Hugging Face's logo

visolex
/
mbert-hsd-span

Token Classification
Safetensors
bert
vietnamese
hate-speech
span-detection
nlp
Eval Results
Model card Files
xet
Community
AnnyNguyen commited on
Commit
d261b5d
·
verified ·
1 Parent(s): c50dc5d

Upload checkpoint-1890/trainer_state.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. checkpoint-1890/trainer_state.json +157 -0
checkpoint-1890/trainer_state.json ADDED
@@ -0,0 +1,157 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1890,
3
+ "best_metric": 0.8640991774392113,
4
+ "best_model_checkpoint": "outputs/runs/mbert/checkpoint-1890",
5
+ "epoch": 6.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1890,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.9968253968253968,
14
+ "grad_norm": 2.8764171600341797,
15
+ "learning_rate": 4.999432333543028e-06,
16
+ "loss": 0.3922,
17
+ "step": 314
18
+ },
19
+ {
20
+ "epoch": 1.0,
21
+ "eval_accuracy": 0.8576303179904168,
22
+ "eval_f1": 0.8463278723071122,
23
+ "eval_loss": 0.34177419543266296,
24
+ "eval_precision": 0.8663865734795833,
25
+ "eval_recall": 0.8368564126768949,
26
+ "eval_runtime": 1.5428,
27
+ "eval_samples_per_second": 843.28,
28
+ "eval_steps_per_second": 26.575,
29
+ "step": 315
30
+ },
31
+ {
32
+ "epoch": 1.9936507936507937,
33
+ "grad_norm": 3.1037893295288086,
34
+ "learning_rate": 4.996525669931999e-06,
35
+ "loss": 0.2939,
36
+ "step": 628
37
+ },
38
+ {
39
+ "epoch": 2.0,
40
+ "eval_accuracy": 0.862621605924205,
41
+ "eval_f1": 0.8567911162439183,
42
+ "eval_loss": 0.30715054273605347,
43
+ "eval_precision": 0.8570806369874044,
44
+ "eval_recall": 0.8565085419444378,
45
+ "eval_runtime": 1.5565,
46
+ "eval_samples_per_second": 835.844,
47
+ "eval_steps_per_second": 26.341,
48
+ "step": 630
49
+ },
50
+ {
51
+ "epoch": 2.9904761904761905,
52
+ "grad_norm": 2.2518210411071777,
53
+ "learning_rate": 4.991155236893945e-06,
54
+ "loss": 0.2786,
55
+ "step": 942
56
+ },
57
+ {
58
+ "epoch": 3.0,
59
+ "eval_accuracy": 0.8685930013068099,
60
+ "eval_f1": 0.8614307154654214,
61
+ "eval_loss": 0.3059610426425934,
62
+ "eval_precision": 0.8669559804904288,
63
+ "eval_recall": 0.8574043561902019,
64
+ "eval_runtime": 1.5618,
65
+ "eval_samples_per_second": 833.027,
66
+ "eval_steps_per_second": 26.252,
67
+ "step": 945
68
+ },
69
+ {
70
+ "epoch": 3.9873015873015873,
71
+ "grad_norm": 5.058016777038574,
72
+ "learning_rate": 4.983326334397891e-06,
73
+ "loss": 0.2663,
74
+ "step": 1256
75
+ },
76
+ {
77
+ "epoch": 4.0,
78
+ "eval_accuracy": 0.8673224916509366,
79
+ "eval_f1": 0.8565473676548805,
80
+ "eval_loss": 0.3232356607913971,
81
+ "eval_precision": 0.8784162677513533,
82
+ "eval_recall": 0.8463548582123183,
83
+ "eval_runtime": 1.5602,
84
+ "eval_samples_per_second": 833.858,
85
+ "eval_steps_per_second": 26.278,
86
+ "step": 1260
87
+ },
88
+ {
89
+ "epoch": 4.984126984126984,
90
+ "grad_norm": 7.031896591186523,
91
+ "learning_rate": 4.97304668862541e-06,
92
+ "loss": 0.2563,
93
+ "step": 1570
94
+ },
95
+ {
96
+ "epoch": 5.0,
97
+ "eval_accuracy": 0.8691012051691593,
98
+ "eval_f1": 0.8625756064004395,
99
+ "eval_loss": 0.300201952457428,
100
+ "eval_precision": 0.8658918299079369,
101
+ "eval_recall": 0.8598985796226895,
102
+ "eval_runtime": 1.5594,
103
+ "eval_samples_per_second": 834.303,
104
+ "eval_steps_per_second": 26.292,
105
+ "step": 1575
106
+ },
107
+ {
108
+ "epoch": 5.980952380952381,
109
+ "grad_norm": 10.076984405517578,
110
+ "learning_rate": 4.9603264443458e-06,
111
+ "loss": 0.2436,
112
+ "step": 1884
113
+ },
114
+ {
115
+ "epoch": 6.0,
116
+ "eval_accuracy": 0.8702991142732684,
117
+ "eval_f1": 0.8640991774392113,
118
+ "eval_loss": 0.29937663674354553,
119
+ "eval_precision": 0.8665212130630453,
120
+ "eval_recall": 0.8620456410474242,
121
+ "eval_runtime": 1.5623,
122
+ "eval_samples_per_second": 832.727,
123
+ "eval_steps_per_second": 26.243,
124
+ "step": 1890
125
+ }
126
+ ],
127
+ "logging_steps": 314,
128
+ "max_steps": 31500,
129
+ "num_input_tokens_seen": 0,
130
+ "num_train_epochs": 100,
131
+ "save_steps": 500,
132
+ "stateful_callbacks": {
133
+ "EarlyStoppingCallback": {
134
+ "args": {
135
+ "early_stopping_patience": 5,
136
+ "early_stopping_threshold": 0.0
137
+ },
138
+ "attributes": {
139
+ "early_stopping_patience_counter": 0
140
+ }
141
+ },
142
+ "TrainerControl": {
143
+ "args": {
144
+ "should_epoch_stop": false,
145
+ "should_evaluate": false,
146
+ "should_log": false,
147
+ "should_save": true,
148
+ "should_training_stop": false
149
+ },
150
+ "attributes": {}
151
+ }
152
+ },
153
+ "total_flos": 1974227645518848.0,
154
+ "train_batch_size": 32,
155
+ "trial_name": null,
156
+ "trial_params": null
157
+ }