Shiyue Zhang
commited on
Commit
•
173ec19
1
Parent(s):
2eb582c
First model version
Browse files- added_tokens.json +1 -0
- all_results.json +20 -0
- config.json +86 -0
- eval_results.json +9 -0
- preprocessor_config.json +9 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +1 -0
- test_results.json +8 -0
- tokenizer_config.json +1 -0
- train_results.json +8 -0
- trainer_state.json +427 -0
- training_args.bin +3 -0
- vocab.json +1 -0
added_tokens.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"<s>": 63, "</s>": 64}
|
all_results.json
ADDED
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 15.0,
|
3 |
+
"eval_loss": 0.5424522757530212,
|
4 |
+
"eval_runtime": 7.8831,
|
5 |
+
"eval_samples": 264,
|
6 |
+
"eval_samples_per_second": 33.49,
|
7 |
+
"eval_steps_per_second": 4.186,
|
8 |
+
"eval_wer": 0.6865671641791045,
|
9 |
+
"test_loss": 0.5638678669929504,
|
10 |
+
"test_runtime": 8.612,
|
11 |
+
"test_samples": 264,
|
12 |
+
"test_samples_per_second": 30.655,
|
13 |
+
"test_steps_per_second": 3.832,
|
14 |
+
"test_wer": 0.6888412017167382,
|
15 |
+
"train_loss": 1.7177187274751209,
|
16 |
+
"train_runtime": 3292.1945,
|
17 |
+
"train_samples": 4465,
|
18 |
+
"train_samples_per_second": 20.344,
|
19 |
+
"train_steps_per_second": 0.638
|
20 |
+
}
|
config.json
ADDED
@@ -0,0 +1,86 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "facebook/wav2vec2-large-xlsr-53",
|
3 |
+
"activation_dropout": 0.0,
|
4 |
+
"apply_spec_augment": true,
|
5 |
+
"architectures": [
|
6 |
+
"Wav2Vec2ForCTC"
|
7 |
+
],
|
8 |
+
"attention_dropout": 0.0,
|
9 |
+
"bos_token_id": 1,
|
10 |
+
"classifier_proj_size": 256,
|
11 |
+
"codevector_dim": 768,
|
12 |
+
"contrastive_logits_temperature": 0.1,
|
13 |
+
"conv_bias": true,
|
14 |
+
"conv_dim": [
|
15 |
+
512,
|
16 |
+
512,
|
17 |
+
512,
|
18 |
+
512,
|
19 |
+
512,
|
20 |
+
512,
|
21 |
+
512
|
22 |
+
],
|
23 |
+
"conv_kernel": [
|
24 |
+
10,
|
25 |
+
3,
|
26 |
+
3,
|
27 |
+
3,
|
28 |
+
3,
|
29 |
+
2,
|
30 |
+
2
|
31 |
+
],
|
32 |
+
"conv_stride": [
|
33 |
+
5,
|
34 |
+
2,
|
35 |
+
2,
|
36 |
+
2,
|
37 |
+
2,
|
38 |
+
2,
|
39 |
+
2
|
40 |
+
],
|
41 |
+
"ctc_loss_reduction": "mean",
|
42 |
+
"ctc_zero_infinity": false,
|
43 |
+
"diversity_loss_weight": 0.1,
|
44 |
+
"do_stable_layer_norm": true,
|
45 |
+
"eos_token_id": 2,
|
46 |
+
"feat_extract_activation": "gelu",
|
47 |
+
"feat_extract_dropout": 0.0,
|
48 |
+
"feat_extract_norm": "layer",
|
49 |
+
"feat_proj_dropout": 0.0,
|
50 |
+
"feat_quantizer_dropout": 0.0,
|
51 |
+
"final_dropout": 0.0,
|
52 |
+
"hidden_act": "gelu",
|
53 |
+
"hidden_dropout": 0.0,
|
54 |
+
"hidden_size": 1024,
|
55 |
+
"initializer_range": 0.02,
|
56 |
+
"intermediate_size": 4096,
|
57 |
+
"layer_norm_eps": 1e-05,
|
58 |
+
"layerdrop": 0.0,
|
59 |
+
"mask_channel_length": 10,
|
60 |
+
"mask_channel_min_space": 1,
|
61 |
+
"mask_channel_other": 0.0,
|
62 |
+
"mask_channel_prob": 0.0,
|
63 |
+
"mask_channel_selection": "static",
|
64 |
+
"mask_feature_length": 10,
|
65 |
+
"mask_feature_prob": 0.0,
|
66 |
+
"mask_time_length": 9,
|
67 |
+
"mask_time_min_space": 1,
|
68 |
+
"mask_time_other": 0.0,
|
69 |
+
"mask_time_prob": 0.01,
|
70 |
+
"mask_time_selection": "static",
|
71 |
+
"model_type": "wav2vec2",
|
72 |
+
"num_attention_heads": 16,
|
73 |
+
"num_codevector_groups": 2,
|
74 |
+
"num_codevectors_per_group": 320,
|
75 |
+
"num_conv_pos_embedding_groups": 16,
|
76 |
+
"num_conv_pos_embeddings": 128,
|
77 |
+
"num_feat_extract_layers": 7,
|
78 |
+
"num_hidden_layers": 24,
|
79 |
+
"num_negatives": 100,
|
80 |
+
"pad_token_id": 62,
|
81 |
+
"proj_codevector_dim": 768,
|
82 |
+
"torch_dtype": "float32",
|
83 |
+
"transformers_version": "4.12.0.dev0",
|
84 |
+
"use_weighted_layer_sum": false,
|
85 |
+
"vocab_size": 65
|
86 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 15.0,
|
3 |
+
"eval_loss": 0.5424522757530212,
|
4 |
+
"eval_runtime": 7.8831,
|
5 |
+
"eval_samples": 264,
|
6 |
+
"eval_samples_per_second": 33.49,
|
7 |
+
"eval_steps_per_second": 4.186,
|
8 |
+
"eval_wer": 0.6865671641791045
|
9 |
+
}
|
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,
|
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:c9fbbaf8c7fae1806960bf43dcdec04415249ab919505a5a60bd8c4089592c6e
|
3 |
+
size 1262195308
|
special_tokens_map.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "[UNK]", "pad_token": "[PAD]", "additional_special_tokens": [{"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}]}
|
test_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"test_loss": 0.5638678669929504,
|
3 |
+
"test_runtime": 8.612,
|
4 |
+
"test_samples": 264,
|
5 |
+
"test_samples_per_second": 30.655,
|
6 |
+
"test_steps_per_second": 3.832,
|
7 |
+
"test_wer": 0.6888412017167382
|
8 |
+
}
|
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": "|", "special_tokens_map_file": null, "tokenizer_file": null, "name_or_path": "./wav2vec2-chr_voice_private-ep15-lr0.0003-repeat0-mask0.01", "tokenizer_class": "Wav2Vec2CTCTokenizer"}
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 15.0,
|
3 |
+
"train_loss": 1.7177187274751209,
|
4 |
+
"train_runtime": 3292.1945,
|
5 |
+
"train_samples": 4465,
|
6 |
+
"train_samples_per_second": 20.344,
|
7 |
+
"train_steps_per_second": 0.638
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,427 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 15.0,
|
5 |
+
"global_step": 2100,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.36,
|
12 |
+
"eval_loss": 4.132408142089844,
|
13 |
+
"eval_runtime": 7.7892,
|
14 |
+
"eval_samples_per_second": 33.893,
|
15 |
+
"eval_steps_per_second": 4.237,
|
16 |
+
"eval_wer": 1.0,
|
17 |
+
"step": 50
|
18 |
+
},
|
19 |
+
{
|
20 |
+
"epoch": 0.71,
|
21 |
+
"eval_loss": 3.357618570327759,
|
22 |
+
"eval_runtime": 7.7699,
|
23 |
+
"eval_samples_per_second": 33.977,
|
24 |
+
"eval_steps_per_second": 4.247,
|
25 |
+
"eval_wer": 1.0,
|
26 |
+
"step": 100
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 1.07,
|
30 |
+
"eval_loss": 3.0935420989990234,
|
31 |
+
"eval_runtime": 8.3088,
|
32 |
+
"eval_samples_per_second": 31.773,
|
33 |
+
"eval_steps_per_second": 3.972,
|
34 |
+
"eval_wer": 1.0,
|
35 |
+
"step": 150
|
36 |
+
},
|
37 |
+
{
|
38 |
+
"epoch": 1.43,
|
39 |
+
"eval_loss": 3.000765085220337,
|
40 |
+
"eval_runtime": 7.8476,
|
41 |
+
"eval_samples_per_second": 33.641,
|
42 |
+
"eval_steps_per_second": 4.205,
|
43 |
+
"eval_wer": 1.2494669509594882,
|
44 |
+
"step": 200
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 1.79,
|
48 |
+
"eval_loss": 2.8665499687194824,
|
49 |
+
"eval_runtime": 7.9301,
|
50 |
+
"eval_samples_per_second": 33.291,
|
51 |
+
"eval_steps_per_second": 4.161,
|
52 |
+
"eval_wer": 1.0874200426439233,
|
53 |
+
"step": 250
|
54 |
+
},
|
55 |
+
{
|
56 |
+
"epoch": 2.14,
|
57 |
+
"eval_loss": 2.7977161407470703,
|
58 |
+
"eval_runtime": 7.8406,
|
59 |
+
"eval_samples_per_second": 33.671,
|
60 |
+
"eval_steps_per_second": 4.209,
|
61 |
+
"eval_wer": 1.537313432835821,
|
62 |
+
"step": 300
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 2.5,
|
66 |
+
"eval_loss": 2.618842601776123,
|
67 |
+
"eval_runtime": 8.0288,
|
68 |
+
"eval_samples_per_second": 32.882,
|
69 |
+
"eval_steps_per_second": 4.11,
|
70 |
+
"eval_wer": 1.2345415778251598,
|
71 |
+
"step": 350
|
72 |
+
},
|
73 |
+
{
|
74 |
+
"epoch": 2.86,
|
75 |
+
"eval_loss": 2.3096766471862793,
|
76 |
+
"eval_runtime": 8.0346,
|
77 |
+
"eval_samples_per_second": 32.858,
|
78 |
+
"eval_steps_per_second": 4.107,
|
79 |
+
"eval_wer": 1.1279317697228144,
|
80 |
+
"step": 400
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 3.21,
|
84 |
+
"eval_loss": 1.8073012828826904,
|
85 |
+
"eval_runtime": 7.9049,
|
86 |
+
"eval_samples_per_second": 33.397,
|
87 |
+
"eval_steps_per_second": 4.175,
|
88 |
+
"eval_wer": 1.251599147121535,
|
89 |
+
"step": 450
|
90 |
+
},
|
91 |
+
{
|
92 |
+
"epoch": 3.57,
|
93 |
+
"learning_rate": 0.0002409,
|
94 |
+
"loss": 3.5589,
|
95 |
+
"step": 500
|
96 |
+
},
|
97 |
+
{
|
98 |
+
"epoch": 3.57,
|
99 |
+
"eval_loss": 1.3744713068008423,
|
100 |
+
"eval_runtime": 7.9139,
|
101 |
+
"eval_samples_per_second": 33.359,
|
102 |
+
"eval_steps_per_second": 4.17,
|
103 |
+
"eval_wer": 1.0895522388059702,
|
104 |
+
"step": 500
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 3.93,
|
108 |
+
"eval_loss": 1.1971436738967896,
|
109 |
+
"eval_runtime": 8.4443,
|
110 |
+
"eval_samples_per_second": 31.264,
|
111 |
+
"eval_steps_per_second": 3.908,
|
112 |
+
"eval_wer": 1.2921108742004264,
|
113 |
+
"step": 550
|
114 |
+
},
|
115 |
+
{
|
116 |
+
"epoch": 4.29,
|
117 |
+
"eval_loss": 1.0361448526382446,
|
118 |
+
"eval_runtime": 7.8907,
|
119 |
+
"eval_samples_per_second": 33.457,
|
120 |
+
"eval_steps_per_second": 4.182,
|
121 |
+
"eval_wer": 0.9872068230277186,
|
122 |
+
"step": 600
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 4.64,
|
126 |
+
"eval_loss": 1.0113328695297241,
|
127 |
+
"eval_runtime": 8.6808,
|
128 |
+
"eval_samples_per_second": 30.412,
|
129 |
+
"eval_steps_per_second": 3.802,
|
130 |
+
"eval_wer": 1.1556503198294243,
|
131 |
+
"step": 650
|
132 |
+
},
|
133 |
+
{
|
134 |
+
"epoch": 5.0,
|
135 |
+
"eval_loss": 0.9761010408401489,
|
136 |
+
"eval_runtime": 8.374,
|
137 |
+
"eval_samples_per_second": 31.526,
|
138 |
+
"eval_steps_per_second": 3.941,
|
139 |
+
"eval_wer": 0.9509594882729211,
|
140 |
+
"step": 700
|
141 |
+
},
|
142 |
+
{
|
143 |
+
"epoch": 5.36,
|
144 |
+
"eval_loss": 0.8795022368431091,
|
145 |
+
"eval_runtime": 8.0897,
|
146 |
+
"eval_samples_per_second": 32.634,
|
147 |
+
"eval_steps_per_second": 4.079,
|
148 |
+
"eval_wer": 1.1279317697228144,
|
149 |
+
"step": 750
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 5.71,
|
153 |
+
"eval_loss": 0.8115519881248474,
|
154 |
+
"eval_runtime": 7.9853,
|
155 |
+
"eval_samples_per_second": 33.061,
|
156 |
+
"eval_steps_per_second": 4.133,
|
157 |
+
"eval_wer": 0.8869936034115139,
|
158 |
+
"step": 800
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 6.07,
|
162 |
+
"eval_loss": 0.7683095932006836,
|
163 |
+
"eval_runtime": 8.0654,
|
164 |
+
"eval_samples_per_second": 32.732,
|
165 |
+
"eval_steps_per_second": 4.092,
|
166 |
+
"eval_wer": 0.9275053304904051,
|
167 |
+
"step": 850
|
168 |
+
},
|
169 |
+
{
|
170 |
+
"epoch": 6.43,
|
171 |
+
"eval_loss": 0.7249290943145752,
|
172 |
+
"eval_runtime": 7.8449,
|
173 |
+
"eval_samples_per_second": 33.652,
|
174 |
+
"eval_steps_per_second": 4.207,
|
175 |
+
"eval_wer": 1.0255863539445629,
|
176 |
+
"step": 900
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 6.79,
|
180 |
+
"eval_loss": 0.7122178077697754,
|
181 |
+
"eval_runtime": 8.1435,
|
182 |
+
"eval_samples_per_second": 32.419,
|
183 |
+
"eval_steps_per_second": 4.052,
|
184 |
+
"eval_wer": 0.9211087420042644,
|
185 |
+
"step": 950
|
186 |
+
},
|
187 |
+
{
|
188 |
+
"epoch": 7.14,
|
189 |
+
"learning_rate": 0.00016634999999999998,
|
190 |
+
"loss": 1.5095,
|
191 |
+
"step": 1000
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 7.14,
|
195 |
+
"eval_loss": 0.7041318416595459,
|
196 |
+
"eval_runtime": 9.0145,
|
197 |
+
"eval_samples_per_second": 29.286,
|
198 |
+
"eval_steps_per_second": 3.661,
|
199 |
+
"eval_wer": 1.0319829424307037,
|
200 |
+
"step": 1000
|
201 |
+
},
|
202 |
+
{
|
203 |
+
"epoch": 7.5,
|
204 |
+
"eval_loss": 0.678531289100647,
|
205 |
+
"eval_runtime": 8.7862,
|
206 |
+
"eval_samples_per_second": 30.047,
|
207 |
+
"eval_steps_per_second": 3.756,
|
208 |
+
"eval_wer": 0.8699360341151386,
|
209 |
+
"step": 1050
|
210 |
+
},
|
211 |
+
{
|
212 |
+
"epoch": 7.86,
|
213 |
+
"eval_loss": 0.7056036591529846,
|
214 |
+
"eval_runtime": 8.0033,
|
215 |
+
"eval_samples_per_second": 32.986,
|
216 |
+
"eval_steps_per_second": 4.123,
|
217 |
+
"eval_wer": 0.9680170575692963,
|
218 |
+
"step": 1100
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 8.21,
|
222 |
+
"eval_loss": 0.6487303972244263,
|
223 |
+
"eval_runtime": 8.7104,
|
224 |
+
"eval_samples_per_second": 30.309,
|
225 |
+
"eval_steps_per_second": 3.789,
|
226 |
+
"eval_wer": 0.8550106609808102,
|
227 |
+
"step": 1150
|
228 |
+
},
|
229 |
+
{
|
230 |
+
"epoch": 8.57,
|
231 |
+
"eval_loss": 0.5972908139228821,
|
232 |
+
"eval_runtime": 7.951,
|
233 |
+
"eval_samples_per_second": 33.204,
|
234 |
+
"eval_steps_per_second": 4.15,
|
235 |
+
"eval_wer": 0.7889125799573561,
|
236 |
+
"step": 1200
|
237 |
+
},
|
238 |
+
{
|
239 |
+
"epoch": 8.93,
|
240 |
+
"eval_loss": 0.5955255627632141,
|
241 |
+
"eval_runtime": 7.9427,
|
242 |
+
"eval_samples_per_second": 33.238,
|
243 |
+
"eval_steps_per_second": 4.155,
|
244 |
+
"eval_wer": 0.8443496801705757,
|
245 |
+
"step": 1250
|
246 |
+
},
|
247 |
+
{
|
248 |
+
"epoch": 9.29,
|
249 |
+
"eval_loss": 0.5822768211364746,
|
250 |
+
"eval_runtime": 7.8596,
|
251 |
+
"eval_samples_per_second": 33.59,
|
252 |
+
"eval_steps_per_second": 4.199,
|
253 |
+
"eval_wer": 0.8017057569296375,
|
254 |
+
"step": 1300
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 9.64,
|
258 |
+
"eval_loss": 0.5886873006820679,
|
259 |
+
"eval_runtime": 7.8662,
|
260 |
+
"eval_samples_per_second": 33.561,
|
261 |
+
"eval_steps_per_second": 4.195,
|
262 |
+
"eval_wer": 0.7569296375266524,
|
263 |
+
"step": 1350
|
264 |
+
},
|
265 |
+
{
|
266 |
+
"epoch": 10.0,
|
267 |
+
"eval_loss": 0.5869713425636292,
|
268 |
+
"eval_runtime": 7.8909,
|
269 |
+
"eval_samples_per_second": 33.456,
|
270 |
+
"eval_steps_per_second": 4.182,
|
271 |
+
"eval_wer": 0.7569296375266524,
|
272 |
+
"step": 1400
|
273 |
+
},
|
274 |
+
{
|
275 |
+
"epoch": 10.36,
|
276 |
+
"eval_loss": 0.5845889449119568,
|
277 |
+
"eval_runtime": 7.8445,
|
278 |
+
"eval_samples_per_second": 33.654,
|
279 |
+
"eval_steps_per_second": 4.207,
|
280 |
+
"eval_wer": 0.7484008528784648,
|
281 |
+
"step": 1450
|
282 |
+
},
|
283 |
+
{
|
284 |
+
"epoch": 10.71,
|
285 |
+
"learning_rate": 9.18e-05,
|
286 |
+
"loss": 1.1157,
|
287 |
+
"step": 1500
|
288 |
+
},
|
289 |
+
{
|
290 |
+
"epoch": 10.71,
|
291 |
+
"eval_loss": 0.5864734053611755,
|
292 |
+
"eval_runtime": 8.0229,
|
293 |
+
"eval_samples_per_second": 32.906,
|
294 |
+
"eval_steps_per_second": 4.113,
|
295 |
+
"eval_wer": 0.7547974413646056,
|
296 |
+
"step": 1500
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 11.07,
|
300 |
+
"eval_loss": 0.5586370825767517,
|
301 |
+
"eval_runtime": 7.8673,
|
302 |
+
"eval_samples_per_second": 33.557,
|
303 |
+
"eval_steps_per_second": 4.195,
|
304 |
+
"eval_wer": 0.7334754797441365,
|
305 |
+
"step": 1550
|
306 |
+
},
|
307 |
+
{
|
308 |
+
"epoch": 11.43,
|
309 |
+
"eval_loss": 0.5573432445526123,
|
310 |
+
"eval_runtime": 7.8679,
|
311 |
+
"eval_samples_per_second": 33.554,
|
312 |
+
"eval_steps_per_second": 4.194,
|
313 |
+
"eval_wer": 0.744136460554371,
|
314 |
+
"step": 1600
|
315 |
+
},
|
316 |
+
{
|
317 |
+
"epoch": 11.79,
|
318 |
+
"eval_loss": 0.5594019889831543,
|
319 |
+
"eval_runtime": 7.9618,
|
320 |
+
"eval_samples_per_second": 33.158,
|
321 |
+
"eval_steps_per_second": 4.145,
|
322 |
+
"eval_wer": 0.7292110874200426,
|
323 |
+
"step": 1650
|
324 |
+
},
|
325 |
+
{
|
326 |
+
"epoch": 12.14,
|
327 |
+
"eval_loss": 0.5614868998527527,
|
328 |
+
"eval_runtime": 7.8272,
|
329 |
+
"eval_samples_per_second": 33.729,
|
330 |
+
"eval_steps_per_second": 4.216,
|
331 |
+
"eval_wer": 0.7569296375266524,
|
332 |
+
"step": 1700
|
333 |
+
},
|
334 |
+
{
|
335 |
+
"epoch": 12.5,
|
336 |
+
"eval_loss": 0.5569693446159363,
|
337 |
+
"eval_runtime": 8.0892,
|
338 |
+
"eval_samples_per_second": 32.636,
|
339 |
+
"eval_steps_per_second": 4.08,
|
340 |
+
"eval_wer": 0.7654584221748401,
|
341 |
+
"step": 1750
|
342 |
+
},
|
343 |
+
{
|
344 |
+
"epoch": 12.86,
|
345 |
+
"eval_loss": 0.5408880710601807,
|
346 |
+
"eval_runtime": 7.8701,
|
347 |
+
"eval_samples_per_second": 33.545,
|
348 |
+
"eval_steps_per_second": 4.193,
|
349 |
+
"eval_wer": 0.7121535181236673,
|
350 |
+
"step": 1800
|
351 |
+
},
|
352 |
+
{
|
353 |
+
"epoch": 13.21,
|
354 |
+
"eval_loss": 0.5358032584190369,
|
355 |
+
"eval_runtime": 7.8977,
|
356 |
+
"eval_samples_per_second": 33.427,
|
357 |
+
"eval_steps_per_second": 4.178,
|
358 |
+
"eval_wer": 0.6652452025586354,
|
359 |
+
"step": 1850
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 13.57,
|
363 |
+
"eval_loss": 0.5394359827041626,
|
364 |
+
"eval_runtime": 7.8219,
|
365 |
+
"eval_samples_per_second": 33.751,
|
366 |
+
"eval_steps_per_second": 4.219,
|
367 |
+
"eval_wer": 0.6823027718550106,
|
368 |
+
"step": 1900
|
369 |
+
},
|
370 |
+
{
|
371 |
+
"epoch": 13.93,
|
372 |
+
"eval_loss": 0.5434439778327942,
|
373 |
+
"eval_runtime": 7.8516,
|
374 |
+
"eval_samples_per_second": 33.624,
|
375 |
+
"eval_steps_per_second": 4.203,
|
376 |
+
"eval_wer": 0.6993603411513859,
|
377 |
+
"step": 1950
|
378 |
+
},
|
379 |
+
{
|
380 |
+
"epoch": 14.29,
|
381 |
+
"learning_rate": 1.74e-05,
|
382 |
+
"loss": 0.8658,
|
383 |
+
"step": 2000
|
384 |
+
},
|
385 |
+
{
|
386 |
+
"epoch": 14.29,
|
387 |
+
"eval_loss": 0.5396074056625366,
|
388 |
+
"eval_runtime": 7.9269,
|
389 |
+
"eval_samples_per_second": 33.304,
|
390 |
+
"eval_steps_per_second": 4.163,
|
391 |
+
"eval_wer": 0.6823027718550106,
|
392 |
+
"step": 2000
|
393 |
+
},
|
394 |
+
{
|
395 |
+
"epoch": 14.64,
|
396 |
+
"eval_loss": 0.5431792736053467,
|
397 |
+
"eval_runtime": 7.8451,
|
398 |
+
"eval_samples_per_second": 33.651,
|
399 |
+
"eval_steps_per_second": 4.206,
|
400 |
+
"eval_wer": 0.6780383795309168,
|
401 |
+
"step": 2050
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 15.0,
|
405 |
+
"eval_loss": 0.5424522757530212,
|
406 |
+
"eval_runtime": 7.9063,
|
407 |
+
"eval_samples_per_second": 33.391,
|
408 |
+
"eval_steps_per_second": 4.174,
|
409 |
+
"eval_wer": 0.6865671641791045,
|
410 |
+
"step": 2100
|
411 |
+
},
|
412 |
+
{
|
413 |
+
"epoch": 15.0,
|
414 |
+
"step": 2100,
|
415 |
+
"total_flos": 2.38289603930769e+18,
|
416 |
+
"train_loss": 1.7177187274751209,
|
417 |
+
"train_runtime": 3292.1945,
|
418 |
+
"train_samples_per_second": 20.344,
|
419 |
+
"train_steps_per_second": 0.638
|
420 |
+
}
|
421 |
+
],
|
422 |
+
"max_steps": 2100,
|
423 |
+
"num_train_epochs": 15,
|
424 |
+
"total_flos": 2.38289603930769e+18,
|
425 |
+
"trial_name": null,
|
426 |
+
"trial_params": null
|
427 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d0c99ff9e12aa55a94bfb7bbe4e77587309782cf5256f9afb21a7656a737b0b6
|
3 |
+
size 2863
|
vocab.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{":": 1, ";": 2, "a": 3, "b": 4, "c": 5, "d": 6, "e": 7, "g": 8, "h": 9, "i": 10, "j": 11, "k": 12, "l": 13, "m": 14, "n": 15, "o": 16, "r": 17, "s": 18, "t": 19, "u": 20, "v": 21, "w": 22, "y": 23, "à": 24, "á": 25, "â": 26, "è": 27, "é": 28, "ê": 29, "ì": 30, "í": 31, "î": 32, "ò": 33, "ó": 34, "ô": 35, "ö": 36, "ù": 37, "ú": 38, "û": 39, "ā": 40, "ē": 41, "ě": 42, "ī": 43, "ō": 44, "ő": 45, "ū": 46, "ű": 47, "ǎ": 48, "ǐ": 49, "ǒ": 50, "ǔ": 51, "ɂ": 52, "̀": 53, "́": 54, "̂": 55, "̄": 56, "̋": 57, "̌": 58, "“": 59, "”": 60, "|": 0, "[UNK]": 61, "[PAD]": 62}
|