204
+ "step": 1395
205
+ },
206
+ {
207
+ "epoch": 32.0,
208
+ "learning_rate": 0.00020923076923076922,
209
+ "loss": 1.07,
210
+ "step": 1440
211
+ },
212
+ {
213
+ "epoch": 33.0,
214
+ "learning_rate": 0.0001976923076923077,
215
+ "loss": 1.0118,
216
+ "step": 1485
217
+ },
218
+ {
219
+ "epoch": 34.0,
220
+ "learning_rate": 0.00018615384615384615,
221
+ "loss": 0.9947,
222
+ "step": 1530
223
+ },
224
+ {
225
+ "epoch": 35.0,
226
+ "learning_rate": 0.00017461538461538463,
227
+ "loss": 0.9557,
228
+ "step": 1575
229
+ },
230
+ {
231
+ "epoch": 36.0,
232
+ "learning_rate": 0.00016307692307692307,
233
+ "loss": 0.8984,
234
+ "step": 1620
235
+ },
236
+ {
237
+ "epoch": 37.0,
238
+ "learning_rate": 0.00015153846153846155,
239
+ "loss": 0.8694,
240
+ "step": 1665
241
+ },
242
+ {
243
+ "epoch": 38.0,
244
+ "learning_rate": 0.00014000000000000001,
245
+ "loss": 0.8245,
246
+ "step": 1710
247
+ },
248
+ {
249
+ "epoch": 39.0,
250
+ "learning_rate": 0.00012846153846153848,
251
+ "loss": 0.7914,
252
+ "step": 1755
253
+ },
254
+ {
255
+ "epoch": 40.0,
256
+ "learning_rate": 0.00011692307692307692,
257
+ "loss": 0.7883,
258
+ "step": 1800
259
+ },
260
+ {
261
+ "epoch": 41.0,
262
+ "learning_rate": 0.00010538461538461538,
263
+ "loss": 0.7628,
264
+ "step": 1845
265
+ },
266
+ {
267
+ "epoch": 42.0,
268
+ "learning_rate": 9.384615384615384e-05,
269
+ "loss": 0.7449,
270
+ "step": 1890
271
+ },
272
+ {
273
+ "epoch": 43.0,
274
+ "learning_rate": 8.23076923076923e-05,
275
+ "loss": 0.7114,
276
+ "step": 1935
277
+ },
278
+ {
279
+ "epoch": 44.0,
280
+ "learning_rate": 7.076923076923076e-05,
281
+ "loss": 0.7224,
282
+ "step": 1980
283
+ },
284
+ {
285
+ "epoch": 44.44,
286
+ "eval_cer": 0.3664073996216103,
287
+ "eval_loss": 1.5880260467529297,
288
+ "eval_runtime": 2.3696,
289
+ "eval_samples_per_second": 75.118,
290
+ "eval_steps_per_second": 5.064,
291
+ "step": 2000
292
+ },
293
+ {
294
+ "epoch": 45.0,
295
+ "learning_rate": 5.923076923076923e-05,
296
+ "loss": 0.703,
297
+ "step": 2025
298
+ },
299
+ {
300
+ "epoch": 46.0,
301
+ "learning_rate": 4.769230769230769e-05,
302
+ "loss": 0.6685,
303
+ "step": 2070
304
+ },
305
+ {
306
+ "epoch": 47.0,
307
+ "learning_rate": 3.615384615384615e-05,
308
+ "loss": 0.6755,
309
+ "step": 2115
310
+ },
311
+ {
312
+ "epoch": 48.0,
313
+ "learning_rate": 2.4615384615384616e-05,
314
+ "loss": 0.6411,
315
+ "step": 2160
316
+ },
317
+ {
318
+ "epoch": 49.0,
319
+ "learning_rate": 1.3076923076923077e-05,
320
+ "loss": 0.6266,
321
+ "step": 2205
322
+ },
323
+ {
324
+ "epoch": 50.0,
325
+ "learning_rate": 1.5384615384615385e-06,
326
+ "loss": 0.6484,
327
+ "step": 2250
328
+ }
329
+ ],
330
+ "logging_steps": 500,
331
+ "max_steps": 2250,
332
+ "num_train_epochs": 50,
333
+ "save_steps": 500,
334
+ "total_flos": 9.519369792641827e+18,
335
+ "trial_name": null,
336
+ "trial_params": null
337
+ }
checkpoint-2250/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:37b6ee5799ff6b41ff6cf71255a8b47d6164c730da9afd8c02d1565ac3df39a9
3
+ size 4027
checkpoint-2250/vocab.json ADDED
@@ -0,0 +1,759 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "0": 1,
3
+ "1": 2,
4
+ "2": 3,
5
+ "3": 4,
6
+ "4": 5,
7
+ "5": 6,
8
+ "6": 7,
9
+ "7": 8,
10
+ "8": 9,
11
+ "9": 10,
12
+ "[PAD]": 756,
13
+ "[UNK]": 755,
14
+ "a": 11,
15
+ "c": 12,
16
+ "d": 13,
17
+ "e": 14,
18
+ "k": 15,
19
+ "n": 16,
20
+ "o": 17,
21
+ "u": 18,
22
+ "|": 0,
23
+ "ใ„ฑ": 19,
24
+ "ใ„ด": 20,
25
+ "๊ฐ€": 21,
26
+ "๊ฐ": 22,
27
+ "๊ฐ„": 23,
28
+ "๊ฐˆ": 24,
29
+ "๊ฐ": 25,
30
+ "๊ฐ‘": 26,
31
+ "๊ฐ’": 27,
32
+ "๊ฐ”": 28,
33
+ "๊ฐ•": 29,
34
+ "๊ฐ–": 30,
35
+ "๊ฐ™": 31,
36
+ "๊ฐš": 32,
37
+ "๊ฐ›": 33,
38
+ "๊ฐœ": 34,
39
+ "๊ฐ": 35,
40
+ "๊ฐฑ": 36,
41
+ "๊ฑ”": 37,
42
+ "๊ฑฐ": 38,
43
+ "๊ฑด": 39,
44
+ "๊ฑท": 40,
45
+ "๊ฑธ": 41,
46
+ "๊ฒ€": 42,
47
+ "๊ฒ": 43,
48
+ "๊ฒƒ": 44,
49
+ "๊ฒŒ": 45,
50
+ "๊ฒ ": 46,
51
+ "๊ฒจ": 47,
52
+ "๊ฒฉ": 48,
53
+ "๊ฒฌ": 49,
54
+ "๊ฒฐ": 50,
55
+ "๊ฒน": 51,
56
+ "๊ฒผ": 52,
57
+ "๊ฒฝ": 53,
58
+ "๊ณ„": 54,
59
+ "๊ณ ": 55,
60
+ "๊ณก": 56,
61
+ "๊ณค": 57,
62
+ "๊ณจ": 58,
63
+ "๊ณฐ": 59,
64
+ "๊ณฑ": 60,
65
+ "๊ณณ": 61,
66
+ "๊ณต": 62,
67
+ "๊ณผ": 63,
68
+ "๊ด€": 64,
69
+ "๊ด„": 65,
70
+ "๊ด‘": 66,
71
+ "๊ดœ": 67,
72
+ "๊ต‰": 68,
73
+ "๊ต": 69,
74
+ "๊ตฌ": 70,
75
+ "๊ตญ": 71,
76
+ "๊ตฐ": 72,
77
+ "๊ตณ": 73,
78
+ "๊ตด": 74,
79
+ "๊ตฝ": 75,
80
+ "๊ถ": 76,
81
+ "๊ถŒ": 77,
82
+ "๊ทœ": 78,
83
+ "๊ท ": 79,
84
+ "๊ทธ": 80,
85
+ "๊ทน": 81,
86
+ "๊ทผ": 82,
87
+ "๊ทฟ": 83,
88
+ "๊ธ€": 84,
89
+ "๊ธˆ": 85,
90
+ "๊ธ‰": 86,
91
+ "๊ธ": 87,
92
+ "๊ธฐ": 88,
93
+ "๊ธด": 89,
94
+ "๊ธธ": 90,
95
+ "๊น€": 91,
96
+ "๊นŠ": 92,
97
+ "๊นŒ": 93,
98
+ "๊น": 94,
99
+ "๊น”": 95,
100
+ "๊นœ": 96,
101
+ "๊ป": 97,
102
+ "๊ป˜": 98,
103
+ "๊ปด": 99,
104
+ "๊ผˆ": 100,
105
+ "๊ผฌ": 101,
106
+ "๊ผญ": 102,
107
+ "๊พธ": 103,
108
+ "๊ฟ€": 104,
109
+ "๊ฟฐ": 105,
110
+ "๋ˆ": 106,
111
+ "๋Š": 107,
112
+ "๋Œ": 108,
113
+ "๋”": 109,
114
+ "๋": 110,
115
+ "๋ผ": 111,
116
+ "๋‚€": 112,
117
+ "๋‚„": 113,
118
+ "๋‚Œ": 114,
119
+ "๋‚˜": 115,
120
+ "๋‚œ": 116,
121
+ "๋‚ ": 117,
122
+ "๋‚จ": 118,
123
+ "๋‚ฉ": 119,
124
+ "๋‚ซ": 120,
125
+ "๋‚ฌ": 121,
126
+ "๋‚ด": 122,
127
+ "๋‚ธ": 123,
128
+ "๋ƒ‰": 124,
129
+ "๋ƒ": 125,
130
+ "๋ƒฅ": 126,
131
+ "๋„ˆ": 127,
132
+ "๋„˜": 128,
133
+ "๋„ฃ": 129,
134
+ "๋„ค": 130,
135
+ "๋„ท": 131,
136
+ "๋…€": 132,
137
+ "๋…": 133,
138
+ "๋…„": 134,
139
+ "๋…": 135,
140
+ "๋…”": 136,
141
+ "๋…•": 137,
142
+ "๋…ธ": 138,
143
+ "๋…น": 139,
144
+ "๋…ผ": 140,
145
+ "๋†€": 141,
146
+ "๋†ˆ": 142,
147
+ "๋†": 143,
148
+ "๋†’": 144,
149
+ "๋†“": 145,
150
+ "๋†”": 146,
151
+ "๋†จ": 147,
152
+ "๋ˆ„": 148,
153
+ "๋ˆˆ": 149,
154
+ "๋ˆŒ": 150,
155
+ "๋ˆ ": 151,
156
+ "๋ˆด": 152,
157
+ "๋‰˜": 153,
158
+ "๋Š": 154,
159
+ "๋Š”": 155,
160
+ "๋Š˜": 156,
161
+ "๋Šฅ": 157,
162
+ "๋Šฆ": 158,
163
+ "๋Šฌ": 159,
164
+ "๋‹ˆ": 160,
165
+ "๋‹‰": 161,
166
+ "๋‹Œ": 162,
167
+ "๋‹˜": 163,
168
+ "๋‹™": 164,
169
+ "๋‹ค": 165,
170
+ "๋‹ฅ": 166,
171
+ "๋‹จ": 167,
172
+ "๋‹ฌ": 168,
173
+ "๋‹ญ": 169,
174
+ "๋‹ด": 170,
175
+ "๋‹ต": 171,
176
+ "๋‹น": 172,
177
+ "๋Œ€": 173,
178
+ "๋”": 174,
179
+ "๋˜": 175,
180
+ "๋œ": 176,
181
+ "๋Ÿ": 177,
182
+ "๋ฐ": 178,
183
+ "๋„": 179,
184
+ "๋…": 180,
185
+ "๋ˆ": 181,
186
+ "๋Œ": 182,
187
+ "๋™": 183,
188
+ "๋ผ": 184,
189
+ "๋": 185,
190
+ "๋˜": 186,
191
+ "๋œ": 187,
192
+ "๋ ": 188,
193
+ "๋ฉ": 189,
194
+ "๋‘": 190,
195
+ "๋‘˜": 191,
196
+ "๋‘ก": 192,
197
+ "๋’ค": 193,
198
+ "๋’ท": 194,
199
+ "๋“€": 195,
200
+ "๋“œ": 196,
201
+ "๋“": 197,
202
+ "๋“ ": 198,
203
+ "๋“ฃ": 199,
204
+ "๋“ค": 200,
205
+ "๋“ญ": 201,
206
+ "๋“ฑ": 202,
207
+ "๋””": 203,
208
+ "๋”ฉ": 204,
209
+ "๋”ฐ": 205,
210
+ "๋”ฑ": 206,
211
+ "๋”ธ": 207,
212
+ "๋•…": 208,
213
+ "๋•Œ": 209,
214
+ "๋– ": 210,
215
+ "๋–ค": 211,
216
+ "๋–จ": 212,
217
+ "๋–ด": 213,
218
+ "๋–ป": 214,
219
+ "๋˜": 215,
220
+ "๋˜‘": 216,
221
+ "๋šœ": 217,
222
+ "๋šซ": 218,
223
+ "๋›ฐ": 219,
224
+ "๋œฌ": 220,
225
+ "๋œฏ": 221,
226
+ "๋œป": 222,
227
+ "๋ต": 223,
228
+ "๋ผ": 224,
229
+ "๋ฝ": 225,
230
+ "๋ž€": 226,
231
+ "๋žŒ": 227,
232
+ "๋ž": 228,
233
+ "๋ž": 229,
234
+ "๋ž‘": 230,
235
+ "๋ž˜": 231,
236
+ "๋žจ": 232,
237
+ "๋žฉ": 233,
238
+ "๋žฌ": 234,
239
+ "๋Ÿ‰": 235,
240
+ "๋Ÿฌ": 236,
241
+ "๋Ÿฐ": 237,
242
+ "๋Ÿด": 238,
243
+ "๋Ÿผ": 239,
244
+ "๋Ÿฝ": 240,
245
+ "๋ €": 241,
246
+ "๋ ‡": 242,
247
+ "๋ ˆ": 243,
248
+ "๋ ค": 244,
249
+ "๋ ฅ": 245,
250
+ "๋ จ": 246,
251
+ "๋ ท": 247,
252
+ "๋ ธ": 248,
253
+ "๋ น": 249,
254
+ "๋ก€": 250,
255
+ "๋กœ": 251,
256
+ "๋ก": 252,
257
+ "๋ก ": 253,
258
+ "๋กฌ": 254,
259
+ "๋กญ": 255,
260
+ "๋กฏ": 256,
261
+ "๋กท": 257,
262
+ "๋ขฐ": 258,
263
+ "๋ฃŒ": 259,
264
+ "๋ฃจ": 260,
265
+ "๋ฅ˜": 261,
266
+ "๋ฅ ": 262,
267
+ "๋ฅด": 263,
268
+ "๋ฅธ": 264,
269
+ "๋ฅผ": 265,
270
+ "๋ฆ„": 266,
271
+ "๋ฆ…": 267,
272
+ "๋ฆฌ": 268,
273
+ "๋ฆฐ": 269,
274
+ "๋ฆด": 270,
275
+ "๋ฆผ": 271,
276
+ "๋ฆฝ": 272,
277
+ "๋ง": 273,
278
+ "๋งˆ": 274,
279
+ "๋ง‰": 275,
280
+ "๋งŒ": 276,
281
+ "๋งŽ": 277,
282
+ "๋ง": 278,
283
+ "๋ง›": 279,
284
+ "๋ง": 280,
285
+ "๋งž": 281,
286
+ "๋งค": 282,
287
+ "๋งจ": 283,
288
+ "๋งน": 284,
289
+ "๋งบ": 285,
290
+ "๋จธ": 286,
291
+ "๋จน": 287,
292
+ "๋จผ": 288,
293
+ "๋ฉˆ": 289,
294
+ "๋ฉ": 290,
295
+ "๋ฉ”": 291,
296
+ "๋ฉ˜": 292,
297
+ "๋ฉฐ": 293,
298
+ "๋ฉด": 294,
299
+ "๋ช…": 295,
300
+ "๋ช‡": 296,
301
+ "๋ชจ": 297,
302
+ "๋ชฉ": 298,
303
+ "๋ชฌ": 299,
304
+ "๋ชฐ": 300,
305
+ "๋ชธ": 301,
306
+ "๋ชป": 302,
307
+ "๋ชฝ": 303,
308
+ "๋ฌด": 304,
309
+ "๋ฌธ": 305,
310
+ "๋ฌป": 306,
311
+ "๋ฌผ": 307,
312
+ "๋ญ": 308,
313
+ "๋ญ”": 309,
314
+ "๋ญ˜": 310,
315
+ "๋ญก": 311,
316
+ "๋ญฃ": 312,
317
+ "๋ฏ€": 313,
318
+ "๋ฏ„": 314,
319
+ "๋ฏธ": 315,
320
+ "๋ฏผ": 316,
321
+ "๋ฏฟ": 317,
322
+ "๋ฐ€": 318,
323
+ "๋ฐ": 319,
324
+ "๋ฐ‘": 320,
325
+ "๋ฐ”": 321,
326
+ "๋ฐ•": 322,
327
+ "๋ฐ–": 323,
328
+ "๋ฐ˜": 324,
329
+ "๋ฐ›": 325,
330
+ "๋ฐœ": 326,
331
+ "๋ฐ": 327,
332
+ "๋ฐค": 328,
333
+ "๋ฐฅ": 329,
334
+ "๋ฐฉ": 330,
335
+ "๋ฐฐ": 331,
336
+ "๋ฐฑ": 332,
337
+ "๋ฑ…": 333,
338
+ "๋ฒ„": 334,
339
+ "๋ฒˆ": 335,
340
+ "๋ฒŒ": 336,
341
+ "๋ฒ”": 337,
342
+ "๋ฒ•": 338,
343
+ "๋ฒ ": 339,
344
+ "๋ฒค": 340,
345
+ "๋ฒจ": 341,
346
+ "๋ฒฝ": 342,
347
+ "๋ณ€": 343,
348
+ "๋ณ„": 344,
349
+ "๋ณ‘": 345,
350
+ "๋ณด": 346,
351
+ "๋ณต": 347,
352
+ "๋ณธ": 348,
353
+ "๋ณผ": 349,
354
+ "๋ด…": 350,
355
+ "๋ด": 351,
356
+ "๋ดค": 352,
357
+ "๋ต™": 353,
358
+ "๋ถ€": 354,
359
+ "๋ถ": 355,
360
+ "๋ถ„": 356,
361
+ "๋ถˆ": 357,
362
+ "๋ถ™": 358,
363
+ "๋ธŒ": 359,
364
+ "๋ธ”": 360,
365
+ "๋น„": 361,
366
+ "๋นˆ": 362,
367
+ "๋น›": 363,
368
+ "๋น ": 364,
369
+ "๋นก": 365,
370
+ "๋นจ": 366,
371
+ "๋นผ": 367,
372
+ "๋บ": 368,
373
+ "๋ป”": 369,
374
+ "๋ป—": 370,
375
+ "๋ปฅ": 371,
376
+ "๋ฝ€": 372,
377
+ "๋ฟ": 373,
378
+ "์˜": 374,
379
+ "์œ": 375,
380
+ "์‚ฌ": 376,
381
+ "์‚ฐ": 377,
382
+ "์‚ด": 378,
383
+ "์‚ถ": 379,
384
+ "์‚ผ": 380,
385
+ "์ƒ": 381,
386
+ "์ƒˆ": 382,
387
+ "์ƒ‰": 383,
388
+ "์ƒ˜": 384,
389
+ "์ƒ": 385,
390
+ "์„œ": 386,
391
+ "์„": 387,
392
+ "์„ž": 388,
393
+ "์„ ": 389,
394
+ "์„ค": 390,
395
+ "์„ฏ": 391,
396
+ "์„ฐ": 392,
397
+ "์„ฑ": 393,
398
+ "์„ธ": 394,
399
+ "์„ผ": 395,
400
+ "์…‹": 396,
401
+ "์…”": 397,
402
+ "์…จ": 398,
403
+ "์†Œ": 399,
404
+ "์†": 400,
405
+ "์†": 401,
406
+ "์†”": 402,
407
+ "์†ก": 403,
408
+ "์‡ ": 404,
409
+ "์ˆ˜": 405,
410
+ "์ˆ™": 406,
411
+ "์ˆœ": 407,
412
+ "์ˆ ": 408,
413
+ "์‰ฌ": 409,
414
+ "์‰ฝ": 410,
415
+ "์Šˆ": 411,
416
+ "์Šค": 412,
417
+ "์Šจ": 413,
418
+ "์Šต": 414,
419
+ "์Šท": 415,
420
+ "์‹œ": 416,
421
+ "์‹": 417,
422
+ "์‹ ": 418,
423
+ "์‹ค": 419,
424
+ "์‹ฌ": 420,
425
+ "์‹ญ": 421,
426
+ "์‹ถ": 422,
427
+ "์‹ธ": 423,
428
+ "์‹ผ": 424,
429
+ "์Œ€": 425,
430
+ "์จ": 426,
431
+ "์ผ": 427,
432
+ "์˜": 428,
433
+ "์“ฐ": 429,
434
+ "์“ธ": 430,
435
+ "์”€": 431,
436
+ "์”": 432,
437
+ "์”Œ": 433,
438
+ "์”จ": 434,
439
+ "์”ฉ": 435,
440
+ "์”ฌ": 436,
441
+ "์”ป": 437,
442
+ "์•„": 438,
443
+ "์•…": 439,
444
+ "์•ˆ": 440,
445
+ "์•‰": 441,
446
+ "์•Š": 442,
447
+ "์•Œ": 443,
448
+ "์•”": 444,
449
+ "์•—": 445,
450
+ "์•˜": 446,
451
+ "์•™": 447,
452
+ "์•ž": 448,
453
+ "์• ": 449,
454
+ "์•ก": 450,
455
+ "์•ค": 451,
456
+ "์•ผ": 452,
457
+ "์•ฝ": 453,
458
+ "์–€": 454,
459
+ "์–„": 455,
460
+ "์–‘": 456,
461
+ "์–˜": 457,
462
+ "์–œ": 458,
463
+ "์–ด": 459,
464
+ "์–ต": 460,
465
+ "์–ธ": 461,
466
+ "์–ป": 462,
467
+ "์–ผ": 463,
468
+ "์—„": 464,
469
+ "์—…": 465,
470
+ "์—†": 466,
471
+ "์—‡": 467,
472
+ "์—ˆ": 468,
473
+ "์—": 469,
474
+ "์—”": 470,
475
+ "์—ฌ": 471,
476
+ "์—ญ": 472,
477
+ "์—ฐ": 473,
478
+ "์—ด": 474,
479
+ "์—ผ": 475,
480
+ "์˜€": 476,
481
+ "์˜": 477,
482
+ "์˜†": 478,
483
+ "์˜ˆ": 479,
484
+ "์˜›": 480,
485
+ "์˜ค": 481,
486
+ "์˜จ": 482,
487
+ "์˜ฌ": 483,
488
+ "์˜ฎ": 484,
489
+ "์˜ณ": 485,
490
+ "์˜ต": 486,
491
+ "์™€": 487,
492
+ "์™„": 488,
493
+ "์™”": 489,
494
+ "์™•": 490,
495
+ "์™œ": 491,
496
+ "์™ธ": 492,
497
+ "์™ผ": 493,
498
+ "์š”": 494,
499
+ "์šฉ": 495,
500
+ "์šฐ": 496,
501
+ "์šฑ": 497,
502
+ "์šด": 498,
503
+ "์šธ": 499,
504
+ "์›€": 500,
505
+ "์›ƒ": 501,
506
+ "์›…": 502,
507
+ "์›Œ": 503,
508
+ "์›": 504,
509
+ "์›”": 505,
510
+ "์› ": 506,
511
+ "์›ฌ": 507,
512
+ "์›น": 508,
513
+ "์œ„": 509,
514
+ "์œ—": 510,
515
+ "์œ ": 511,
516
+ "์œก": 512,
517
+ "์œจ": 513,
518
+ "์œผ": 514,
519
+ "์œฝ": 515,
520
+ "์€": 516,
521
+ "์„": 517,
522
+ "์Œ": 518,
523
+ "์": 519,
524
+ "์‘": 520,
525
+ "์˜": 521,
526
+ "์ด": 522,
527
+ "์ต": 523,
528
+ "์ธ": 524,
529
+ "์ผ": 525,
530
+ "์ฝ": 526,
531
+ "์ž„": 527,
532
+ "์ž…": 528,
533
+ "์žˆ": 529,
534
+ "์ž‰": 530,
535
+ "์ž": 531,
536
+ "์ž‘": 532,
537
+ "์ž”": 533,
538
+ "์ž–": 534,
539
+ "์ž˜": 535,
540
+ "์ž ": 536,
541
+ "์žก": 537,
542
+ "์žฅ": 538,
543
+ "์žฌ": 539,
544
+ "์Ÿ": 540,
545
+ "์ €": 541,
546
+ "์ ": 542,
547
+ "์ „": 543,
548
+ "์ ˆ": 544,
549
+ "์ ": 545,
550
+ "์ ‘": 546,
551
+ "์ •": 547,
552
+ "์ œ": 548,
553
+ "์ ": 549,
554
+ "์  ": 550,
555
+ "์ ค": 551,
556
+ "์ ธ": 552,
557
+ "์กŒ": 553,
558
+ "์กฐ": 554,
559
+ "์กฑ": 555,
560
+ "์กด": 556,
561
+ "์ข€": 557,
562
+ "์ข…": 558,
563
+ "์ข‹": 559,
564
+ "์ขŒ": 560,
565
+ "์ฃ„": 561,
566
+ "์ฃ ": 562,
567
+ "์ฃผ": 563,
568
+ "์ค€": 564,
569
+ "์ค„": 565,
570
+ "์ค‘": 566,
571
+ "์ค˜": 567,
572
+ "์คฌ": 568,
573
+ "์ฆˆ": 569,
574
+ "์ฆ": 570,
575
+ "์ฆ˜": 571,
576
+ "์ฆ": 572,
577
+ "์ง€": 573,
578
+ "์ง": 574,
579
+ "์ง„": 575,
580
+ "์งˆ": 576,
581
+ "์ง": 577,
582
+ "์ง‘": 578,
583
+ "์ง“": 579,
584
+ "์ง•": 580,
585
+ "์งœ": 581,
586
+ "์ง": 582,
587
+ "์งค": 583,
588
+ "์งญ": 584,
589
+ "์งฑ": 585,
590
+ "์งธ": 586,
591
+ "์ฉŒ": 587,
592
+ "์ฉœ": 588,
593
+ "์ชผ": 589,
594
+ "์ชฝ": 590,
595
+ "์ญ‰": 591,
596
+ "์ฐŒ": 592,
597
+ "์ฐ": 593,
598
+ "์ฐœ": 594,
599
+ "์ฐจ": 595,
600
+ "์ฐฉ": 596,
601
+ "์ฐฌ": 597,
602
+ "์ฐฎ": 598,
603
+ "์ฐฐ": 599,
604
+ "์ฐธ": 600,
605
+ "์ฐป": 601,
606
+ "์ฐฝ": 602,
607
+ "์ฐพ": 603,
608
+ "์ฑ„": 604,
609
+ "์ฑ…": 605,
610
+ "์ฒ˜": 606,
611
+ "์ฒ™": 607,
612
+ "์ฒœ": 608,
613
+ "์ฒ ": 609,
614
+ "์ฒฉ": 610,
615
+ "์ฒซ": 611,
616
+ "์ฒญ": 612,
617
+ "์ฒด": 613,
618
+ "์ณ": 614,
619
+ "์ณค": 615,
620
+ "์ดˆ": 616,
621
+ "์ด›": 617,
622
+ "์ด": 618,
623
+ "์ดฌ": 619,
624
+ "์ตœ": 620,
625
+ "์ถ”": 621,
626
+ "์ถ˜": 622,
627
+ "์ถœ": 623,
628
+ "์ถค": 624,
629
+ "์ถฅ": 625,
630
+ "์ถฉ": 626,
631
+ "์ถฐ": 627,
632
+ "์ทจ": 628,
633
+ "์ธ ": 629,
634
+ "์ธก": 630,
635
+ "์ธต": 631,
636
+ "์น˜": 632,
637
+ "์น™": 633,
638
+ "์นœ": 634,
639
+ "์น ": 635,
640
+ "์นจ": 636,
641
+ "์นฉ": 637,
642
+ "์นญ": 638,
643
+ "์นด": 639,
644
+ "์บ": 640,
645
+ "์ปค": 641,
646
+ "์ปฅ": 642,
647
+ "์ผ€": 643,
648
+ "์ผœ": 644,
649
+ "์ผฐ": 645,
650
+ "์ฝ”": 646,
651
+ "์ฝ•": 647,
652
+ "์ฝœ": 648,
653
+ "์ฝค": 649,
654
+ "์ฝฉ": 650,
655
+ "์ฟจ": 651,
656
+ "ํฌ": 652,
657
+ "ํฐ": 653,
658
+ "ํด": 654,
659
+ "ํผ": 655,
660
+ "ํ‚ค": 656,
661
+ "ํ‚จ": 657,
662
+ "ํ‚ต": 658,
663
+ "ํƒ€": 659,
664
+ "ํƒ": 660,
665
+ "ํƒ„": 661,
666
+ "ํƒˆ": 662,
667
+ "ํƒ": 663,
668
+ "ํƒœ": 664,
669
+ "ํƒ": 665,
670
+ "ํ„ฐ": 666,
671
+ "ํ„ฑ": 667,
672
+ "ํ…Œ": 668,
673
+ "ํ…": 669,
674
+ "ํ† ": 670,
675
+ "ํ†ต": 671,
676
+ "ํˆฌ": 672,
677
+ "ํˆญ": 673,
678
+ "ํ‰": 674,
679
+ "ํŠธ": 675,
680
+ "ํŠน": 676,
681
+ "ํŠผ": 677,
682
+ "ํ‹€": 678,
683
+ "ํ‹ฐ": 679,
684
+ "ํŒ": 680,
685
+ "ํŒŒ": 681,
686
+ "ํŒ": 682,
687
+ "ํŒ”": 683,
688
+ "ํŒก": 684,
689
+ "ํŒจ": 685,
690
+ "ํผ": 686,
691
+ "ํŽ˜": 687,
692
+ "ํŽœ": 688,
693
+ "ํŽธ": 689,
694
+ "ํŽผ": 690,
695
+ "ํ‰": 691,
696
+ "ํฌ": 692,
697
+ "ํด": 693,
698
+ "ํ‘œ": 694,
699
+ "ํ‘ธ": 695,
700
+ "ํ‘น": 696,
701
+ "ํ’€": 697,
702
+ "ํ”„": 698,
703
+ "ํ”Œ": 699,
704
+ "ํ”ผ": 700,
705
+ "ํ”ฝ": 701,
706
+ "ํ•„": 702,
707
+ "ํ•‘": 703,
708
+ "ํ•˜": 704,
709
+ "ํ•™": 705,
710
+ "ํ•œ": 706,
711
+ "ํ• ": 707,
712
+ "ํ•จ": 708,
713
+ "ํ•ฉ": 709,
714
+ "ํ•ญ": 710,
715
+ "ํ•ด": 711,
716
+ "ํ•ต": 712,
717
+ "ํ–‡": 713,
718
+ "ํ–ˆ": 714,
719
+ "ํ–‰": 715,
720
+ "ํ–ฅ": 716,
721
+ "ํ—ˆ": 717,
722
+ "ํ—Œ": 718,
723
+ "ํ—˜": 719,
724
+ "ํ—ค": 720,
725
+ "ํ—ท": 721,
726
+ "ํ˜€": 722,
727
+ "ํ˜„": 723,
728
+ "ํ˜‘": 724,
729
+ "ํ˜•": 725,
730
+ "ํ˜œ": 726,
731
+ "ํ˜ธ": 727,
732
+ "ํ˜น": 728,
733
+ "ํ˜ผ": 729,
734
+ "ํ™€": 730,
735
+ "ํ™": 731,
736
+ "ํ™”": 732,
737
+ "ํ™•": 733,
738
+ "ํ™˜": 734,
739
+ "ํ™œ": 735,
740
+ "ํ™ฉ": 736,
741
+ "ํšŒ": 737,
742
+ "ํš": 738,
743
+ "ํšจ": 739,
744
+ "ํ›„": 740,
745
+ "ํ›ˆ": 741,
746
+ "ํ›จ": 742,
747
+ "ํœด": 743,
748
+ "ํ‰": 744,
749
+ "ํ": 745,
750
+ "ํ”": 746,
751
+ "ํ ": 747,
752
+ "ํก": 748,
753
+ "ํฅ": 749,
754
+ "ํฌ": 750,
755
+ "ํžˆ": 751,
756
+ "ํžŒ": 752,
757
+ "ํž˜": 753,
758
+ "ํž™": 754
759
+ }
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": 0.1,
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": false,
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": 0.1,
54
+ "feat_quantizer_dropout": 0.0,
55
+ "final_dropout": 0.0,
56
+ "gradient_checkpointing": false,
57
+ "hidden_act": "gelu",
58
+ "hidden_dropout": 0.1,
59
+ "hidden_size": 1024,
60
+ "initializer_range": 0.02,
61
+ "intermediate_size": 4096,
62
+ "layer_norm_eps": 1e-05,
63
+ "layerdrop": 0.1,
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": 64,
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": 0.3,
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": 756,
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": "4.33.1",
114
+ "use_weighted_layer_sum": false,
115
+ "vocab_size": 759,
116
+ "xvector_output_dim": 512
117
+ }
eval_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 50.0,
3
+ "eval_cer": 0.35694765608576834,
4
+ "eval_loss": 1.5648640394210815,
5
+ "eval_runtime": 2.3295,
6
+ "eval_samples": 178,
7
+ "eval_samples_per_second": 76.41,
8
+ "eval_steps_per_second": 5.151
9
+ }
preprocessor_config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "feature_extractor_type": "Wav2Vec2FeatureExtractor",
4
+ "feature_size": 1,
5
+ "padding_side": "right",
6
+ "padding_value": 0,
7
+ "processor_class": "Wav2Vec2Processor",
8
+ "return_attention_mask": true,
9
+ "sampling_rate": 16000
10
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:97c86e323d40ef74a32ad8abfac0ae57d2d2f3a5d9d6c9e1cf59fa08fee1a0d1
3
+ size 1265013741
special_tokens_map.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": true,
7
+ "rstrip": false,
8
+ "single_word": false
9
+ },
10
+ {
11
+ "content": "</s>",
12
+ "lstrip": false,
13
+ "normalized": true,
14
+ "rstrip": false,
15
+ "single_word": false
16
+ },
17
+ {
18
+ "content": "<s>",
19
+ "lstrip": false,
20
+ "normalized": true,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ },
24
+ {
25
+ "content": "</s>",
26
+ "lstrip": false,
27
+ "normalized": true,
28
+ "rstrip": false,
29
+ "single_word": false
30
+ }
31
+ ],
32
+ "bos_token": "<s>",
33
+ "eos_token": "</s>",
34
+ "pad_token": "[PAD]",
35
+ "unk_token": "[UNK]"
36
+ }
tokenizer_config.json ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "clean_up_tokenization_spaces": true,
4
+ "config": null,
5
+ "do_lower_case": false,
6
+ "eos_token": "</s>",
7
+ "model_max_length": 1000000000000000019884624838656,
8
+ "pad_token": "[PAD]",
9
+ "processor_class": "Wav2Vec2Processor",
10
+ "replace_word_delimiter_char": " ",
11
+ "target_lang": null,
12
+ "tokenizer_class": "Wav2Vec2CTCTokenizer",
13
+ "tokenizer_type": "wav2vec2",
14
+ "trust_remote_code": false,
15
+ "unk_token": "[UNK]",
16
+ "word_delimiter_token": "|"
17
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 50.0,
3
+ "train_loss": 3.618984514872233,
4
+ "train_runtime": 1635.0894,
5
+ "train_samples": 1427,
6
+ "train_samples_per_second": 43.637,
7
+ "train_steps_per_second": 1.376
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,346 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 50.0,
5
+ "eval_steps": 1000,
6
+ "global_step": 2250,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 1.0,
13
+ "learning_rate": 6.500000000000001e-05,
14
+ "loss": 42.9058,
15
+ "step": 45
16
+ },
17
+ {
18
+ "epoch": 2.0,
19
+ "learning_rate": 0.00014000000000000001,
20
+ "loss": 21.7572,
21
+ "step": 90
22
+ },
23
+ {
24
+ "epoch": 3.0,
25
+ "learning_rate": 0.000215,
26
+ "loss": 6.6528,
27
+ "step": 135
28
+ },
29
+ {
30
+ "epoch": 4.0,
31
+ "learning_rate": 0.00029,
32
+ "loss": 4.7079,
33
+ "step": 180
34
+ },
35
+ {
36
+ "epoch": 5.0,
37
+ "learning_rate": 0.000365,
38
+ "loss": 4.6824,
39
+ "step": 225
40
+ },
41
+ {
42
+ "epoch": 6.0,
43
+ "learning_rate": 0.00044,
44
+ "loss": 4.6861,
45
+ "step": 270
46
+ },
47
+ {
48
+ "epoch": 7.0,
49
+ "learning_rate": 0.0004976923076923077,
50
+ "loss": 4.6781,
51
+ "step": 315
52
+ },
53
+ {
54
+ "epoch": 8.0,
55
+ "learning_rate": 0.0004861538461538462,
56
+ "loss": 4.657,
57
+ "step": 360
58
+ },
59
+ {
60
+ "epoch": 9.0,
61
+ "learning_rate": 0.00047461538461538463,
62
+ "loss": 4.6171,
63
+ "step": 405
64
+ },
65
+ {
66
+ "epoch": 10.0,
67
+ "learning_rate": 0.00046307692307692304,
68
+ "loss": 4.5707,
69
+ "step": 450
70
+ },
71
+ {
72
+ "epoch": 11.0,
73
+ "learning_rate": 0.00045153846153846156,
74
+ "loss": 4.4978,
75
+ "step": 495
76
+ },
77
+ {
78
+ "epoch": 12.0,
79
+ "learning_rate": 0.00044,
80
+ "loss": 4.4697,
81
+ "step": 540
82
+ },
83
+ {
84
+ "epoch": 13.0,
85
+ "learning_rate": 0.0004284615384615385,
86
+ "loss": 4.4129,
87
+ "step": 585
88
+ },
89
+ {
90
+ "epoch": 14.0,
91
+ "learning_rate": 0.0004169230769230769,
92
+ "loss": 4.3792,
93
+ "step": 630
94
+ },
95
+ {
96
+ "epoch": 15.0,
97
+ "learning_rate": 0.0004053846153846154,
98
+ "loss": 4.3352,
99
+ "step": 675
100
+ },
101
+ {
102
+ "epoch": 16.0,
103
+ "learning_rate": 0.00039384615384615386,
104
+ "loss": 4.3007,
105
+ "step": 720
106
+ },
107
+ {
108
+ "epoch": 17.0,
109
+ "learning_rate": 0.0003823076923076923,
110
+ "loss": 4.2279,
111
+ "step": 765
112
+ },
113
+ {
114
+ "epoch": 18.0,
115
+ "learning_rate": 0.0003707692307692308,
116
+ "loss": 4.1306,
117
+ "step": 810
118
+ },
119
+ {
120
+ "epoch": 19.0,
121
+ "learning_rate": 0.00035923076923076924,
122
+ "loss": 3.9656,
123
+ "step": 855
124
+ },
125
+ {
126
+ "epoch": 20.0,
127
+ "learning_rate": 0.0003476923076923077,
128
+ "loss": 3.62,
129
+ "step": 900
130
+ },
131
+ {
132
+ "epoch": 21.0,
133
+ "learning_rate": 0.00033615384615384616,
134
+ "loss": 3.0615,
135
+ "step": 945
136
+ },
137
+ {
138
+ "epoch": 22.0,
139
+ "learning_rate": 0.0003246153846153846,
140
+ "loss": 2.5849,
141
+ "step": 990
142
+ },
143
+ {
144
+ "epoch": 22.22,
145
+ "eval_cer": 0.5984864410342653,
146
+ "eval_loss": 2.5845649242401123,
147
+ "eval_runtime": 2.3789,
148
+ "eval_samples_per_second": 74.826,
149
+ "eval_steps_per_second": 5.044,
150
+ "step": 1000
151
+ },
152
+ {
153
+ "epoch": 23.0,
154
+ "learning_rate": 0.0003130769230769231,
155
+ "loss": 2.1788,
156
+ "step": 1035
157
+ },
158
+ {
159
+ "epoch": 24.0,
160
+ "learning_rate": 0.00030153846153846154,
161
+ "loss": 1.9325,
162
+ "step": 1080
163
+ },
164
+ {
165
+ "epoch": 25.0,
166
+ "learning_rate": 0.00029,
167
+ "loss": 1.7492,
168
+ "step": 1125
169
+ },
170
+ {
171
+ "epoch": 26.0,
172
+ "learning_rate": 0.00027846153846153846,
173
+ "loss": 1.6023,
174
+ "step": 1170
175
+ },
176
+ {
177
+ "epoch": 27.0,
178
+ "learning_rate": 0.0002669230769230769,
179
+ "loss": 1.4718,
180
+ "step": 1215
181
+ },
182
+ {
183
+ "epoch": 28.0,
184
+ "learning_rate": 0.0002553846153846154,
185
+ "loss": 1.3751,
186
+ "step": 1260
187
+ },
188
+ {
189
+ "epoch": 29.0,
190
+ "learning_rate": 0.00024384615384615384,
191
+ "loss": 1.2758,
192
+ "step": 1305
193
+ },
194
+ {
195
+ "epoch": 30.0,
196
+ "learning_rate": 0.0002323076923076923,
197
+ "loss": 1.2088,
198
+ "step": 1350
199
+ },
200
+ {
201
+ "epoch": 31.0,
202
+ "learning_rate": 0.00022076923076923076,
203
+ "loss": 1.1447,
204
+ "step": 1395
205
+ },
206
+ {
207
+ "epoch": 32.0,
208
+ "learning_rate": 0.00020923076923076922,
209
+ "loss": 1.07,
210
+ "step": 1440
211
+ },
212
+ {
213
+ "epoch": 33.0,
214
+ "learning_rate": 0.0001976923076923077,
215
+ "loss": 1.0118,
216
+ "step": 1485
217
+ },
218
+ {
219
+ "epoch": 34.0,
220
+ "learning_rate": 0.00018615384615384615,
221
+ "loss": 0.9947,
222
+ "step": 1530
223
+ },
224
+ {
225
+ "epoch": 35.0,
226
+ "learning_rate": 0.00017461538461538463,
227
+ "loss": 0.9557,
228
+ "step": 1575
229
+ },
230
+ {
231
+ "epoch": 36.0,
232
+ "learning_rate": 0.00016307692307692307,
233
+ "loss": 0.8984,
234
+ "step": 1620
235
+ },
236
+ {
237
+ "epoch": 37.0,
238
+ "learning_rate": 0.00015153846153846155,
239
+ "loss": 0.8694,
240
+ "step": 1665
241
+ },
242
+ {
243
+ "epoch": 38.0,
244
+ "learning_rate": 0.00014000000000000001,
245
+ "loss": 0.8245,
246
+ "step": 1710
247
+ },
248
+ {
249
+ "epoch": 39.0,
250
+ "learning_rate": 0.00012846153846153848,
251
+ "loss": 0.7914,
252
+ "step": 1755
253
+ },
254
+ {
255
+ "epoch": 40.0,
256
+ "learning_rate": 0.00011692307692307692,
257
+ "loss": 0.7883,
258
+ "step": 1800
259
+ },
260
+ {
261
+ "epoch": 41.0,
262
+ "learning_rate": 0.00010538461538461538,
263
+ "loss": 0.7628,
264
+ "step": 1845
265
+ },
266
+ {
267
+ "epoch": 42.0,
268
+ "learning_rate": 9.384615384615384e-05,
269
+ "loss": 0.7449,
270
+ "step": 1890
271
+ },
272
+ {
273
+ "epoch": 43.0,
274
+ "learning_rate": 8.23076923076923e-05,
275
+ "loss": 0.7114,
276
+ "step": 1935
277
+ },
278
+ {
279
+ "epoch": 44.0,
280
+ "learning_rate": 7.076923076923076e-05,
281
+ "loss": 0.7224,
282
+ "step": 1980
283
+ },
284
+ {
285
+ "epoch": 44.44,
286
+ "eval_cer": 0.3664073996216103,
287
+ "eval_loss": 1.5880260467529297,
288
+ "eval_runtime": 2.3696,
289
+ "eval_samples_per_second": 75.118,
290
+ "eval_steps_per_second": 5.064,
291
+ "step": 2000
292
+ },
293
+ {
294
+ "epoch": 45.0,
295
+ "learning_rate": 5.923076923076923e-05,
296
+ "loss": 0.703,
297
+ "step": 2025
298
+ },
299
+ {
300
+ "epoch": 46.0,
301
+ "learning_rate": 4.769230769230769e-05,
302
+ "loss": 0.6685,
303
+ "step": 2070
304
+ },
305
+ {
306
+ "epoch": 47.0,
307
+ "learning_rate": 3.615384615384615e-05,
308
+ "loss": 0.6755,
309
+ "step": 2115
310
+ },
311
+ {
312
+ "epoch": 48.0,
313
+ "learning_rate": 2.4615384615384616e-05,
314
+ "loss": 0.6411,
315
+ "step": 2160
316
+ },
317
+ {
318
+ "epoch": 49.0,
319
+ "learning_rate": 1.3076923076923077e-05,
320
+ "loss": 0.6266,
321
+ "step": 2205
322
+ },
323
+ {
324
+ "epoch": 50.0,
325
+ "learning_rate": 1.5384615384615385e-06,
326
+ "loss": 0.6484,
327
+ "step": 2250
328
+ },
329
+ {
330
+ "epoch": 50.0,
331
+ "step": 2250,
332
+ "total_flos": 9.519369792641827e+18,
333
+ "train_loss": 3.618984514872233,
334
+ "train_runtime": 1635.0894,
335
+ "train_samples_per_second": 43.637,
336
+ "train_steps_per_second": 1.376
337
+ }
338
+ ],
339
+ "logging_steps": 500,
340
+ "max_steps": 2250,
341
+ "num_train_epochs": 50,
342
+ "save_steps": 500,
343
+ "total_flos": 9.519369792641827e+18,
344
+ "trial_name": null,
345
+ "trial_params": null
346
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:37b6ee5799ff6b41ff6cf71255a8b47d6164c730da9afd8c02d1565ac3df39a9
3
+ size 4027
vocab.json ADDED
@@ -0,0 +1,759 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "0": 1,
3
+ "1": 2,
4
+ "2": 3,
5
+ "3": 4,
6
+ "4": 5,
7
+ "5": 6,
8
+ "6": 7,
9
+ "7": 8,
10
+ "8": 9,
11
+ "9": 10,
12
+ "[PAD]": 756,
13
+ "[UNK]": 755,
14
+ "a": 11,
15
+ "c": 12,
16
+ "d": 13,
17
+ "e": 14,
18
+ "k": 15,
19
+ "n": 16,
20
+ "o": 17,
21
+ "u": 18,
22
+ "|": 0,
23
+ "ใ„ฑ": 19,
24
+ "ใ„ด": 20,
25
+ "๊ฐ€": 21,
26
+ "๊ฐ": 22,
27
+ "๊ฐ„": 23,
28
+ "๊ฐˆ": 24,
29
+ "๊ฐ": 25,
30
+ "๊ฐ‘": 26,
31
+ "๊ฐ’": 27,
32
+ "๊ฐ”": 28,
33
+ "๊ฐ•": 29,
34
+ "๊ฐ–": 30,
35
+ "๊ฐ™": 31,
36
+ "๊ฐš": 32,
37
+ "๊ฐ›": 33,
38
+ "๊ฐœ": 34,
39
+ "๊ฐ": 35,
40
+ "๊ฐฑ": 36,
41
+ "๊ฑ”": 37,
42
+ "๊ฑฐ": 38,
43
+ "๊ฑด": 39,
44
+ "๊ฑท": 40,
45
+ "๊ฑธ": 41,
46
+ "๊ฒ€": 42,
47
+ "๊ฒ": 43,
48
+ "๊ฒƒ": 44,
49
+ "๊ฒŒ": 45,
50
+ "๊ฒ ": 46,
51
+ "๊ฒจ": 47,
52
+ "๊ฒฉ": 48,
53
+ "๊ฒฌ": 49,
54
+ "๊ฒฐ": 50,
55
+ "๊ฒน": 51,
56
+ "๊ฒผ": 52,
57
+ "๊ฒฝ": 53,
58
+ "๊ณ„": 54,
59
+ "๊ณ ": 55,
60
+ "๊ณก": 56,
61
+ "๊ณค": 57,
62
+ "๊ณจ": 58,
63
+ "๊ณฐ": 59,
64
+ "๊ณฑ": 60,
65
+ "๊ณณ": 61,
66
+ "๊ณต": 62,
67
+ "๊ณผ": 63,
68
+ "๊ด€": 64,
69
+ "๊ด„": 65,
70
+ "๊ด‘": 66,
71
+ "๊ดœ": 67,
72
+ "๊ต‰": 68,
73
+ "๊ต": 69,
74
+ "๊ตฌ": 70,
75
+ "๊ตญ": 71,
76
+ "๊ตฐ": 72,
77
+ "๊ตณ": 73,
78
+ "๊ตด": 74,
79
+ "๊ตฝ": 75,
80
+ "๊ถ": 76,
81
+ "๊ถŒ": 77,
82
+ "๊ทœ": 78,
83
+ "๊ท ": 79,
84
+ "๊ทธ": 80,
85
+ "๊ทน": 81,
86
+ "๊ทผ": 82,
87
+ "๊ทฟ": 83,
88
+ "๊ธ€": 84,
89
+ "๊ธˆ": 85,
90
+ "๊ธ‰": 86,
91
+ "๊ธ": 87,
92
+ "๊ธฐ": 88,
93
+ "๊ธด": 89,
94
+ "๊ธธ": 90,
95
+ "๊น€": 91,
96
+ "๊นŠ": 92,
97
+ "๊นŒ": 93,
98
+ "๊น": 94,
99
+ "๊น”": 95,
100
+ "๊นœ": 96,
101
+ "๊ป": 97,
102
+ "๊ป˜": 98,
103
+ "๊ปด": 99,
104
+ "๊ผˆ": 100,
105
+ "๊ผฌ": 101,
106
+ "๊ผญ": 102,
107
+ "๊พธ": 103,
108
+ "๊ฟ€": 104,
109
+ "๊ฟฐ": 105,
110
+ "๋ˆ": 106,
111
+ "๋Š": 107,
112
+ "๋Œ": 108,
113
+ "๋”": 109,
114
+ "๋": 110,
115
+ "๋ผ": 111,
116
+ "๋‚€": 112,
117
+ "๋‚„": 113,
118
+ "๋‚Œ": 114,
119
+ "๋‚˜": 115,
120
+ "๋‚œ": 116,
121
+ "๋‚ ": 117,
122
+ "๋‚จ": 118,
123
+ "๋‚ฉ": 119,
124
+ "๋‚ซ": 120,
125
+ "๋‚ฌ": 121,
126
+ "๋‚ด": 122,
127
+ "๋‚ธ": 123,
128
+ "๋ƒ‰": 124,
129
+ "๋ƒ": 125,
130
+ "๋ƒฅ": 126,
131
+ "๋„ˆ": 127,
132
+ "๋„˜": 128,
133
+ "๋„ฃ": 129,
134
+ "๋„ค": 130,
135
+ "๋„ท": 131,
136
+ "๋…€": 132,
137
+ "๋…": 133,
138
+ "๋…„": 134,
139
+ "๋…": 135,
140
+ "๋…”": 136,
141
+ "๋…•": 137,
142
+ "๋…ธ": 138,
143
+ "๋…น": 139,
144
+ "๋…ผ": 140,
145
+ "๋†€": 141,
146
+ "๋†ˆ": 142,
147
+ "๋†": 143,
148
+ "๋†’": 144,
149
+ "๋†“": 145,
150
+ "๋†”": 146,
151
+ "๋†จ": 147,
152
+ "๋ˆ„": 148,
153
+ "๋ˆˆ": 149,
154
+ "๋ˆŒ": 150,
155
+ "๋ˆ ": 151,
156
+ "๋ˆด": 152,
157
+ "๋‰˜": 153,
158
+ "๋Š": 154,
159
+ "๋Š”": 155,
160
+ "๋Š˜": 156,
161
+ "๋Šฅ": 157,
162
+ "๋Šฆ": 158,
163
+ "๋Šฌ": 159,
164
+ "๋‹ˆ": 160,
165
+ "๋‹‰": 161,
166
+ "๋‹Œ": 162,
167
+ "๋‹˜": 163,
168
+ "๋‹™": 164,
169
+ "๋‹ค": 165,
170
+ "๋‹ฅ": 166,
171
+ "๋‹จ": 167,
172
+ "๋‹ฌ": 168,
173
+ "๋‹ญ": 169,
174
+ "๋‹ด": 170,
175
+ "๋‹ต": 171,
176
+ "๋‹น": 172,
177
+ "๋Œ€": 173,
178
+ "๋”": 174,
179
+ "๋˜": 175,
180
+ "๋œ": 176,
181
+ "๋Ÿ": 177,
182
+ "๋ฐ": 178,
183
+ "๋„": 179,
184
+ "๋…": 180,
185
+ "๋ˆ": 181,
186
+ "๋Œ": 182,
187
+ "๋™": 183,
188
+ "๋ผ": 184,
189
+ "๋": 185,
190
+ "๋˜": 186,
191
+ "๋œ": 187,
192
+ "๋ ": 188,
193
+ "๋ฉ": 189,
194
+ "๋‘": 190,
195
+ "๋‘˜": 191,
196
+ "๋‘ก": 192,
197
+ "๋’ค": 193,
198
+ "๋’ท": 194,
199
+ "๋“€": 195,
200
+ "๋“œ": 196,
201
+ "๋“": 197,
202
+ "๋“ ": 198,
203
+ "๋“ฃ": 199,
204
+ "๋“ค": 200,
205
+ "๋“ญ": 201,
206
+ "๋“ฑ": 202,
207
+ "๋””": 203,
208
+ "๋”ฉ": 204,
209
+ "๋”ฐ": 205,
210
+ "๋”ฑ": 206,
211
+ "๋”ธ": 207,
212
+ "๋•…": 208,
213
+ "๋•Œ": 209,
214
+ "๋– ": 210,
215
+ "๋–ค": 211,
216
+ "๋–จ": 212,
217
+ "๋–ด": 213,
218
+ "๋–ป": 214,
219
+ "๋˜": 215,
220
+ "๋˜‘": 216,
221
+ "๋šœ": 217,
222
+ "๋šซ": 218,
223
+ "๋›ฐ": 219,
224
+ "๋œฌ": 220,
225
+ "๋œฏ": 221,
226
+ "๋œป": 222,
227
+ "๋ต": 223,
228
+ "๋ผ": 224,
229
+ "๋ฝ": 225,
230
+ "๋ž€": 226,
231
+ "๋žŒ": 227,
232
+ "๋ž": 228,
233
+ "๋ž": 229,
234
+ "๋ž‘": 230,
235
+ "๋ž˜": 231,
236
+ "๋žจ": 232,
237
+ "๋žฉ": 233,
238
+ "๋žฌ": 234,
239
+ "๋Ÿ‰": 235,
240
+ "๋Ÿฌ": 236,
241
+ "๋Ÿฐ": 237,
242
+ "๋Ÿด": 238,
243
+ "๋Ÿผ": 239,
244
+ "๋Ÿฝ": 240,
245
+ "๋ €": 241,
246
+ "๋ ‡": 242,
247
+ "๋ ˆ": 243,
248
+ "๋ ค": 244,
249
+ "๋ ฅ": 245,
250
+ "๋ จ": 246,
251
+ "๋ ท": 247,
252
+ "๋ ธ": 248,
253
+ "๋ น": 249,
254
+ "๋ก€": 250,
255
+ "๋กœ": 251,
256
+ "๋ก": 252,
257
+ "๋ก ": 253,
258
+ "๋กฌ": 254,
259
+ "๋กญ": 255,
260
+ "๋กฏ": 256,
261
+ "๋กท": 257,
262
+ "๋ขฐ": 258,
263
+ "๋ฃŒ": 259,
264
+ "๋ฃจ": 260,
265
+ "๋ฅ˜": 261,
266
+ "๋ฅ ": 262,
267
+ "๋ฅด": 263,
268
+ "๋ฅธ": 264,
269
+ "๋ฅผ": 265,
270
+ "๋ฆ„": 266,
271
+ "๋ฆ…": 267,
272
+ "๋ฆฌ": 268,
273
+ "๋ฆฐ": 269,
274
+ "๋ฆด": 270,
275
+ "๋ฆผ": 271,
276
+ "๋ฆฝ": 272,
277
+ "๋ง": 273,
278
+ "๋งˆ": 274,
279
+ "๋ง‰": 275,
280
+ "๋งŒ": 276,
281
+ "๋งŽ": 277,
282
+ "๋ง": 278,
283
+ "๋ง›": 279,
284
+ "๋ง": 280,
285
+ "๋งž": 281,
286
+ "๋งค": 282,
287
+ "๋งจ": 283,
288
+ "๋งน": 284,
289
+ "๋งบ": 285,
290
+ "๋จธ": 286,
291
+ "๋จน": 287,
292
+ "๋จผ": 288,
293
+ "๋ฉˆ": 289,
294
+ "๋ฉ": 290,
295
+ "๋ฉ”": 291,
296
+ "๋ฉ˜": 292,
297
+ "๋ฉฐ": 293,
298
+ "๋ฉด": 294,
299
+ "๋ช…": 295,
300
+ "๋ช‡": 296,
301
+ "๋ชจ": 297,
302
+ "๋ชฉ": 298,
303
+ "๋ชฌ": 299,
304
+ "๋ชฐ": 300,
305
+ "๋ชธ": 301,
306
+ "๋ชป": 302,
307
+ "๋ชฝ": 303,
308
+ "๋ฌด": 304,
309
+ "๋ฌธ": 305,
310
+ "๋ฌป": 306,
311
+ "๋ฌผ": 307,
312
+ "๋ญ": 308,
313
+ "๋ญ”": 309,
314
+ "๋ญ˜": 310,
315
+ "๋ญก": 311,
316
+ "๋ญฃ": 312,
317
+ "๋ฏ€": 313,
318
+ "๋ฏ„": 314,
319
+ "๋ฏธ": 315,
320
+ "๋ฏผ": 316,
321
+ "๋ฏฟ": 317,
322
+ "๋ฐ€": 318,
323
+ "๋ฐ": 319,
324
+ "๋ฐ‘": 320,
325
+ "๋ฐ”": 321,
326
+ "๋ฐ•": 322,
327
+ "๋ฐ–": 323,
328
+ "๋ฐ˜": 324,
329
+ "๋ฐ›": 325,
330
+ "๋ฐœ": 326,
331
+ "๋ฐ": 327,
332
+ "๋ฐค": 328,
333
+ "๋ฐฅ": 329,
334
+ "๋ฐฉ": 330,
335
+ "๋ฐฐ": 331,
336
+ "๋ฐฑ": 332,
337
+ "๋ฑ…": 333,
338
+ "๋ฒ„": 334,
339
+ "๋ฒˆ": 335,
340
+ "๋ฒŒ": 336,
341
+ "๋ฒ”": 337,
342
+ "๋ฒ•": 338,
343
+ "๋ฒ ": 339,
344
+ "๋ฒค": 340,
345
+ "๋ฒจ": 341,
346
+ "๋ฒฝ": 342,
347
+ "๋ณ€": 343,
348
+ "๋ณ„": 344,
349
+ "๋ณ‘": 345,
350
+ "๋ณด": 346,
351
+ "๋ณต": 347,
352
+ "๋ณธ": 348,
353
+ "๋ณผ": 349,
354
+ "๋ด…": 350,
355
+ "๋ด": 351,
356
+ "๋ดค": 352,
357
+ "๋ต™": 353,
358
+ "๋ถ€": 354,
359
+ "๋ถ": 355,
360
+ "๋ถ„": 356,
361
+ "๋ถˆ": 357,
362
+ "๋ถ™": 358,
363
+ "๋ธŒ": 359,
364
+ "๋ธ”": 360,
365
+ "๋น„": 361,
366
+ "๋นˆ": 362,
367
+ "๋น›": 363,
368
+ "๋น ": 364,
369
+ "๋นก": 365,
370
+ "๋นจ": 366,
371
+ "๋นผ": 367,
372
+ "๋บ": 368,
373
+ "๋ป”": 369,
374
+ "๋ป—": 370,
375
+ "๋ปฅ": 371,
376
+ "๋ฝ€": 372,
377
+ "๋ฟ": 373,
378
+ "์˜": 374,
379
+ "์œ": 375,
380
+ "์‚ฌ": 376,
381
+ "์‚ฐ": 377,
382
+ "์‚ด": 378,
383
+ "์‚ถ": 379,
384
+ "์‚ผ": 380,
385
+ "์ƒ": 381,
386
+ "์ƒˆ": 382,
387
+ "์ƒ‰": 383,
388
+ "์ƒ˜": 384,
389
+ "์ƒ": 385,
390
+ "์„œ": 386,
391
+ "์„": 387,
392
+ "์„ž": 388,
393
+ "์„ ": 389,
394
+ "์„ค": 390,
395
+ "์„ฏ": 391,
396
+ "์„ฐ": 392,
397
+ "์„ฑ": 393,
398
+ "์„ธ": 394,
399
+ "์„ผ": 395,
400
+ "์…‹": 396,
401
+ "์…”": 397,
402
+ "์…จ": 398,
403
+ "์†Œ": 399,
404
+ "์†": 400,
405
+ "์†": 401,
406
+ "์†”": 402,
407
+ "์†ก": 403,
408
+ "์‡ ": 404,
409
+ "์ˆ˜": 405,
410
+ "์ˆ™": 406,
411
+ "์ˆœ": 407,
412
+ "์ˆ ": 408,
413
+ "์‰ฌ": 409,
414
+ "์‰ฝ": 410,
415
+ "์Šˆ": 411,
416
+ "์Šค": 412,
417
+ "์Šจ": 413,
418
+ "์Šต": 414,
419
+ "์Šท": 415,
420
+ "์‹œ": 416,
421
+ "์‹": 417,
422
+ "์‹ ": 418,
423
+ "์‹ค": 419,
424
+ "์‹ฌ": 420,
425
+ "์‹ญ": 421,
426
+ "์‹ถ": 422,
427
+ "์‹ธ": 423,
428
+ "์‹ผ": 424,
429
+ "์Œ€": 425,
430
+ "์จ": 426,
431
+ "์ผ": 427,
432
+ "์˜": 428,
433
+ "์“ฐ": 429,
434
+ "์“ธ": 430,
435
+ "์”€": 431,
436
+ "์”": 432,
437
+ "์”Œ": 433,
438
+ "์”จ": 434,
439
+ "์”ฉ": 435,
440
+ "์”ฌ": 436,
441
+ "์”ป": 437,
442
+ "์•„": 438,
443
+ "์•…": 439,
444
+ "์•ˆ": 440,
445
+ "์•‰": 441,
446
+ "์•Š": 442,
447
+ "์•Œ": 443,
448
+ "์•”": 444,
449
+ "์•—": 445,
450
+ "์•˜": 446,
451
+ "์•™": 447,
452
+ "์•ž": 448,
453
+ "์• ": 449,
454
+ "์•ก": 450,
455
+ "์•ค": 451,
456
+ "์•ผ": 452,
457
+ "์•ฝ": 453,
458
+ "์–€": 454,
459
+ "์–„": 455,
460
+ "์–‘": 456,
461
+ "์–˜": 457,
462
+ "์–œ": 458,
463
+ "์–ด": 459,
464
+ "์–ต": 460,
465
+ "์–ธ": 461,
466
+ "์–ป": 462,
467
+ "์–ผ": 463,
468
+ "์—„": 464,
469
+ "์—…": 465,
470
+ "์—†": 466,
471
+ "์—‡": 467,
472
+ "์—ˆ": 468,
473
+ "์—": 469,
474
+ "์—”": 470,
475
+ "์—ฌ": 471,
476
+ "์—ญ": 472,
477
+ "์—ฐ": 473,
478
+ "์—ด": 474,
479
+ "์—ผ": 475,
480
+ "์˜€": 476,
481
+ "์˜": 477,
482
+ "์˜†": 478,
483
+ "์˜ˆ": 479,
484
+ "์˜›": 480,
485
+ "์˜ค": 481,
486
+ "์˜จ": 482,
487
+ "์˜ฌ": 483,
488
+ "์˜ฎ": 484,
489
+ "์˜ณ": 485,
490
+ "์˜ต": 486,
491
+ "์™€": 487,
492
+ "์™„": 488,
493
+ "์™”": 489,
494
+ "์™•": 490,
495
+ "์™œ": 491,
496
+ "์™ธ": 492,
497
+ "์™ผ": 493,
498
+ "์š”": 494,
499
+ "์šฉ": 495,
500
+ "์šฐ": 496,
501
+ "์šฑ": 497,
502
+ "์šด": 498,
503
+ "์šธ": 499,
504
+ "์›€": 500,
505
+ "์›ƒ": 501,
506
+ "์›…": 502,
507
+ "์›Œ": 503,
508
+ "์›": 504,
509
+ "์›”": 505,
510
+ "์› ": 506,
511
+ "์›ฌ": 507,
512
+ "์›น": 508,
513
+ "์œ„": 509,
514
+ "์œ—": 510,
515
+ "์œ ": 511,
516
+ "์œก": 512,
517
+ "์œจ": 513,
518
+ "์œผ": 514,
519
+ "์œฝ": 515,
520
+ "์€": 516,
521
+ "์„": 517,
522
+ "์Œ": 518,
523
+ "์": 519,
524
+ "์‘": 520,
525
+ "์˜": 521,
526
+ "์ด": 522,
527
+ "์ต": 523,
528
+ "์ธ": 524,
529
+ "์ผ": 525,
530
+ "์ฝ": 526,
531
+ "์ž„": 527,
532
+ "์ž…": 528,
533
+ "์žˆ": 529,
534
+ "์ž‰": 530,
535
+ "์ž": 531,
536
+ "์ž‘": 532,
537
+ "์ž”": 533,
538
+ "์ž–": 534,
539
+ "์ž˜": 535,
540
+ "์ž ": 536,
541
+ "์žก": 537,
542
+ "์žฅ": 538,
543
+ "์žฌ": 539,
544
+ "์Ÿ": 540,
545
+ "์ €": 541,
546
+ "์ ": 542,
547
+ "์ „": 543,
548
+ "์ ˆ": 544,
549
+ "์ ": 545,
550
+ "์ ‘": 546,
551
+ "์ •": 547,
552
+ "์ œ": 548,
553
+ "์ ": 549,
554
+ "์  ": 550,
555
+ "์ ค": 551,
556
+ "์ ธ": 552,
557
+ "์กŒ": 553,
558
+ "์กฐ": 554,
559
+ "์กฑ": 555,
560
+ "์กด": 556,
561
+ "์ข€": 557,
562
+ "์ข…": 558,
563
+ "์ข‹": 559,
564
+ "์ขŒ": 560,
565
+ "์ฃ„": 561,
566
+ "์ฃ ": 562,
567
+ "์ฃผ": 563,
568
+ "์ค€": 564,
569
+ "์ค„": 565,
570
+ "์ค‘": 566,
571
+ "์ค˜": 567,
572
+ "์คฌ": 568,
573
+ "์ฆˆ": 569,
574
+ "์ฆ": 570,
575
+ "์ฆ˜": 571,
576
+ "์ฆ": 572,
577
+ "์ง€": 573,
578
+ "์ง": 574,
579
+ "์ง„": 575,
580
+ "์งˆ": 576,
581
+ "์ง": 577,
582
+ "์ง‘": 578,
583
+ "์ง“": 579,
584
+ "์ง•": 580,
585
+ "์งœ": 581,
586
+ "์ง": 582,
587
+ "์งค": 583,
588
+ "์งญ": 584,
589
+ "์งฑ": 585,
590
+ "์งธ": 586,
591
+ "์ฉŒ": 587,
592
+ "์ฉœ": 588,
593
+ "์ชผ": 589,
594
+ "์ชฝ": 590,
595
+ "์ญ‰": 591,
596
+ "์ฐŒ": 592,
597
+ "์ฐ": 593,
598
+ "์ฐœ": 594,
599
+ "์ฐจ": 595,
600
+ "์ฐฉ": 596,
601
+ "์ฐฌ": 597,
602
+ "์ฐฎ": 598,
603
+ "์ฐฐ": 599,
604
+ "์ฐธ": 600,
605
+ "์ฐป": 601,
606
+ "์ฐฝ": 602,
607
+ "์ฐพ": 603,
608
+ "์ฑ„": 604,
609
+ "์ฑ…": 605,
610
+ "์ฒ˜": 606,
611
+ "์ฒ™": 607,
612
+ "์ฒœ": 608,
613
+ "์ฒ ": 609,
614
+ "์ฒฉ": 610,
615
+ "์ฒซ": 611,
616
+ "์ฒญ": 612,
617
+ "์ฒด": 613,
618
+ "์ณ": 614,
619
+ "์ณค": 615,
620
+ "์ดˆ": 616,
621
+ "์ด›": 617,
622
+ "์ด": 618,
623
+ "์ดฌ": 619,
624
+ "์ตœ": 620,
625
+ "์ถ”": 621,
626
+ "์ถ˜": 622,
627
+ "์ถœ": 623,
628
+ "์ถค": 624,
629
+ "์ถฅ": 625,
630
+ "์ถฉ": 626,
631
+ "์ถฐ": 627,
632
+ "์ทจ": 628,
633
+ "์ธ ": 629,
634
+ "์ธก": 630,
635
+ "์ธต": 631,
636
+ "์น˜": 632,
637
+ "์น™": 633,
638
+ "์นœ": 634,
639
+ "์น ": 635,
640
+ "์นจ": 636,
641
+ "์นฉ": 637,
642
+ "์นญ": 638,
643
+ "์นด": 639,
644
+ "์บ": 640,
645
+ "์ปค": 641,
646
+ "์ปฅ": 642,
647
+ "์ผ€": 643,
648
+ "์ผœ": 644,
649
+ "์ผฐ": 645,
650
+ "์ฝ”": 646,
651
+ "์ฝ•": 647,
652
+ "์ฝœ": 648,
653
+ "์ฝค": 649,
654
+ "์ฝฉ": 650,
655
+ "์ฟจ": 651,
656
+ "ํฌ": 652,
657
+ "ํฐ": 653,
658
+ "ํด": 654,
659
+ "ํผ": 655,
660
+ "ํ‚ค": 656,
661
+ "ํ‚จ": 657,
662
+ "ํ‚ต": 658,
663
+ "ํƒ€": 659,
664
+ "ํƒ": 660,
665
+ "ํƒ„": 661,
666
+ "ํƒˆ": 662,
667
+ "ํƒ": 663,
668
+ "ํƒœ": 664,
669
+ "ํƒ": 665,
670
+ "ํ„ฐ": 666,
671
+ "ํ„ฑ": 667,
672
+ "ํ…Œ": 668,
673
+ "ํ…": 669,
674
+ "ํ† ": 670,
675
+ "ํ†ต": 671,
676
+ "ํˆฌ": 672,
677
+ "ํˆญ": 673,
678
+ "ํ‰": 674,
679
+ "ํŠธ": 675,
680
+ "ํŠน": 676,
681
+ "ํŠผ": 677,
682
+ "ํ‹€": 678,
683
+ "ํ‹ฐ": 679,
684
+ "ํŒ": 680,
685
+ "ํŒŒ": 681,
686
+ "ํŒ": 682,
687
+ "ํŒ”": 683,
688
+ "ํŒก": 684,
689
+ "ํŒจ": 685,
690
+ "ํผ": 686,
691
+ "ํŽ˜": 687,
692
+ "ํŽœ": 688,
693
+ "ํŽธ": 689,
694
+ "ํŽผ": 690,
695
+ "ํ‰": 691,
696
+ "ํฌ": 692,
697
+ "ํด": 693,
698
+ "ํ‘œ": 694,
699
+ "ํ‘ธ": 695,
700
+ "ํ‘น": 696,
701
+ "ํ’€": 697,
702
+ "ํ”„": 698,
703
+ "ํ”Œ": 699,
704
+ "ํ”ผ": 700,
705
+ "ํ”ฝ": 701,
706
+ "ํ•„": 702,
707
+ "ํ•‘": 703,
708
+ "ํ•˜": 704,
709
+ "ํ•™": 705,
710
+ "ํ•œ": 706,
711
+ "ํ• ": 707,
712
+ "ํ•จ": 708,
713
+ "ํ•ฉ": 709,
714
+ "ํ•ญ": 710,
715
+ "ํ•ด": 711,
716
+ "ํ•ต": 712,
717
+ "ํ–‡": 713,
718
+ "ํ–ˆ": 714,
719
+ "ํ–‰": 715,
720
+ "ํ–ฅ": 716,
721
+ "ํ—ˆ": 717,
722
+ "ํ—Œ": 718,
723
+ "ํ—˜": 719,
724
+ "ํ—ค": 720,
725
+ "ํ—ท": 721,
726
+ "ํ˜€": 722,
727
+ "ํ˜„": 723,
728
+ "ํ˜‘": 724,
729
+ "ํ˜•": 725,
730
+ "ํ˜œ": 726,
731
+ "ํ˜ธ": 727,
732
+ "ํ˜น": 728,
733
+ "ํ˜ผ": 729,
734
+ "ํ™€": 730,
735
+ "ํ™": 731,
736
+ "ํ™”": 732,
737
+ "ํ™•": 733,
738
+ "ํ™˜": 734,
739
+ "ํ™œ": 735,
740
+ "ํ™ฉ": 736,
741
+ "ํšŒ": 737,
742
+ "ํš": 738,
743
+ "ํšจ": 739,
744
+ "ํ›„": 740,
745
+ "ํ›ˆ": 741,
746
+ "ํ›จ": 742,
747
+ "ํœด": 743,
748
+ "ํ‰": 744,
749
+ "ํ": 745,
750
+ "ํ”": 746,
751
+ "ํ ": 747,
752
+ "ํก": 748,
753
+ "ํฅ": 749,
754
+ "ํฌ": 750,
755
+ "ํžˆ": 751,
756
+ "ํžŒ": 752,
757
+ "ํž˜": 753,
758
+ "ํž™": 754
759
+ }