Hugging Face's logo

visolex
/
phobert-v1-hsd

Text Classification
PyTorch
roberta
vietnamese
hate-speech-detection
offensive-language-detection
custom_code
Eval Results (legacy)
Model card Files
xet
 Community
6
AnnyNguyen commited on
Commit
313fe0d
·
verified ·
1 Parent(s): 2bea19e

Upload trainer_state.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. trainer_state.json +176 -0
trainer_state.json ADDED
@@ -0,0 +1,176 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 6433,
3
+ "best_metric": 0.869164826258844,
4
+ "best_model_checkpoint": "outputs/hate-speech-detection/phobert-v1/checkpoint-6433",
5
+ "epoch": 7.0,
6
+ "eval_steps": 500,
7
+ "global_step": 6433,
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": 3.996723175048828,
15
+ "learning_rate": 1.9996091589167148e-05,
16
+ "loss": 0.4454,
17
+ "step": 918
18
+ },
19
+ {
20
+ "epoch": 1.0,
21
+ "eval_accuracy": 0.8648837209302326,
22
+ "eval_f1": 0.8541055666665719,
23
+ "eval_loss": 0.36985549330711365,
24
+ "eval_precision": 0.8482981349301894,
25
+ "eval_recall": 0.8648837209302326,
26
+ "eval_runtime": 18.8079,
27
+ "eval_samples_per_second": 228.627,
28
+ "eval_steps_per_second": 7.178,
29
+ "step": 919
30
+ },
31
+ {
32
+ "epoch": 1.997823721436344,
33
+ "grad_norm": 5.397847652435303,
34
+ "learning_rate": 1.9982377977480694e-05,
35
+ "loss": 0.3302,
36
+ "step": 1836
37
+ },
38
+ {
39
+ "epoch": 2.0,
40
+ "eval_accuracy": 0.8790697674418605,
41
+ "eval_f1": 0.8639193560209791,
42
+ "eval_loss": 0.3517931401729584,
43
+ "eval_precision": 0.8613674120453191,
44
+ "eval_recall": 0.8790697674418605,
45
+ "eval_runtime": 18.812,
46
+ "eval_samples_per_second": 228.577,
47
+ "eval_steps_per_second": 7.176,
48
+ "step": 1838
49
+ },
50
+ {
51
+ "epoch": 2.996735582154516,
52
+ "grad_norm": 4.222469329833984,
53
+ "learning_rate": 1.9958812963917564e-05,
54
+ "loss": 0.2643,
55
+ "step": 2754
56
+ },
57
+ {
58
+ "epoch": 3.0,
59
+ "eval_accuracy": 0.8790697674418605,
60
+ "eval_f1": 0.8643239770760321,
61
+ "eval_loss": 0.34931325912475586,
62
+ "eval_precision": 0.8637571927551355,
63
+ "eval_recall": 0.8790697674418605,
64
+ "eval_runtime": 18.8951,
65
+ "eval_samples_per_second": 227.573,
66
+ "eval_steps_per_second": 7.145,
67
+ "step": 2757
68
+ },
69
+ {
70
+ "epoch": 3.995647442872688,
71
+ "grad_norm": 21.62336540222168,
72
+ "learning_rate": 1.992541980430111e-05,
73
+ "loss": 0.2096,
74
+ "step": 3672
75
+ },
76
+ {
77
+ "epoch": 4.0,
78
+ "eval_accuracy": 0.868139534883721,
79
+ "eval_f1": 0.8680505813983537,
80
+ "eval_loss": 0.4169940650463104,
81
+ "eval_precision": 0.8679632185986601,
82
+ "eval_recall": 0.868139534883721,
83
+ "eval_runtime": 18.8275,
84
+ "eval_samples_per_second": 228.389,
85
+ "eval_steps_per_second": 7.17,
86
+ "step": 3676
87
+ },
88
+ {
89
+ "epoch": 4.99455930359086,
90
+ "grad_norm": 19.167633056640625,
91
+ "learning_rate": 1.9882231453648264e-05,
92
+ "loss": 0.166,
93
+ "step": 4590
94
+ },
95
+ {
96
+ "epoch": 5.0,
97
+ "eval_accuracy": 0.8748837209302326,
98
+ "eval_f1": 0.8683907897636799,
99
+ "eval_loss": 0.45102834701538086,
100
+ "eval_precision": 0.8642072805764893,
101
+ "eval_recall": 0.8748837209302326,
102
+ "eval_runtime": 18.8141,
103
+ "eval_samples_per_second": 228.552,
104
+ "eval_steps_per_second": 7.175,
105
+ "step": 4595
106
+ },
107
+ {
108
+ "epoch": 5.993471164309032,
109
+ "grad_norm": 14.512689590454102,
110
+ "learning_rate": 1.982929053364693e-05,
111
+ "loss": 0.1268,
112
+ "step": 5508
113
+ },
114
+ {
115
+ "epoch": 6.0,
116
+ "eval_accuracy": 0.8767441860465116,
117
+ "eval_f1": 0.8584101867027553,
118
+ "eval_loss": 0.5603283047676086,
119
+ "eval_precision": 0.8576597180464702,
120
+ "eval_recall": 0.8767441860465116,
121
+ "eval_runtime": 18.8828,
122
+ "eval_samples_per_second": 227.72,
123
+ "eval_steps_per_second": 7.149,
124
+ "step": 5514
125
+ },
126
+ {
127
+ "epoch": 6.992383025027204,
128
+ "grad_norm": 0.22124843299388885,
129
+ "learning_rate": 1.9766649290593513e-05,
130
+ "loss": 0.1009,
131
+ "step": 6426
132
+ },
133
+ {
134
+ "epoch": 7.0,
135
+ "eval_accuracy": 0.88,
136
+ "eval_f1": 0.869164826258844,
137
+ "eval_loss": 0.6517873406410217,
138
+ "eval_precision": 0.8655995996061706,
139
+ "eval_recall": 0.88,
140
+ "eval_runtime": 18.8384,
141
+ "eval_samples_per_second": 228.257,
142
+ "eval_steps_per_second": 7.166,
143
+ "step": 6433
144
+ }
145
+ ],
146
+ "logging_steps": 918,
147
+ "max_steps": 91900,
148
+ "num_input_tokens_seen": 0,
149
+ "num_train_epochs": 100,
150
+ "save_steps": 500,
151
+ "stateful_callbacks": {
152
+ "EarlyStoppingCallback": {
153
+ "args": {
154
+ "early_stopping_patience": 5,
155
+ "early_stopping_threshold": 0.0
156
+ },
157
+ "attributes": {
158
+ "early_stopping_patience_counter": 0
159
+ }
160
+ },
161
+ "TrainerControl": {
162
+ "args": {
163
+ "should_epoch_stop": false,
164
+ "should_evaluate": false,
165
+ "should_log": false,
166
+ "should_save": true,
167
+ "should_training_stop": false
168
+ },
169
+ "attributes": {}
170
+ }
171
+ },
172
+ "total_flos": 0.0,
173
+ "train_batch_size": 32,
174
+ "trial_name": null,
175
+ "trial_params": null
176
+ }