Hugging Face's logo

JessicaOjo
/
mt5-asr-corrector-mlm

Safetensors
Mulam
mt5
asr
speech-correction
multilingual
Model card Files
xet
Community
JessicaOjo commited on
Commit
c65a074
·
verified ·
1 Parent(s): f3baa4d

Upload trainer_state.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. trainer_state.json +128 -0
trainer_state.json ADDED
@@ -0,0 +1,128 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1143,
3
+ "best_metric": 1.135260820388794,
4
+ "best_model_checkpoint": "./mt5_base/indic_loss/mlm/checkpoint-1143",
5
+ "epoch": 5.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1905,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 1.0,
14
+ "eval_bleu": 8.4002,
15
+ "eval_gen_len": 19.9018,
16
+ "eval_loss": 1.3717433214187622,
17
+ "eval_runtime": 43.2354,
18
+ "eval_samples_per_second": 22.135,
19
+ "eval_steps_per_second": 2.776,
20
+ "eval_wer": 0.6943,
21
+ "step": 381
22
+ },
23
+ {
24
+ "epoch": 1.3123359580052494,
25
+ "grad_norm": 2.9348294734954834,
26
+ "learning_rate": 4.67257217847769e-05,
27
+ "loss": 2.0859,
28
+ "step": 500
29
+ },
30
+ {
31
+ "epoch": 2.0,
32
+ "eval_bleu": 9.9408,
33
+ "eval_gen_len": 19.9739,
34
+ "eval_loss": 1.248307704925537,
35
+ "eval_runtime": 43.1078,
36
+ "eval_samples_per_second": 22.2,
37
+ "eval_steps_per_second": 2.784,
38
+ "eval_wer": 0.6486,
39
+ "step": 762
40
+ },
41
+ {
42
+ "epoch": 2.6246719160104988,
43
+ "grad_norm": 2.1083076000213623,
44
+ "learning_rate": 4.3444881889763784e-05,
45
+ "loss": 0.6856,
46
+ "step": 1000
47
+ },
48
+ {
49
+ "epoch": 3.0,
50
+ "eval_bleu": 10.3503,
51
+ "eval_gen_len": 19.9739,
52
+ "eval_loss": 1.135260820388794,
53
+ "eval_runtime": 43.0868,
54
+ "eval_samples_per_second": 22.211,
55
+ "eval_steps_per_second": 2.785,
56
+ "eval_wer": 0.6484,
57
+ "step": 1143
58
+ },
59
+ {
60
+ "epoch": 3.937007874015748,
61
+ "grad_norm": 2.1635806560516357,
62
+ "learning_rate": 4.016404199475066e-05,
63
+ "loss": 0.5485,
64
+ "step": 1500
65
+ },
66
+ {
67
+ "epoch": 4.0,
68
+ "eval_bleu": 10.3966,
69
+ "eval_gen_len": 19.9739,
70
+ "eval_loss": 1.1630462408065796,
71
+ "eval_runtime": 43.0721,
72
+ "eval_samples_per_second": 22.219,
73
+ "eval_steps_per_second": 2.786,
74
+ "eval_wer": 0.6476,
75
+ "step": 1524
76
+ },
77
+ {
78
+ "epoch": 5.0,
79
+ "eval_bleu": 10.585,
80
+ "eval_gen_len": 19.9739,
81
+ "eval_loss": 1.1526703834533691,
82
+ "eval_runtime": 42.8826,
83
+ "eval_samples_per_second": 22.317,
84
+ "eval_steps_per_second": 2.798,
85
+ "eval_wer": 0.6481,
86
+ "step": 1905
87
+ },
88
+ {
89
+ "epoch": 5.0,
90
+ "step": 1905,
91
+ "total_flos": 2480404906856448.0,
92
+ "train_loss": 0.9748498230781455,
93
+ "train_runtime": 626.9519,
94
+ "train_samples_per_second": 97.073,
95
+ "train_steps_per_second": 12.154
96
+ }
97
+ ],
98
+ "logging_steps": 500,
99
+ "max_steps": 7620,
100
+ "num_input_tokens_seen": 0,
101
+ "num_train_epochs": 20,
102
+ "save_steps": 50000.0,
103
+ "stateful_callbacks": {
104
+ "EarlyStoppingCallback": {
105
+ "args": {
106
+ "early_stopping_patience": 2,
107
+ "early_stopping_threshold": 0.0
108
+ },
109
+ "attributes": {
110
+ "early_stopping_patience_counter": 2
111
+ }
112
+ },
113
+ "TrainerControl": {
114
+ "args": {
115
+ "should_epoch_stop": false,
116
+ "should_evaluate": false,
117
+ "should_log": false,
118
+ "should_save": true,
119
+ "should_training_stop": true
120
+ },
121
+ "attributes": {}
122
+ }
123
+ },
124
+ "total_flos": 2480404906856448.0,
125
+ "train_batch_size": 8,
126
+ "trial_name": null,
127
+ "trial_params": null
128
+ }