Hugging Face's logo

visolex
/
mbert-hsd

Text Classification
PyTorch
Vietnamese
hate-speech-detection
vietnamese-nlp
offensive-speech
Model card Files
xet
Community
AnnyNguyen commited on
Commit
ccd5c50
·
verified ·
1 Parent(s): 1f7138d

Upload trainer_state.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. trainer_state.json +157 -0
trainer_state.json ADDED
@@ -0,0 +1,157 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 5514,
3
+ "best_metric": 0.861991895146672,
4
+ "best_model_checkpoint": "outputs/hate-speech-detection/mbert/checkpoint-5514",
5
+ "epoch": 6.0,
6
+ "eval_steps": 500,
7
+ "global_step": 5514,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.998911860718172,
14
+ "grad_norm": 4.496480941772461,
15
+ "learning_rate": 1.9996091589167148e-05,
16
+ "loss": 0.4497,
17
+ "step": 918
18
+ },
19
+ {
20
+ "epoch": 1.0,
21
+ "eval_accuracy": 0.8672093023255814,
22
+ "eval_f1": 0.8381693329314983,
23
+ "eval_loss": 0.38074544072151184,
24
+ "eval_precision": 0.839596315015221,
25
+ "eval_recall": 0.8672093023255814,
26
+ "eval_runtime": 19.5248,
27
+ "eval_samples_per_second": 220.233,
28
+ "eval_steps_per_second": 6.914,
29
+ "step": 919
30
+ },
31
+ {
32
+ "epoch": 1.997823721436344,
33
+ "grad_norm": 3.883493185043335,
34
+ "learning_rate": 1.9982377977480694e-05,
35
+ "loss": 0.3285,
36
+ "step": 1836
37
+ },
38
+ {
39
+ "epoch": 2.0,
40
+ "eval_accuracy": 0.871860465116279,
41
+ "eval_f1": 0.8487258188734706,
42
+ "eval_loss": 0.38101404905319214,
43
+ "eval_precision": 0.8488176209662102,
44
+ "eval_recall": 0.871860465116279,
45
+ "eval_runtime": 19.6307,
46
+ "eval_samples_per_second": 219.044,
47
+ "eval_steps_per_second": 6.877,
48
+ "step": 1838
49
+ },
50
+ {
51
+ "epoch": 2.996735582154516,
52
+ "grad_norm": 6.906253814697266,
53
+ "learning_rate": 1.9958812963917564e-05,
54
+ "loss": 0.2625,
55
+ "step": 2754
56
+ },
57
+ {
58
+ "epoch": 3.0,
59
+ "eval_accuracy": 0.8690697674418605,
60
+ "eval_f1": 0.8595973310447063,
61
+ "eval_loss": 0.38094526529312134,
62
+ "eval_precision": 0.8602939065897677,
63
+ "eval_recall": 0.8690697674418605,
64
+ "eval_runtime": 19.6152,
65
+ "eval_samples_per_second": 219.218,
66
+ "eval_steps_per_second": 6.882,
67
+ "step": 2757
68
+ },
69
+ {
70
+ "epoch": 3.995647442872688,
71
+ "grad_norm": 5.721479415893555,
72
+ "learning_rate": 1.992541980430111e-05,
73
+ "loss": 0.2028,
74
+ "step": 3672
75
+ },
76
+ {
77
+ "epoch": 4.0,
78
+ "eval_accuracy": 0.8455813953488373,
79
+ "eval_f1": 0.8481157607486833,
80
+ "eval_loss": 0.4412923753261566,
81
+ "eval_precision": 0.8528228084873011,
82
+ "eval_recall": 0.8455813953488373,
83
+ "eval_runtime": 19.5818,
84
+ "eval_samples_per_second": 219.592,
85
+ "eval_steps_per_second": 6.894,
86
+ "step": 3676
87
+ },
88
+ {
89
+ "epoch": 4.99455930359086,
90
+ "grad_norm": 2.1352882385253906,
91
+ "learning_rate": 1.9882231453648264e-05,
92
+ "loss": 0.1518,
93
+ "step": 4590
94
+ },
95
+ {
96
+ "epoch": 5.0,
97
+ "eval_accuracy": 0.858139534883721,
98
+ "eval_f1": 0.8538781067265747,
99
+ "eval_loss": 0.5418818593025208,
100
+ "eval_precision": 0.8503660980046817,
101
+ "eval_recall": 0.858139534883721,
102
+ "eval_runtime": 19.6103,
103
+ "eval_samples_per_second": 219.273,
104
+ "eval_steps_per_second": 6.884,
105
+ "step": 4595
106
+ },
107
+ {
108
+ "epoch": 5.993471164309032,
109
+ "grad_norm": 1.7267584800720215,
110
+ "learning_rate": 1.982929053364693e-05,
111
+ "loss": 0.1191,
112
+ "step": 5508
113
+ },
114
+ {
115
+ "epoch": 6.0,
116
+ "eval_accuracy": 0.8748837209302326,
117
+ "eval_f1": 0.861991895146672,
118
+ "eval_loss": 0.5352126955986023,
119
+ "eval_precision": 0.85848454601395,
120
+ "eval_recall": 0.8748837209302326,
121
+ "eval_runtime": 40.9448,
122
+ "eval_samples_per_second": 105.02,
123
+ "eval_steps_per_second": 3.297,
124
+ "step": 5514
125
+ }
126
+ ],
127
+ "logging_steps": 918,
128
+ "max_steps": 91900,
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": 0.0,
154
+ "train_batch_size": 32,
155
+ "trial_name": null,
156
+ "trial_params": null
157
+ }