Hugging Face's logo

visolex
/
mbert-spam-binary

Text Classification
PyTorch
bert
vietnamese
spam-detection
e-commerce
Eval Results
Model card Files
xet
Community
2
AnnyNguyen commited on
Commit
7e4160d
·
verified ·
1 Parent(s): 3ba0e09

Upload trainer_state.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. trainer_state.json +252 -0
trainer_state.json ADDED
@@ -0,0 +1,252 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 4906,
3
+ "best_metric": 0.8757205622114191,
4
+ "best_model_checkpoint": "outputs/mbert-spam-binary/checkpoint-4906",
5
+ "epoch": 11.0,
6
+ "eval_steps": 500,
7
+ "global_step": 4906,
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": 0.5657051801681519,
15
+ "learning_rate": 4.998771926094749e-05,
16
+ "loss": 0.3358,
17
+ "step": 446
18
+ },
19
+ {
20
+ "epoch": 1.0,
21
+ "eval_accuracy": 0.8972904851921865,
22
+ "eval_f1": 0.8565157558602947,
23
+ "eval_loss": 0.3799394369125366,
24
+ "eval_precision": 0.9093794413813263,
25
+ "eval_recall": 0.8265391650419087,
26
+ "eval_runtime": 1.8927,
27
+ "eval_samples_per_second": 838.483,
28
+ "eval_steps_per_second": 26.417,
29
+ "step": 446
30
+ },
31
+ {
32
+ "epoch": 2.0,
33
+ "grad_norm": 9.706869125366211,
34
+ "learning_rate": 4.995077872180951e-05,
35
+ "loss": 0.239,
36
+ "step": 892
37
+ },
38
+ {
39
+ "epoch": 2.0,
40
+ "eval_accuracy": 0.8853182104599874,
41
+ "eval_f1": 0.8486677201394943,
42
+ "eval_loss": 0.3240993022918701,
43
+ "eval_precision": 0.86617355096017,
44
+ "eval_recall": 0.8351329621515146,
45
+ "eval_runtime": 1.8802,
46
+ "eval_samples_per_second": 844.062,
47
+ "eval_steps_per_second": 26.593,
48
+ "step": 892
49
+ },
50
+ {
51
+ "epoch": 3.0,
52
+ "grad_norm": 5.252721309661865,
53
+ "learning_rate": 4.988921477641713e-05,
54
+ "loss": 0.1874,
55
+ "step": 1338
56
+ },
57
+ {
58
+ "epoch": 3.0,
59
+ "eval_accuracy": 0.8859483301827347,
60
+ "eval_f1": 0.8551266047827633,
61
+ "eval_loss": 0.31036534905433655,
62
+ "eval_precision": 0.8564533765313355,
63
+ "eval_recall": 0.853831078772286,
64
+ "eval_runtime": 1.8827,
65
+ "eval_samples_per_second": 842.955,
66
+ "eval_steps_per_second": 26.558,
67
+ "step": 1338
68
+ },
69
+ {
70
+ "epoch": 4.0,
71
+ "grad_norm": 7.976099967956543,
72
+ "learning_rate": 4.980308818095173e-05,
73
+ "loss": 0.1425,
74
+ "step": 1784
75
+ },
76
+ {
77
+ "epoch": 4.0,
78
+ "eval_accuracy": 0.8645242596093258,
79
+ "eval_f1": 0.833393799625288,
80
+ "eval_loss": 0.3800713121891022,
81
+ "eval_precision": 0.8251404013423738,
82
+ "eval_recall": 0.8435217382565174,
83
+ "eval_runtime": 1.8889,
84
+ "eval_samples_per_second": 840.172,
85
+ "eval_steps_per_second": 26.47,
86
+ "step": 1784
87
+ },
88
+ {
89
+ "epoch": 5.0,
90
+ "grad_norm": 19.17702293395996,
91
+ "learning_rate": 4.9692483931964865e-05,
92
+ "loss": 0.1182,
93
+ "step": 2230
94
+ },
95
+ {
96
+ "epoch": 5.0,
97
+ "eval_accuracy": 0.8928796471329553,
98
+ "eval_f1": 0.8623913575713062,
99
+ "eval_loss": 0.38209661841392517,
100
+ "eval_precision": 0.8681797459435936,
101
+ "eval_recall": 0.857123474905027,
102
+ "eval_runtime": 1.891,
103
+ "eval_samples_per_second": 839.244,
104
+ "eval_steps_per_second": 26.441,
105
+ "step": 2230
106
+ },
107
+ {
108
+ "epoch": 6.0,
109
+ "grad_norm": 0.38656169176101685,
110
+ "learning_rate": 4.9557511182496865e-05,
111
+ "loss": 0.0966,
112
+ "step": 2676
113
+ },
114
+ {
115
+ "epoch": 6.0,
116
+ "eval_accuracy": 0.9042218021424071,
117
+ "eval_f1": 0.8754265647593472,
118
+ "eval_loss": 0.439815878868103,
119
+ "eval_precision": 0.8869994596227571,
120
+ "eval_recall": 0.8656328516009728,
121
+ "eval_runtime": 1.889,
122
+ "eval_samples_per_second": 840.135,
123
+ "eval_steps_per_second": 26.469,
124
+ "step": 2676
125
+ },
126
+ {
127
+ "epoch": 7.0,
128
+ "grad_norm": 15.262068748474121,
129
+ "learning_rate": 4.939830313435599e-05,
130
+ "loss": 0.0787,
131
+ "step": 3122
132
+ },
133
+ {
134
+ "epoch": 7.0,
135
+ "eval_accuracy": 0.8985507246376812,
136
+ "eval_f1": 0.8709438719717468,
137
+ "eval_loss": 0.500139594078064,
138
+ "eval_precision": 0.8729009773098388,
139
+ "eval_recall": 0.8690498683443548,
140
+ "eval_runtime": 1.8913,
141
+ "eval_samples_per_second": 839.098,
142
+ "eval_steps_per_second": 26.437,
143
+ "step": 3122
144
+ },
145
+ {
146
+ "epoch": 8.0,
147
+ "grad_norm": 2.0876944065093994,
148
+ "learning_rate": 4.9215016906664344e-05,
149
+ "loss": 0.0827,
150
+ "step": 3568
151
+ },
152
+ {
153
+ "epoch": 8.0,
154
+ "eval_accuracy": 0.8979206049149339,
155
+ "eval_f1": 0.8670210785043945,
156
+ "eval_loss": 0.4308617115020752,
157
+ "eval_precision": 0.8790773631628384,
158
+ "eval_recall": 0.8569274989447448,
159
+ "eval_runtime": 1.8901,
160
+ "eval_samples_per_second": 839.645,
161
+ "eval_steps_per_second": 26.454,
162
+ "step": 3568
163
+ },
164
+ {
165
+ "epoch": 9.0,
166
+ "grad_norm": 0.03602065145969391,
167
+ "learning_rate": 4.900783338080024e-05,
168
+ "loss": 0.0695,
169
+ "step": 4014
170
+ },
171
+ {
172
+ "epoch": 9.0,
173
+ "eval_accuracy": 0.9010712035286704,
174
+ "eval_f1": 0.8671410632063625,
175
+ "eval_loss": 0.650521457195282,
176
+ "eval_precision": 0.8951755990488386,
177
+ "eval_recall": 0.847398042250407,
178
+ "eval_runtime": 1.8893,
179
+ "eval_samples_per_second": 840.014,
180
+ "eval_steps_per_second": 26.465,
181
+ "step": 4014
182
+ },
183
+ {
184
+ "epoch": 10.0,
185
+ "grad_norm": 0.15067948400974274,
186
+ "learning_rate": 4.877695702189014e-05,
187
+ "loss": 0.0551,
188
+ "step": 4460
189
+ },
190
+ {
191
+ "epoch": 10.0,
192
+ "eval_accuracy": 0.8821676118462508,
193
+ "eval_f1": 0.8511983345308572,
194
+ "eval_loss": 0.53841632604599,
195
+ "eval_precision": 0.8504388791271249,
196
+ "eval_recall": 0.8519688046471428,
197
+ "eval_runtime": 1.8908,
198
+ "eval_samples_per_second": 839.349,
199
+ "eval_steps_per_second": 26.445,
200
+ "step": 4460
201
+ },
202
+ {
203
+ "epoch": 11.0,
204
+ "grad_norm": 0.37308648228645325,
205
+ "learning_rate": 4.852261567702625e-05,
206
+ "loss": 0.0544,
207
+ "step": 4906
208
+ },
209
+ {
210
+ "epoch": 11.0,
211
+ "eval_accuracy": 0.9029615626969124,
212
+ "eval_f1": 0.8757205622114191,
213
+ "eval_loss": 0.5222230553627014,
214
+ "eval_precision": 0.8804347826086957,
215
+ "eval_recall": 0.8713442945870435,
216
+ "eval_runtime": 1.8866,
217
+ "eval_samples_per_second": 841.185,
218
+ "eval_steps_per_second": 26.502,
219
+ "step": 4906
220
+ }
221
+ ],
222
+ "logging_steps": 446,
223
+ "max_steps": 44600,
224
+ "num_input_tokens_seen": 0,
225
+ "num_train_epochs": 100,
226
+ "save_steps": 500,
227
+ "stateful_callbacks": {
228
+ "EarlyStoppingCallback": {
229
+ "args": {
230
+ "early_stopping_patience": 5,
231
+ "early_stopping_threshold": 0.0
232
+ },
233
+ "attributes": {
234
+ "early_stopping_patience_counter": 0
235
+ }
236
+ },
237
+ "TrainerControl": {
238
+ "args": {
239
+ "should_epoch_stop": false,
240
+ "should_evaluate": false,
241
+ "should_log": false,
242
+ "should_save": true,
243
+ "should_training_stop": false
244
+ },
245
+ "attributes": {}
246
+ }
247
+ },
248
+ "total_flos": 0.0,
249
+ "train_batch_size": 32,
250
+ "trial_name": null,
251
+ "trial_params": null
252
+ }