Hugging Face's logo

visolex
/
visobert-hsd

Text Classification
Safetensors
Vietnamese
xlm-roberta
hate-speech-detection
vietnamese
transformer
Eval Results
Model card Files
xet
Community
AnnyNguyen commited on
Commit
78fea98
·
verified ·
1 Parent(s): 38f579f

Upload trainer_state.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. trainer_state.json +155 -0
trainer_state.json ADDED
@@ -0,0 +1,155 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 6461,
3
+ "best_metric": 0.885767355467843,
4
+ "best_model_checkpoint": "outputs/visobert-hsd/checkpoint-best/checkpoint-6461",
5
+ "epoch": 7.0,
6
+ "eval_steps": 500,
7
+ "global_step": 6461,
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
+ "grad_norm": 7.377974510192871,
15
+ "learning_rate": 4.998769072267837e-05,
16
+ "loss": 0.3319,
17
+ "step": 923
18
+ },
19
+ {
20
+ "epoch": 1.0,
21
+ "eval_accuracy": 0.8841420942651498,
22
+ "eval_loss": 0.3208531141281128,
23
+ "eval_runtime": 19.7374,
24
+ "eval_samples_per_second": 218.215,
25
+ "eval_steps_per_second": 6.84,
26
+ "step": 923
27
+ },
28
+ {
29
+ "epoch": 2.0,
30
+ "grad_norm": 5.012074947357178,
31
+ "learning_rate": 4.995072162589518e-05,
32
+ "loss": 0.1938,
33
+ "step": 1846
34
+ },
35
+ {
36
+ "epoch": 2.0,
37
+ "eval_accuracy": 0.885070814952403,
38
+ "eval_loss": 0.41746047139167786,
39
+ "eval_runtime": 19.7546,
40
+ "eval_samples_per_second": 218.025,
41
+ "eval_steps_per_second": 6.834,
42
+ "step": 1846
43
+ },
44
+ {
45
+ "epoch": 3.0,
46
+ "grad_norm": 20.185508728027344,
47
+ "learning_rate": 4.988912917920435e-05,
48
+ "loss": 0.097,
49
+ "step": 2769
50
+ },
51
+ {
52
+ "epoch": 3.0,
53
+ "eval_accuracy": 0.8834455537497098,
54
+ "eval_loss": 0.6088711619377136,
55
+ "eval_runtime": 19.7199,
56
+ "eval_samples_per_second": 218.409,
57
+ "eval_steps_per_second": 6.846,
58
+ "step": 2769
59
+ },
60
+ {
61
+ "epoch": 4.0,
62
+ "grad_norm": 0.14329379796981812,
63
+ "learning_rate": 4.980297416691463e-05,
64
+ "loss": 0.0647,
65
+ "step": 3692
66
+ },
67
+ {
68
+ "epoch": 4.0,
69
+ "eval_accuracy": 0.8632458788019504,
70
+ "eval_loss": 0.6463912129402161,
71
+ "eval_runtime": 19.7704,
72
+ "eval_samples_per_second": 217.851,
73
+ "eval_steps_per_second": 6.828,
74
+ "step": 3692
75
+ },
76
+ {
77
+ "epoch": 5.0,
78
+ "grad_norm": 0.18555864691734314,
79
+ "learning_rate": 4.969234161362153e-05,
80
+ "loss": 0.0456,
81
+ "step": 4615
82
+ },
83
+ {
84
+ "epoch": 5.0,
85
+ "eval_accuracy": 0.8769445089389366,
86
+ "eval_loss": 0.674150288105011,
87
+ "eval_runtime": 19.7446,
88
+ "eval_samples_per_second": 218.136,
89
+ "eval_steps_per_second": 6.837,
90
+ "step": 4615
91
+ },
92
+ {
93
+ "epoch": 6.0,
94
+ "grad_norm": 0.1324630081653595,
95
+ "learning_rate": 4.9557340700298316e-05,
96
+ "loss": 0.0368,
97
+ "step": 5538
98
+ },
99
+ {
100
+ "epoch": 6.0,
101
+ "eval_accuracy": 0.8827490132342698,
102
+ "eval_loss": 0.6759707927703857,
103
+ "eval_runtime": 19.7523,
104
+ "eval_samples_per_second": 218.05,
105
+ "eval_steps_per_second": 6.835,
106
+ "step": 5538
107
+ },
108
+ {
109
+ "epoch": 7.0,
110
+ "grad_norm": 0.013418528251349926,
111
+ "learning_rate": 4.939810465654758e-05,
112
+ "loss": 0.0306,
113
+ "step": 6461
114
+ },
115
+ {
116
+ "epoch": 7.0,
117
+ "eval_accuracy": 0.885767355467843,
118
+ "eval_loss": 0.9015969038009644,
119
+ "eval_runtime": 19.7472,
120
+ "eval_samples_per_second": 218.107,
121
+ "eval_steps_per_second": 6.836,
122
+ "step": 6461
123
+ }
124
+ ],
125
+ "logging_steps": 923,
126
+ "max_steps": 92300,
127
+ "num_input_tokens_seen": 0,
128
+ "num_train_epochs": 100,
129
+ "save_steps": 500,
130
+ "stateful_callbacks": {
131
+ "EarlyStoppingCallback": {
132
+ "args": {
133
+ "early_stopping_patience": 5,
134
+ "early_stopping_threshold": 0.0
135
+ },
136
+ "attributes": {
137
+ "early_stopping_patience_counter": 0
138
+ }
139
+ },
140
+ "TrainerControl": {
141
+ "args": {
142
+ "should_epoch_stop": false,
143
+ "should_evaluate": false,
144
+ "should_log": false,
145
+ "should_save": true,
146
+ "should_training_stop": false
147
+ },
148
+ "attributes": {}
149
+ }
150
+ },
151
+ "total_flos": 2.7186720113581056e+16,
152
+ "train_batch_size": 32,
153
+ "trial_name": null,
154
+ "trial_params": null
155
+ }