takedarn commited on
Commit
5790c24
·
verified ·
1 Parent(s): e9d921e

Upload folder using huggingface_hub

Browse files
Files changed (4) hide show
  1. config.json +28 -0
  2. eval_results.txt +1899 -0
  3. pytorch_model.bin +3 -0
  4. vocab.txt +0 -0
config.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "BertForSequenceClassification"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.1,
6
+ "cell": {},
7
+ "classifier_dropout": null,
8
+ "dtype": "float32",
9
+ "hidden_act": "gelu",
10
+ "hidden_dropout_prob": 0.1,
11
+ "hidden_size": 768,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 3072,
14
+ "layer_norm_eps": 1e-12,
15
+ "max_position_embeddings": 512,
16
+ "model_type": "bert",
17
+ "num_attention_heads": 12,
18
+ "num_hidden_layers": 6,
19
+ "pad_token_id": 0,
20
+ "position_embedding_type": "absolute",
21
+ "pre_trained": "",
22
+ "structure": [],
23
+ "training": "",
24
+ "transformers_version": "4.57.0",
25
+ "type_vocab_size": 2,
26
+ "use_cache": true,
27
+ "vocab_size": 30522
28
+ }
eval_results.txt ADDED
@@ -0,0 +1,1899 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ acc = 0.8578431372549019
2
+ acc_and_f1 = 0.8802638505066456
3
+ att_loss = 0.0
4
+ cls_loss = 0.3242976481866355
5
+ eval_loss = 0.4957661720422598
6
+ f1 = 0.9026845637583892
7
+ global_step = 99
8
+ loss = 0.3242976481866355
9
+ rep_loss = 0.0
10
+ acc = 0.8700980392156863
11
+ acc_and_f1 = 0.8899042155192571
12
+ att_loss = 0.0
13
+ cls_loss = 0.29263367926954625
14
+ eval_loss = 0.37604412551109606
15
+ f1 = 0.909710391822828
16
+ global_step = 199
17
+ loss = 0.29263367926954625
18
+ rep_loss = 0.0
19
+ acc = 0.875
20
+ acc_and_f1 = 0.893458549222798
21
+ att_loss = 0.0
22
+ cls_loss = 0.2793534170823751
23
+ eval_loss = 0.3502044482873036
24
+ f1 = 0.9119170984455959
25
+ global_step = 299
26
+ loss = 0.2793534170823751
27
+ rep_loss = 0.0
28
+ acc = 0.8480392156862745
29
+ acc_and_f1 = 0.8721801429601941
30
+ att_loss = 0.0
31
+ cls_loss = 0.27228835038373944
32
+ eval_loss = 0.3637411135893602
33
+ f1 = 0.8963210702341137
34
+ global_step = 399
35
+ loss = 0.27228835038373944
36
+ rep_loss = 0.0
37
+ acc = 0.8529411764705882
38
+ acc_and_f1 = 0.8759655377302435
39
+ att_loss = 0.0
40
+ cls_loss = 0.26796382716399636
41
+ eval_loss = 0.3581725668448668
42
+ f1 = 0.898989898989899
43
+ global_step = 499
44
+ loss = 0.26796382716399636
45
+ rep_loss = 0.0
46
+ acc = 0.8578431372549019
47
+ acc_and_f1 = 0.8797690262545697
48
+ att_loss = 0.0
49
+ cls_loss = 0.2651721050275586
50
+ eval_loss = 0.3489295714176618
51
+ f1 = 0.9016949152542373
52
+ global_step = 599
53
+ loss = 0.2651721050275586
54
+ rep_loss = 0.0
55
+ acc = 0.8357843137254902
56
+ acc_and_f1 = 0.8628839466821212
57
+ att_loss = 0.0
58
+ cls_loss = 0.26308564340778345
59
+ eval_loss = 0.36577051877975464
60
+ f1 = 0.8899835796387521
61
+ global_step = 699
62
+ loss = 0.26308564340778345
63
+ rep_loss = 0.0
64
+ acc = 0.8602941176470589
65
+ acc_and_f1 = 0.8819237085697224
66
+ att_loss = 0.0
67
+ cls_loss = 0.26133335688534903
68
+ eval_loss = 0.3564658004503984
69
+ f1 = 0.9035532994923858
70
+ global_step = 799
71
+ loss = 0.26133335688534903
72
+ rep_loss = 0.0
73
+ acc = 0.8676470588235294
74
+ acc_and_f1 = 0.8877484440875326
75
+ att_loss = 0.0
76
+ cls_loss = 0.26052062065901027
77
+ eval_loss = 0.3444621574420195
78
+ f1 = 0.9078498293515358
79
+ global_step = 899
80
+ loss = 0.26052062065901027
81
+ rep_loss = 0.0
82
+ acc = 0.8382352941176471
83
+ acc_and_f1 = 0.8650192864030859
84
+ att_loss = 0.0
85
+ cls_loss = 0.25977209680252245
86
+ eval_loss = 0.37112209773980653
87
+ f1 = 0.8918032786885246
88
+ global_step = 999
89
+ loss = 0.25977209680252245
90
+ rep_loss = 0.0
91
+ acc = 0.8578431372549019
92
+ acc_and_f1 = 0.8801000198059021
93
+ att_loss = 0.0
94
+ cls_loss = 0.2590724386688359
95
+ eval_loss = 0.3469302596954199
96
+ f1 = 0.9023569023569024
97
+ global_step = 1099
98
+ loss = 0.2590724386688359
99
+ rep_loss = 0.0
100
+ acc = 0.8504901960784313
101
+ acc_and_f1 = 0.8743269010442241
102
+ att_loss = 0.0
103
+ cls_loss = 0.25856810821852555
104
+ eval_loss = 0.3481288712758284
105
+ f1 = 0.8981636060100167
106
+ global_step = 1199
107
+ loss = 0.25856810821852555
108
+ rep_loss = 0.0
109
+ acc = 0.875
110
+ acc_and_f1 = 0.8937607204116638
111
+ att_loss = 0.0
112
+ cls_loss = 0.2580898127893561
113
+ eval_loss = 0.333905516908719
114
+ f1 = 0.9125214408233276
115
+ global_step = 1299
116
+ loss = 0.2580898127893561
117
+ rep_loss = 0.0
118
+ acc = 0.8455882352941176
119
+ acc_and_f1 = 0.8705553116769096
120
+ att_loss = 0.0
121
+ cls_loss = 0.257552234284191
122
+ eval_loss = 0.36380689304608566
123
+ f1 = 0.8955223880597015
124
+ global_step = 1399
125
+ loss = 0.257552234284191
126
+ rep_loss = 0.0
127
+ acc = 0.8431372549019608
128
+ acc_and_f1 = 0.8680569217653616
129
+ att_loss = 0.0
130
+ cls_loss = 0.2570895803141705
131
+ eval_loss = 0.35713368539626783
132
+ f1 = 0.8929765886287625
133
+ global_step = 1499
134
+ loss = 0.2570895803141705
135
+ rep_loss = 0.0
136
+ acc = 0.8529411764705882
137
+ acc_and_f1 = 0.8761350177654954
138
+ att_loss = 0.0
139
+ cls_loss = 0.2569112215305284
140
+ eval_loss = 0.3565815102595549
141
+ f1 = 0.8993288590604027
142
+ global_step = 1599
143
+ loss = 0.2569112215305284
144
+ rep_loss = 0.0
145
+ acc = 0.8725490196078431
146
+ acc_and_f1 = 0.891753961858716
147
+ att_loss = 0.0
148
+ cls_loss = 0.2566968538502934
149
+ eval_loss = 0.3385407466154832
150
+ f1 = 0.910958904109589
151
+ global_step = 1699
152
+ loss = 0.2566968538502934
153
+ rep_loss = 0.0
154
+ acc = 0.8627450980392157
155
+ acc_and_f1 = 0.883262583383869
156
+ att_loss = 0.0
157
+ cls_loss = 0.256454997994689
158
+ eval_loss = 0.3509576893769778
159
+ f1 = 0.9037800687285223
160
+ global_step = 1799
161
+ loss = 0.256454997994689
162
+ rep_loss = 0.0
163
+ acc = 0.8578431372549019
164
+ acc_and_f1 = 0.8799350821409644
165
+ att_loss = 0.0
166
+ cls_loss = 0.25624414355645625
167
+ eval_loss = 0.34759315045980305
168
+ f1 = 0.902027027027027
169
+ global_step = 1899
170
+ loss = 0.25624414355645625
171
+ rep_loss = 0.0
172
+ acc = 0.8651960784313726
173
+ acc_and_f1 = 0.8857496576143234
174
+ att_loss = 0.0
175
+ cls_loss = 0.2560090167842071
176
+ eval_loss = 0.347777607349249
177
+ f1 = 0.9063032367972743
178
+ global_step = 1999
179
+ loss = 0.2560090167842071
180
+ rep_loss = 0.0
181
+ acc = 0.8259803921568627
182
+ acc_and_f1 = 0.855266618842659
183
+ att_loss = 0.0
184
+ cls_loss = 0.2559245532540948
185
+ eval_loss = 0.3720426639685264
186
+ f1 = 0.8845528455284553
187
+ global_step = 2099
188
+ loss = 0.2559245532540948
189
+ rep_loss = 0.0
190
+ acc = 0.8382352941176471
191
+ acc_and_f1 = 0.8653717187200614
192
+ att_loss = 0.0
193
+ cls_loss = 0.2558124039852386
194
+ eval_loss = 0.3727398125024942
195
+ f1 = 0.8925081433224755
196
+ global_step = 2199
197
+ loss = 0.2558124039852386
198
+ rep_loss = 0.0
199
+ acc = 0.8553921568627451
200
+ acc_and_f1 = 0.8782823430879889
201
+ att_loss = 0.0
202
+ cls_loss = 0.2555689057120555
203
+ eval_loss = 0.34686333628801197
204
+ f1 = 0.9011725293132329
205
+ global_step = 2299
206
+ loss = 0.2555689057120555
207
+ rep_loss = 0.0
208
+ acc = 0.8578431372549019
209
+ acc_and_f1 = 0.880426585349859
210
+ att_loss = 0.0
211
+ cls_loss = 0.2553385552352744
212
+ eval_loss = 0.3510417170249499
213
+ f1 = 0.903010033444816
214
+ global_step = 2399
215
+ loss = 0.2553385552352744
216
+ rep_loss = 0.0
217
+ acc = 0.8651960784313726
218
+ acc_and_f1 = 0.8863795518207283
219
+ att_loss = 0.0
220
+ cls_loss = 0.25519898026382604
221
+ eval_loss = 0.3500976963685109
222
+ f1 = 0.907563025210084
223
+ global_step = 2499
224
+ loss = 0.25519898026382604
225
+ rep_loss = 0.0
226
+ acc = 0.8725490196078431
227
+ acc_and_f1 = 0.8920568227290917
228
+ att_loss = 0.0
229
+ cls_loss = 0.25513598423776557
230
+ eval_loss = 0.33592245326592374
231
+ f1 = 0.9115646258503401
232
+ global_step = 2599
233
+ loss = 0.25513598423776557
234
+ rep_loss = 0.0
235
+ acc = 0.8799019607843137
236
+ acc_and_f1 = 0.8977823056933616
237
+ att_loss = 0.0
238
+ cls_loss = 0.2549745513194576
239
+ eval_loss = 0.33671250022374666
240
+ f1 = 0.9156626506024096
241
+ global_step = 2699
242
+ loss = 0.2549745513194576
243
+ rep_loss = 0.0
244
+ acc = 0.8455882352941176
245
+ acc_and_f1 = 0.8703814720563766
246
+ att_loss = 0.0
247
+ cls_loss = 0.2548747558270492
248
+ eval_loss = 0.35129279127487767
249
+ f1 = 0.8951747088186356
250
+ global_step = 2799
251
+ loss = 0.2548747558270492
252
+ rep_loss = 0.0
253
+ acc = 0.8700980392156863
254
+ acc_and_f1 = 0.8899042155192571
255
+ att_loss = 0.0
256
+ cls_loss = 0.254759354796357
257
+ eval_loss = 0.3461922097664613
258
+ f1 = 0.909710391822828
259
+ global_step = 2899
260
+ loss = 0.254759354796357
261
+ rep_loss = 0.0
262
+ acc = 0.8455882352941176
263
+ acc_and_f1 = 0.8705553116769096
264
+ att_loss = 0.0
265
+ cls_loss = 0.2546514504182017
266
+ eval_loss = 0.3605812146113469
267
+ f1 = 0.8955223880597015
268
+ global_step = 2999
269
+ loss = 0.2546514504182017
270
+ rep_loss = 0.0
271
+ acc = 0.8529411764705882
272
+ acc_and_f1 = 0.8759655377302435
273
+ att_loss = 0.0
274
+ cls_loss = 0.25449483129296546
275
+ eval_loss = 0.3575789022904176
276
+ f1 = 0.898989898989899
277
+ global_step = 3099
278
+ loss = 0.25449483129296546
279
+ rep_loss = 0.0
280
+ acc = 0.875
281
+ acc_and_f1 = 0.8936101549053357
282
+ att_loss = 0.0
283
+ cls_loss = 0.25449092466315465
284
+ eval_loss = 0.3449848878842134
285
+ f1 = 0.9122203098106713
286
+ global_step = 3199
287
+ loss = 0.25449092466315465
288
+ rep_loss = 0.0
289
+ acc = 0.8455882352941176
290
+ acc_and_f1 = 0.8705553116769096
291
+ att_loss = 0.0
292
+ cls_loss = 0.25450513546901893
293
+ eval_loss = 0.3649230289917726
294
+ f1 = 0.8955223880597015
295
+ global_step = 3299
296
+ loss = 0.25450513546901893
297
+ rep_loss = 0.0
298
+ acc = 0.8578431372549019
299
+ acc_and_f1 = 0.8801000198059021
300
+ att_loss = 0.0
301
+ cls_loss = 0.25436776550281187
302
+ eval_loss = 0.3560798065020488
303
+ f1 = 0.9023569023569024
304
+ global_step = 3399
305
+ loss = 0.25436776550281187
306
+ rep_loss = 0.0
307
+ acc = 0.8382352941176471
308
+ acc_and_f1 = 0.8646622015142691
309
+ att_loss = 0.0
310
+ cls_loss = 0.2543743547455656
311
+ eval_loss = 0.36460480667077577
312
+ f1 = 0.8910891089108911
313
+ global_step = 3499
314
+ loss = 0.2543743547455656
315
+ rep_loss = 0.0
316
+ acc = 0.8676470588235294
317
+ acc_and_f1 = 0.887905162064826
318
+ att_loss = 0.0
319
+ cls_loss = 0.2542944018063727
320
+ eval_loss = 0.3417053394592725
321
+ f1 = 0.9081632653061225
322
+ global_step = 3599
323
+ loss = 0.2542944018063727
324
+ rep_loss = 0.0
325
+ acc = 0.8651960784313726
326
+ acc_and_f1 = 0.8860667363392057
327
+ att_loss = 0.0
328
+ cls_loss = 0.25428640324784535
329
+ eval_loss = 0.3461264268710063
330
+ f1 = 0.9069373942470389
331
+ global_step = 3699
332
+ loss = 0.25428640324784535
333
+ rep_loss = 0.0
334
+ acc = 0.8651960784313726
335
+ acc_and_f1 = 0.8855894922071392
336
+ att_loss = 0.0
337
+ cls_loss = 0.2542373033009507
338
+ eval_loss = 0.3465757690943204
339
+ f1 = 0.905982905982906
340
+ global_step = 3799
341
+ loss = 0.2542373033009507
342
+ rep_loss = 0.0
343
+ acc = 0.8627450980392157
344
+ acc_and_f1 = 0.8837535014005602
345
+ att_loss = 0.0
346
+ cls_loss = 0.2541301912315444
347
+ eval_loss = 0.3509358007174272
348
+ f1 = 0.9047619047619048
349
+ global_step = 3899
350
+ loss = 0.2541301912315444
351
+ rep_loss = 0.0
352
+ acc = 0.8553921568627451
353
+ acc_and_f1 = 0.8782823430879889
354
+ att_loss = 0.0
355
+ cls_loss = 0.25408553400093925
356
+ eval_loss = 0.3561211767104956
357
+ f1 = 0.9011725293132329
358
+ global_step = 3999
359
+ loss = 0.25408553400093925
360
+ rep_loss = 0.0
361
+ acc = 0.8455882352941176
362
+ acc_and_f1 = 0.8700302985515814
363
+ att_loss = 0.0
364
+ cls_loss = 0.25409953523391804
365
+ eval_loss = 0.3504564188993894
366
+ f1 = 0.8944723618090452
367
+ global_step = 4099
368
+ loss = 0.25409953523391804
369
+ rep_loss = 0.0
370
+ acc = 0.8578431372549019
371
+ acc_and_f1 = 0.8802638505066456
372
+ att_loss = 0.0
373
+ cls_loss = 0.2540690626671383
374
+ eval_loss = 0.3497544423891948
375
+ f1 = 0.9026845637583892
376
+ global_step = 4199
377
+ loss = 0.2540690626671383
378
+ rep_loss = 0.0
379
+ acc = 0.8553921568627451
380
+ acc_and_f1 = 0.8779490295274939
381
+ att_loss = 0.0
382
+ cls_loss = 0.25401594868840993
383
+ eval_loss = 0.35735770716116977
384
+ f1 = 0.9005059021922428
385
+ global_step = 4299
386
+ loss = 0.25401594868840993
387
+ rep_loss = 0.0
388
+ acc = 0.8602941176470589
389
+ acc_and_f1 = 0.8817599620493359
390
+ att_loss = 0.0
391
+ cls_loss = 0.2539481226931542
392
+ eval_loss = 0.3534167305781291
393
+ f1 = 0.9032258064516129
394
+ global_step = 4399
395
+ loss = 0.2539481226931542
396
+ rep_loss = 0.0
397
+ acc = 0.8455882352941176
398
+ acc_and_f1 = 0.8703814720563766
399
+ att_loss = 0.0
400
+ cls_loss = 0.25384427584205954
401
+ eval_loss = 0.35798397889504063
402
+ f1 = 0.8951747088186356
403
+ global_step = 4499
404
+ loss = 0.25384427584205954
405
+ rep_loss = 0.0
406
+ acc = 0.8676470588235294
407
+ acc_and_f1 = 0.8880608175473579
408
+ att_loss = 0.0
409
+ cls_loss = 0.2538821630618914
410
+ eval_loss = 0.3465126913327437
411
+ f1 = 0.9084745762711864
412
+ global_step = 4599
413
+ loss = 0.2538821630618914
414
+ rep_loss = 0.0
415
+ acc = 0.8700980392156863
416
+ acc_and_f1 = 0.8897498743086978
417
+ att_loss = 0.0
418
+ cls_loss = 0.2538268513363913
419
+ eval_loss = 0.34550271125940174
420
+ f1 = 0.9094017094017094
421
+ global_step = 4699
422
+ loss = 0.2538268513363913
423
+ rep_loss = 0.0
424
+ acc = 0.8627450980392157
425
+ acc_and_f1 = 0.8842345018815607
426
+ att_loss = 0.0
427
+ cls_loss = 0.25377951471737015
428
+ eval_loss = 0.3496234290874921
429
+ f1 = 0.9057239057239057
430
+ global_step = 4799
431
+ loss = 0.25377951471737015
432
+ rep_loss = 0.0
433
+ acc = 0.8504901960784313
434
+ acc_and_f1 = 0.8738117084945276
435
+ att_loss = 0.0
436
+ cls_loss = 0.253692247221411
437
+ eval_loss = 0.3484192524964993
438
+ f1 = 0.897133220910624
439
+ global_step = 4899
440
+ loss = 0.253692247221411
441
+ rep_loss = 0.0
442
+ acc = 0.8627450980392157
443
+ acc_and_f1 = 0.8840752517223105
444
+ att_loss = 0.0
445
+ cls_loss = 0.25362403400482286
446
+ eval_loss = 0.34760494988698226
447
+ f1 = 0.9054054054054054
448
+ global_step = 4999
449
+ loss = 0.25362403400482286
450
+ rep_loss = 0.0
451
+ acc = 0.8578431372549019
452
+ acc_and_f1 = 0.880426585349859
453
+ att_loss = 0.0
454
+ cls_loss = 0.25358679795363387
455
+ eval_loss = 0.35553938952776104
456
+ f1 = 0.903010033444816
457
+ global_step = 5099
458
+ loss = 0.25358679795363387
459
+ rep_loss = 0.0
460
+ acc = 0.8504901960784313
461
+ acc_and_f1 = 0.8744963459593488
462
+ att_loss = 0.0
463
+ cls_loss = 0.25357687659310846
464
+ eval_loss = 0.3577815191103862
465
+ f1 = 0.8985024958402662
466
+ global_step = 5199
467
+ loss = 0.25357687659310846
468
+ rep_loss = 0.0
469
+ acc = 0.8602941176470589
470
+ acc_and_f1 = 0.8819237085697224
471
+ att_loss = 0.0
472
+ cls_loss = 0.2535540806918712
473
+ eval_loss = 0.34547018660948825
474
+ f1 = 0.9035532994923858
475
+ global_step = 5299
476
+ loss = 0.2535540806918712
477
+ rep_loss = 0.0
478
+ acc = 0.8406862745098039
479
+ acc_and_f1 = 0.8666241289904392
480
+ att_loss = 0.0
481
+ cls_loss = 0.2535351322031304
482
+ eval_loss = 0.3646775048512679
483
+ f1 = 0.8925619834710744
484
+ global_step = 5399
485
+ loss = 0.2535351322031304
486
+ rep_loss = 0.0
487
+ acc = 0.8651960784313726
488
+ acc_and_f1 = 0.8851023570049782
489
+ att_loss = 0.0
490
+ cls_loss = 0.2534648269620846
491
+ eval_loss = 0.33971722653278935
492
+ f1 = 0.9050086355785838
493
+ global_step = 5499
494
+ loss = 0.2534648269620846
495
+ rep_loss = 0.0
496
+ acc = 0.8382352941176471
497
+ acc_and_f1 = 0.8646622015142691
498
+ att_loss = 0.0
499
+ cls_loss = 0.25345307687048957
500
+ eval_loss = 0.3617634429381444
501
+ f1 = 0.8910891089108911
502
+ global_step = 5599
503
+ loss = 0.25345307687048957
504
+ rep_loss = 0.0
505
+ acc = 0.8382352941176471
506
+ acc_and_f1 = 0.8646622015142691
507
+ att_loss = 0.0
508
+ cls_loss = 0.2534148392865147
509
+ eval_loss = 0.3660709330668816
510
+ f1 = 0.8910891089108911
511
+ global_step = 5699
512
+ loss = 0.2534148392865147
513
+ rep_loss = 0.0
514
+ acc = 0.8382352941176471
515
+ acc_and_f1 = 0.8644818854694196
516
+ att_loss = 0.0
517
+ cls_loss = 0.2533445278618496
518
+ eval_loss = 0.36703427135944366
519
+ f1 = 0.890728476821192
520
+ global_step = 5799
521
+ loss = 0.2533445278618496
522
+ rep_loss = 0.0
523
+ acc = 0.8455882352941176
524
+ acc_and_f1 = 0.8703814720563766
525
+ att_loss = 0.0
526
+ cls_loss = 0.2533070436497425
527
+ eval_loss = 0.3633102167111177
528
+ f1 = 0.8951747088186356
529
+ global_step = 5899
530
+ loss = 0.2533070436497425
531
+ rep_loss = 0.0
532
+ acc = 0.8627450980392157
533
+ acc_and_f1 = 0.8837535014005602
534
+ att_loss = 0.0
535
+ cls_loss = 0.2532717240618793
536
+ eval_loss = 0.34351416849173033
537
+ f1 = 0.9047619047619048
538
+ global_step = 5999
539
+ loss = 0.2532717240618793
540
+ rep_loss = 0.0
541
+ acc = 0.8700980392156863
542
+ acc_and_f1 = 0.8897498743086978
543
+ att_loss = 0.0
544
+ cls_loss = 0.25324679027914826
545
+ eval_loss = 0.3388754427433014
546
+ f1 = 0.9094017094017094
547
+ global_step = 6099
548
+ loss = 0.25324679027914826
549
+ rep_loss = 0.0
550
+ acc = 0.8651960784313726
551
+ acc_and_f1 = 0.8859087353107626
552
+ att_loss = 0.0
553
+ cls_loss = 0.2531998183245927
554
+ eval_loss = 0.3493821873114659
555
+ f1 = 0.9066213921901528
556
+ global_step = 6199
557
+ loss = 0.2531998183245927
558
+ rep_loss = 0.0
559
+ acc = 0.8553921568627451
560
+ acc_and_f1 = 0.8779490295274939
561
+ att_loss = 0.0
562
+ cls_loss = 0.25318595228128576
563
+ eval_loss = 0.34524847108584183
564
+ f1 = 0.9005059021922428
565
+ global_step = 6299
566
+ loss = 0.25318595228128576
567
+ rep_loss = 0.0
568
+ acc = 0.8602941176470589
569
+ acc_and_f1 = 0.8817599620493359
570
+ att_loss = 0.0
571
+ cls_loss = 0.25316377091703984
572
+ eval_loss = 0.34949486416119796
573
+ f1 = 0.9032258064516129
574
+ global_step = 6399
575
+ loss = 0.25316377091703984
576
+ rep_loss = 0.0
577
+ acc = 0.8651960784313726
578
+ acc_and_f1 = 0.8859087353107626
579
+ att_loss = 0.0
580
+ cls_loss = 0.2531305441633154
581
+ eval_loss = 0.3458871142222331
582
+ f1 = 0.9066213921901528
583
+ global_step = 6499
584
+ loss = 0.2531305441633154
585
+ rep_loss = 0.0
586
+ acc = 0.8700980392156863
587
+ acc_and_f1 = 0.8902097641086892
588
+ att_loss = 0.0
589
+ cls_loss = 0.2531191714899452
590
+ eval_loss = 0.3459375523603879
591
+ f1 = 0.9103214890016921
592
+ global_step = 6599
593
+ loss = 0.2531191714899452
594
+ rep_loss = 0.0
595
+ acc = 0.8700980392156863
596
+ acc_and_f1 = 0.8900575085721896
597
+ att_loss = 0.0
598
+ cls_loss = 0.2530899527985261
599
+ eval_loss = 0.34675515500398785
600
+ f1 = 0.9100169779286927
601
+ global_step = 6699
602
+ loss = 0.2530899527985261
603
+ rep_loss = 0.0
604
+ acc = 0.8651960784313726
605
+ acc_and_f1 = 0.8857496576143234
606
+ att_loss = 0.0
607
+ cls_loss = 0.25311452018844677
608
+ eval_loss = 0.34479153614777786
609
+ f1 = 0.9063032367972743
610
+ global_step = 6799
611
+ loss = 0.25311452018844677
612
+ rep_loss = 0.0
613
+ acc = 0.8676470588235294
614
+ acc_and_f1 = 0.8877484440875326
615
+ att_loss = 0.0
616
+ cls_loss = 0.2530607040860823
617
+ eval_loss = 0.3488370054043256
618
+ f1 = 0.9078498293515358
619
+ global_step = 6899
620
+ loss = 0.2530607040860823
621
+ rep_loss = 0.0
622
+ acc = 0.8578431372549019
623
+ acc_and_f1 = 0.8799350821409644
624
+ att_loss = 0.0
625
+ cls_loss = 0.2530439395592168
626
+ eval_loss = 0.3482685432984279
627
+ f1 = 0.902027027027027
628
+ global_step = 6999
629
+ loss = 0.2530439395592168
630
+ rep_loss = 0.0
631
+ acc = 0.8578431372549019
632
+ acc_and_f1 = 0.880426585349859
633
+ att_loss = 0.0
634
+ cls_loss = 0.25111380563332486
635
+ eval_loss = 0.3547634929418564
636
+ f1 = 0.903010033444816
637
+ global_step = 7099
638
+ loss = 0.25111380563332486
639
+ rep_loss = 0.0
640
+ acc = 0.8431372549019608
641
+ acc_and_f1 = 0.868235294117647
642
+ att_loss = 0.0
643
+ cls_loss = 0.25059721966584525
644
+ eval_loss = 0.36289874750834245
645
+ f1 = 0.8933333333333333
646
+ global_step = 7199
647
+ loss = 0.25059721966584525
648
+ rep_loss = 0.0
649
+ acc = 0.8455882352941176
650
+ acc_and_f1 = 0.8705553116769096
651
+ att_loss = 0.0
652
+ cls_loss = 0.2501519975234877
653
+ eval_loss = 0.35943046097572035
654
+ f1 = 0.8955223880597015
655
+ global_step = 7299
656
+ loss = 0.2501519975234877
657
+ rep_loss = 0.0
658
+ acc = 0.8602941176470589
659
+ acc_and_f1 = 0.8817599620493359
660
+ att_loss = 0.0
661
+ cls_loss = 0.25039135756557934
662
+ eval_loss = 0.3471243713910763
663
+ f1 = 0.9032258064516129
664
+ global_step = 7399
665
+ loss = 0.25039135756557934
666
+ rep_loss = 0.0
667
+ acc = 0.8602941176470589
668
+ acc_and_f1 = 0.8822478991596638
669
+ att_loss = 0.0
670
+ cls_loss = 0.25091522460983645
671
+ eval_loss = 0.34930069056841045
672
+ f1 = 0.9042016806722689
673
+ global_step = 7499
674
+ loss = 0.25091522460983645
675
+ rep_loss = 0.0
676
+ acc = 0.8627450980392157
677
+ acc_and_f1 = 0.8839149219009639
678
+ att_loss = 0.0
679
+ cls_loss = 0.2509644917154734
680
+ eval_loss = 0.34623642838918245
681
+ f1 = 0.9050847457627119
682
+ global_step = 7599
683
+ loss = 0.2509644917154734
684
+ rep_loss = 0.0
685
+ acc = 0.8602941176470589
686
+ acc_and_f1 = 0.8820863505604604
687
+ att_loss = 0.0
688
+ cls_loss = 0.2509478765770905
689
+ eval_loss = 0.3377785464892021
690
+ f1 = 0.9038785834738617
691
+ global_step = 7699
692
+ loss = 0.2509478765770905
693
+ rep_loss = 0.0
694
+ acc = 0.8602941176470589
695
+ acc_and_f1 = 0.8822478991596638
696
+ att_loss = 0.0
697
+ cls_loss = 0.25074089159762936
698
+ eval_loss = 0.35502059986958134
699
+ f1 = 0.9042016806722689
700
+ global_step = 7799
701
+ loss = 0.25074089159762936
702
+ rep_loss = 0.0
703
+ acc = 0.8602941176470589
704
+ acc_and_f1 = 0.8822478991596638
705
+ att_loss = 0.0
706
+ cls_loss = 0.25068820375583073
707
+ eval_loss = 0.3541043515388782
708
+ f1 = 0.9042016806722689
709
+ global_step = 7899
710
+ loss = 0.25068820375583073
711
+ rep_loss = 0.0
712
+ acc = 0.8651960784313726
713
+ acc_and_f1 = 0.8865343876243965
714
+ att_loss = 0.0
715
+ cls_loss = 0.25077821322055677
716
+ eval_loss = 0.3519786733847398
717
+ f1 = 0.9078726968174204
718
+ global_step = 7999
719
+ loss = 0.25077821322055677
720
+ rep_loss = 0.0
721
+ acc = 0.8627450980392157
722
+ acc_and_f1 = 0.8840752517223105
723
+ att_loss = 0.0
724
+ cls_loss = 0.2508025012945345
725
+ eval_loss = 0.3526520866614122
726
+ f1 = 0.9054054054054054
727
+ global_step = 8099
728
+ loss = 0.2508025012945345
729
+ rep_loss = 0.0
730
+ acc = 0.8529411764705882
731
+ acc_and_f1 = 0.8763033641550265
732
+ att_loss = 0.0
733
+ cls_loss = 0.25104583394373947
734
+ eval_loss = 0.36234907462046695
735
+ f1 = 0.8996655518394648
736
+ global_step = 8199
737
+ loss = 0.25104583394373947
738
+ rep_loss = 0.0
739
+ acc = 0.8578431372549019
740
+ acc_and_f1 = 0.880426585349859
741
+ att_loss = 0.0
742
+ cls_loss = 0.2509774452259418
743
+ eval_loss = 0.3583677720565062
744
+ f1 = 0.903010033444816
745
+ global_step = 8299
746
+ loss = 0.2509774452259418
747
+ rep_loss = 0.0
748
+ acc = 0.8700980392156863
749
+ acc_and_f1 = 0.8900575085721896
750
+ att_loss = 0.0
751
+ cls_loss = 0.250863101936522
752
+ eval_loss = 0.35542446260268873
753
+ f1 = 0.9100169779286927
754
+ global_step = 8399
755
+ loss = 0.250863101936522
756
+ rep_loss = 0.0
757
+ acc = 0.8602941176470589
758
+ acc_and_f1 = 0.8822478991596638
759
+ att_loss = 0.0
760
+ cls_loss = 0.2508395428136754
761
+ eval_loss = 0.35508807691243977
762
+ f1 = 0.9042016806722689
763
+ global_step = 8499
764
+ loss = 0.2508395428136754
765
+ rep_loss = 0.0
766
+ acc = 0.8602941176470589
767
+ acc_and_f1 = 0.8820863505604604
768
+ att_loss = 0.0
769
+ cls_loss = 0.25077458718142953
770
+ eval_loss = 0.3556858759659987
771
+ f1 = 0.9038785834738617
772
+ global_step = 8599
773
+ loss = 0.25077458718142953
774
+ rep_loss = 0.0
775
+ acc = 0.8480392156862745
776
+ acc_and_f1 = 0.8721801429601941
777
+ att_loss = 0.0
778
+ cls_loss = 0.250895513759719
779
+ eval_loss = 0.35307549513303316
780
+ f1 = 0.8963210702341137
781
+ global_step = 8699
782
+ loss = 0.250895513759719
783
+ rep_loss = 0.0
784
+ acc = 0.8602941176470589
785
+ acc_and_f1 = 0.8822478991596638
786
+ att_loss = 0.0
787
+ cls_loss = 0.25087003551697934
788
+ eval_loss = 0.3517612104232495
789
+ f1 = 0.9042016806722689
790
+ global_step = 8799
791
+ loss = 0.25087003551697934
792
+ rep_loss = 0.0
793
+ acc = 0.8700980392156863
794
+ acc_and_f1 = 0.8899042155192571
795
+ att_loss = 0.0
796
+ cls_loss = 0.2508806363427287
797
+ eval_loss = 0.3515888601541519
798
+ f1 = 0.909710391822828
799
+ global_step = 8899
800
+ loss = 0.2508806363427287
801
+ rep_loss = 0.0
802
+ acc = 0.875
803
+ acc_and_f1 = 0.8937607204116638
804
+ att_loss = 0.0
805
+ cls_loss = 0.25089499888225975
806
+ eval_loss = 0.34325943084863514
807
+ f1 = 0.9125214408233276
808
+ global_step = 8999
809
+ loss = 0.25089499888225975
810
+ rep_loss = 0.0
811
+ acc = 0.8578431372549019
812
+ acc_and_f1 = 0.8799350821409644
813
+ att_loss = 0.0
814
+ cls_loss = 0.250888285849054
815
+ eval_loss = 0.3543619845922177
816
+ f1 = 0.902027027027027
817
+ global_step = 9099
818
+ loss = 0.250888285849054
819
+ rep_loss = 0.0
820
+ acc = 0.8553921568627451
821
+ acc_and_f1 = 0.8779490295274939
822
+ att_loss = 0.0
823
+ cls_loss = 0.2509580318495528
824
+ eval_loss = 0.35584669273633224
825
+ f1 = 0.9005059021922428
826
+ global_step = 9199
827
+ loss = 0.2509580318495528
828
+ rep_loss = 0.0
829
+ acc = 0.8676470588235294
830
+ acc_and_f1 = 0.8877484440875326
831
+ att_loss = 0.0
832
+ cls_loss = 0.2507857592558492
833
+ eval_loss = 0.3540742339996191
834
+ f1 = 0.9078498293515358
835
+ global_step = 9299
836
+ loss = 0.2507857592558492
837
+ rep_loss = 0.0
838
+ acc = 0.8529411764705882
839
+ acc_and_f1 = 0.8764705882352941
840
+ att_loss = 0.0
841
+ cls_loss = 0.2507655027174798
842
+ eval_loss = 0.35924350871489596
843
+ f1 = 0.9
844
+ global_step = 9399
845
+ loss = 0.2507655027174798
846
+ rep_loss = 0.0
847
+ acc = 0.8676470588235294
848
+ acc_and_f1 = 0.887905162064826
849
+ att_loss = 0.0
850
+ cls_loss = 0.2507434476219858
851
+ eval_loss = 0.3500103514928084
852
+ f1 = 0.9081632653061225
853
+ global_step = 9499
854
+ loss = 0.2507434476219858
855
+ rep_loss = 0.0
856
+ acc = 0.8602941176470589
857
+ acc_and_f1 = 0.8819237085697224
858
+ att_loss = 0.0
859
+ cls_loss = 0.25066164429889554
860
+ eval_loss = 0.352135290320103
861
+ f1 = 0.9035532994923858
862
+ global_step = 9599
863
+ loss = 0.25066164429889554
864
+ rep_loss = 0.0
865
+ acc = 0.8504901960784313
866
+ acc_and_f1 = 0.8743269010442241
867
+ att_loss = 0.0
868
+ cls_loss = 0.2507221189996762
869
+ eval_loss = 0.3587732601624269
870
+ f1 = 0.8981636060100167
871
+ global_step = 9699
872
+ loss = 0.2507221189996762
873
+ rep_loss = 0.0
874
+ acc = 0.8406862745098039
875
+ acc_and_f1 = 0.8669769631990727
876
+ att_loss = 0.0
877
+ cls_loss = 0.25068050444018347
878
+ eval_loss = 0.3656560835930017
879
+ f1 = 0.8932676518883416
880
+ global_step = 9799
881
+ loss = 0.25068050444018347
882
+ rep_loss = 0.0
883
+ acc = 0.8651960784313726
884
+ acc_and_f1 = 0.8857496576143234
885
+ att_loss = 0.0
886
+ cls_loss = 0.2507135918939301
887
+ eval_loss = 0.3497274162677618
888
+ f1 = 0.9063032367972743
889
+ global_step = 9899
890
+ loss = 0.2507135918939301
891
+ rep_loss = 0.0
892
+ acc = 0.8578431372549019
893
+ acc_and_f1 = 0.8801000198059021
894
+ att_loss = 0.0
895
+ cls_loss = 0.2506162076535153
896
+ eval_loss = 0.3515304716733786
897
+ f1 = 0.9023569023569024
898
+ global_step = 9999
899
+ loss = 0.2506162076535153
900
+ rep_loss = 0.0
901
+ acc = 0.8651960784313726
902
+ acc_and_f1 = 0.8859087353107626
903
+ att_loss = 0.0
904
+ cls_loss = 0.25055873150440455
905
+ eval_loss = 0.3466983471925442
906
+ f1 = 0.9066213921901528
907
+ global_step = 10099
908
+ loss = 0.25055873150440455
909
+ rep_loss = 0.0
910
+ acc = 0.8602941176470589
911
+ acc_and_f1 = 0.8817599620493359
912
+ att_loss = 0.0
913
+ cls_loss = 0.2505836462145921
914
+ eval_loss = 0.35016662111649144
915
+ f1 = 0.9032258064516129
916
+ global_step = 10199
917
+ loss = 0.2505836462145921
918
+ rep_loss = 0.0
919
+ acc = 0.8504901960784313
920
+ acc_and_f1 = 0.8743269010442241
921
+ att_loss = 0.0
922
+ cls_loss = 0.25066264783350284
923
+ eval_loss = 0.35700984528431523
924
+ f1 = 0.8981636060100167
925
+ global_step = 10299
926
+ loss = 0.25066264783350284
927
+ rep_loss = 0.0
928
+ acc = 0.8480392156862745
929
+ acc_and_f1 = 0.8723529411764706
930
+ att_loss = 0.0
931
+ cls_loss = 0.250681378746635
932
+ eval_loss = 0.3612692677057706
933
+ f1 = 0.8966666666666666
934
+ global_step = 10399
935
+ loss = 0.250681378746635
936
+ rep_loss = 0.0
937
+ acc = 0.8602941176470589
938
+ acc_and_f1 = 0.8822478991596638
939
+ att_loss = 0.0
940
+ cls_loss = 0.25071217013901964
941
+ eval_loss = 0.3546430892669238
942
+ f1 = 0.9042016806722689
943
+ global_step = 10499
944
+ loss = 0.25071217013901964
945
+ rep_loss = 0.0
946
+ acc = 0.8602941176470589
947
+ acc_and_f1 = 0.8824083653561927
948
+ att_loss = 0.0
949
+ cls_loss = 0.25077468947223996
950
+ eval_loss = 0.35997511790348935
951
+ f1 = 0.9045226130653267
952
+ global_step = 10599
953
+ loss = 0.25077468947223996
954
+ rep_loss = 0.0
955
+ acc = 0.8553921568627451
956
+ acc_and_f1 = 0.8779490295274939
957
+ att_loss = 0.0
958
+ cls_loss = 0.25072247814107557
959
+ eval_loss = 0.34856143364539516
960
+ f1 = 0.9005059021922428
961
+ global_step = 10699
962
+ loss = 0.25072247814107557
963
+ rep_loss = 0.0
964
+ acc = 0.8700980392156863
965
+ acc_and_f1 = 0.8902097641086892
966
+ att_loss = 0.0
967
+ cls_loss = 0.2507110400701741
968
+ eval_loss = 0.35061138753707594
969
+ f1 = 0.9103214890016921
970
+ global_step = 10799
971
+ loss = 0.2507110400701741
972
+ rep_loss = 0.0
973
+ acc = 0.8578431372549019
974
+ acc_and_f1 = 0.8802638505066456
975
+ att_loss = 0.0
976
+ cls_loss = 0.2506920118188488
977
+ eval_loss = 0.35475401007212126
978
+ f1 = 0.9026845637583892
979
+ global_step = 10899
980
+ loss = 0.2506920118188488
981
+ rep_loss = 0.0
982
+ acc = 0.8627450980392157
983
+ acc_and_f1 = 0.8840752517223105
984
+ att_loss = 0.0
985
+ cls_loss = 0.2507689426179792
986
+ eval_loss = 0.34655993489118725
987
+ f1 = 0.9054054054054054
988
+ global_step = 10999
989
+ loss = 0.2507689426179792
990
+ rep_loss = 0.0
991
+ acc = 0.8529411764705882
992
+ acc_and_f1 = 0.8764705882352941
993
+ att_loss = 0.0
994
+ cls_loss = 0.2507825035084071
995
+ eval_loss = 0.3536278639848416
996
+ f1 = 0.9
997
+ global_step = 11099
998
+ loss = 0.2507825035084071
999
+ rep_loss = 0.0
1000
+ acc = 0.8651960784313726
1001
+ acc_and_f1 = 0.8859087353107626
1002
+ att_loss = 0.0
1003
+ cls_loss = 0.25069774150275953
1004
+ eval_loss = 0.3487686056357164
1005
+ f1 = 0.9066213921901528
1006
+ global_step = 11199
1007
+ loss = 0.25069774150275953
1008
+ rep_loss = 0.0
1009
+ acc = 0.8651960784313726
1010
+ acc_and_f1 = 0.8862236715934266
1011
+ att_loss = 0.0
1012
+ cls_loss = 0.2506819853958743
1013
+ eval_loss = 0.3487859356861848
1014
+ f1 = 0.9072512647554806
1015
+ global_step = 11299
1016
+ loss = 0.2506819853958743
1017
+ rep_loss = 0.0
1018
+ acc = 0.8627450980392157
1019
+ acc_and_f1 = 0.8839149219009639
1020
+ att_loss = 0.0
1021
+ cls_loss = 0.2506861321056285
1022
+ eval_loss = 0.34572866788277257
1023
+ f1 = 0.9050847457627119
1024
+ global_step = 11399
1025
+ loss = 0.2506861321056285
1026
+ rep_loss = 0.0
1027
+ acc = 0.8676470588235294
1028
+ acc_and_f1 = 0.8880608175473579
1029
+ att_loss = 0.0
1030
+ cls_loss = 0.2506145616127689
1031
+ eval_loss = 0.34923419470970446
1032
+ f1 = 0.9084745762711864
1033
+ global_step = 11499
1034
+ loss = 0.2506145616127689
1035
+ rep_loss = 0.0
1036
+ acc = 0.8553921568627451
1037
+ acc_and_f1 = 0.8781162464985994
1038
+ att_loss = 0.0
1039
+ cls_loss = 0.2506683378660927
1040
+ eval_loss = 0.35191299365117
1041
+ f1 = 0.9008403361344538
1042
+ global_step = 11599
1043
+ loss = 0.2506683378660927
1044
+ rep_loss = 0.0
1045
+ acc = 0.8578431372549019
1046
+ acc_and_f1 = 0.8802638505066456
1047
+ att_loss = 0.0
1048
+ cls_loss = 0.25072587686975156
1049
+ eval_loss = 0.35413290560245514
1050
+ f1 = 0.9026845637583892
1051
+ global_step = 11699
1052
+ loss = 0.25072587686975156
1053
+ rep_loss = 0.0
1054
+ acc = 0.8627450980392157
1055
+ acc_and_f1 = 0.8840752517223105
1056
+ att_loss = 0.0
1057
+ cls_loss = 0.2507821178698965
1058
+ eval_loss = 0.3447670741723134
1059
+ f1 = 0.9054054054054054
1060
+ global_step = 11799
1061
+ loss = 0.2507821178698965
1062
+ rep_loss = 0.0
1063
+ acc = 0.8627450980392157
1064
+ acc_and_f1 = 0.8840752517223105
1065
+ att_loss = 0.0
1066
+ cls_loss = 0.2507828957252855
1067
+ eval_loss = 0.35020356338757735
1068
+ f1 = 0.9054054054054054
1069
+ global_step = 11899
1070
+ loss = 0.2507828957252855
1071
+ rep_loss = 0.0
1072
+ acc = 0.8578431372549019
1073
+ acc_and_f1 = 0.8797690262545697
1074
+ att_loss = 0.0
1075
+ cls_loss = 0.25077833824107293
1076
+ eval_loss = 0.35079979323423827
1077
+ f1 = 0.9016949152542373
1078
+ global_step = 11999
1079
+ loss = 0.25077833824107293
1080
+ rep_loss = 0.0
1081
+ acc = 0.8627450980392157
1082
+ acc_and_f1 = 0.8840752517223105
1083
+ att_loss = 0.0
1084
+ cls_loss = 0.25074108129427913
1085
+ eval_loss = 0.3491762796273598
1086
+ f1 = 0.9054054054054054
1087
+ global_step = 12099
1088
+ loss = 0.25074108129427913
1089
+ rep_loss = 0.0
1090
+ acc = 0.8578431372549019
1091
+ acc_and_f1 = 0.8797690262545697
1092
+ att_loss = 0.0
1093
+ cls_loss = 0.25067708402843236
1094
+ eval_loss = 0.35040095563118273
1095
+ f1 = 0.9016949152542373
1096
+ global_step = 12199
1097
+ loss = 0.25067708402843236
1098
+ rep_loss = 0.0
1099
+ acc = 0.8578431372549019
1100
+ acc_and_f1 = 0.880426585349859
1101
+ att_loss = 0.0
1102
+ cls_loss = 0.2507138960240347
1103
+ eval_loss = 0.36072335220300233
1104
+ f1 = 0.903010033444816
1105
+ global_step = 12299
1106
+ loss = 0.2507138960240347
1107
+ rep_loss = 0.0
1108
+ acc = 0.8602941176470589
1109
+ acc_and_f1 = 0.8820863505604604
1110
+ att_loss = 0.0
1111
+ cls_loss = 0.250679742348383
1112
+ eval_loss = 0.3501311552066069
1113
+ f1 = 0.9038785834738617
1114
+ global_step = 12399
1115
+ loss = 0.250679742348383
1116
+ rep_loss = 0.0
1117
+ acc = 0.8676470588235294
1118
+ acc_and_f1 = 0.887905162064826
1119
+ att_loss = 0.0
1120
+ cls_loss = 0.25065393314269957
1121
+ eval_loss = 0.3466459409548686
1122
+ f1 = 0.9081632653061225
1123
+ global_step = 12499
1124
+ loss = 0.25065393314269957
1125
+ rep_loss = 0.0
1126
+ acc = 0.8700980392156863
1127
+ acc_and_f1 = 0.8902097641086892
1128
+ att_loss = 0.0
1129
+ cls_loss = 0.2506209577270595
1130
+ eval_loss = 0.3507487281010701
1131
+ f1 = 0.9103214890016921
1132
+ global_step = 12599
1133
+ loss = 0.2506209577270595
1134
+ rep_loss = 0.0
1135
+ acc = 0.8529411764705882
1136
+ acc_and_f1 = 0.8763033641550265
1137
+ att_loss = 0.0
1138
+ cls_loss = 0.25056982488623786
1139
+ eval_loss = 0.35089847101615024
1140
+ f1 = 0.8996655518394648
1141
+ global_step = 12699
1142
+ loss = 0.25056982488623786
1143
+ rep_loss = 0.0
1144
+ acc = 0.8627450980392157
1145
+ acc_and_f1 = 0.8835909790537375
1146
+ att_loss = 0.0
1147
+ cls_loss = 0.2505782142501237
1148
+ eval_loss = 0.3467470705509186
1149
+ f1 = 0.9044368600682594
1150
+ global_step = 12799
1151
+ loss = 0.2505782142501237
1152
+ rep_loss = 0.0
1153
+ acc = 0.8602941176470589
1154
+ acc_and_f1 = 0.8820863505604604
1155
+ att_loss = 0.0
1156
+ cls_loss = 0.2505564182641053
1157
+ eval_loss = 0.3530815484432074
1158
+ f1 = 0.9038785834738617
1159
+ global_step = 12899
1160
+ loss = 0.2505564182641053
1161
+ rep_loss = 0.0
1162
+ acc = 0.8602941176470589
1163
+ acc_and_f1 = 0.8820863505604604
1164
+ att_loss = 0.0
1165
+ cls_loss = 0.25060850552221553
1166
+ eval_loss = 0.3507145127424827
1167
+ f1 = 0.9038785834738617
1168
+ global_step = 12999
1169
+ loss = 0.25060850552221553
1170
+ rep_loss = 0.0
1171
+ acc = 0.8627450980392157
1172
+ acc_and_f1 = 0.8842345018815607
1173
+ att_loss = 0.0
1174
+ cls_loss = 0.25058785082138185
1175
+ eval_loss = 0.34966851541629207
1176
+ f1 = 0.9057239057239057
1177
+ global_step = 13099
1178
+ loss = 0.25058785082138185
1179
+ rep_loss = 0.0
1180
+ acc = 0.8700980392156863
1181
+ acc_and_f1 = 0.8897498743086978
1182
+ att_loss = 0.0
1183
+ cls_loss = 0.25056991792024874
1184
+ eval_loss = 0.34772413166669697
1185
+ f1 = 0.9094017094017094
1186
+ global_step = 13199
1187
+ loss = 0.25056991792024874
1188
+ rep_loss = 0.0
1189
+ acc = 0.8602941176470589
1190
+ acc_and_f1 = 0.8819237085697224
1191
+ att_loss = 0.0
1192
+ cls_loss = 0.25056508799980665
1193
+ eval_loss = 0.35361165610643536
1194
+ f1 = 0.9035532994923858
1195
+ global_step = 13299
1196
+ loss = 0.25056508799980665
1197
+ rep_loss = 0.0
1198
+ acc = 0.8700980392156863
1199
+ acc_and_f1 = 0.8897498743086978
1200
+ att_loss = 0.0
1201
+ cls_loss = 0.2505593915905244
1202
+ eval_loss = 0.3432594881607936
1203
+ f1 = 0.9094017094017094
1204
+ global_step = 13399
1205
+ loss = 0.2505593915905244
1206
+ rep_loss = 0.0
1207
+ acc = 0.8700980392156863
1208
+ acc_and_f1 = 0.8903609926263929
1209
+ att_loss = 0.0
1210
+ cls_loss = 0.25059567983770037
1211
+ eval_loss = 0.35125912496676814
1212
+ f1 = 0.9106239460370995
1213
+ global_step = 13499
1214
+ loss = 0.25059567983770037
1215
+ rep_loss = 0.0
1216
+ acc = 0.8602941176470589
1217
+ acc_and_f1 = 0.8819237085697224
1218
+ att_loss = 0.0
1219
+ cls_loss = 0.25064265978581474
1220
+ eval_loss = 0.35306716767641216
1221
+ f1 = 0.9035532994923858
1222
+ global_step = 13599
1223
+ loss = 0.25064265978581474
1224
+ rep_loss = 0.0
1225
+ acc = 0.8725490196078431
1226
+ acc_and_f1 = 0.8922067131937521
1227
+ att_loss = 0.0
1228
+ cls_loss = 0.25063442940874736
1229
+ eval_loss = 0.3472353391922437
1230
+ f1 = 0.911864406779661
1231
+ global_step = 13699
1232
+ loss = 0.25063442940874736
1233
+ rep_loss = 0.0
1234
+ acc = 0.8676470588235294
1235
+ acc_and_f1 = 0.887905162064826
1236
+ att_loss = 0.0
1237
+ cls_loss = 0.2506325423893714
1238
+ eval_loss = 0.34969039949086994
1239
+ f1 = 0.9081632653061225
1240
+ global_step = 13799
1241
+ loss = 0.2506325423893714
1242
+ rep_loss = 0.0
1243
+ acc = 0.8725490196078431
1244
+ acc_and_f1 = 0.8922067131937521
1245
+ att_loss = 0.0
1246
+ cls_loss = 0.250672040370721
1247
+ eval_loss = 0.34627919930678147
1248
+ f1 = 0.911864406779661
1249
+ global_step = 13899
1250
+ loss = 0.250672040370721
1251
+ rep_loss = 0.0
1252
+ acc = 0.8700980392156863
1253
+ acc_and_f1 = 0.8900575085721896
1254
+ att_loss = 0.0
1255
+ cls_loss = 0.250632513108565
1256
+ eval_loss = 0.350477954516044
1257
+ f1 = 0.9100169779286927
1258
+ global_step = 13999
1259
+ loss = 0.250632513108565
1260
+ rep_loss = 0.0
1261
+ acc = 0.8602941176470589
1262
+ acc_and_f1 = 0.8822478991596638
1263
+ att_loss = 0.0
1264
+ cls_loss = 0.24775300006712636
1265
+ eval_loss = 0.3548291176557541
1266
+ f1 = 0.9042016806722689
1267
+ global_step = 14099
1268
+ loss = 0.24775300006712636
1269
+ rep_loss = 0.0
1270
+ acc = 0.8700980392156863
1271
+ acc_and_f1 = 0.8902097641086892
1272
+ att_loss = 0.0
1273
+ cls_loss = 0.25028811759166136
1274
+ eval_loss = 0.3484991811788999
1275
+ f1 = 0.9103214890016921
1276
+ global_step = 14199
1277
+ loss = 0.25028811759166136
1278
+ rep_loss = 0.0
1279
+ acc = 0.8602941176470589
1280
+ acc_and_f1 = 0.8822478991596638
1281
+ att_loss = 0.0
1282
+ cls_loss = 0.2508955852680908
1283
+ eval_loss = 0.35242691062963927
1284
+ f1 = 0.9042016806722689
1285
+ global_step = 14299
1286
+ loss = 0.2508955852680908
1287
+ rep_loss = 0.0
1288
+ acc = 0.8602941176470589
1289
+ acc_and_f1 = 0.8820863505604604
1290
+ att_loss = 0.0
1291
+ cls_loss = 0.2512200890711067
1292
+ eval_loss = 0.3515079812361644
1293
+ f1 = 0.9038785834738617
1294
+ global_step = 14399
1295
+ loss = 0.2512200890711067
1296
+ rep_loss = 0.0
1297
+ acc = 0.8578431372549019
1298
+ acc_and_f1 = 0.8802638505066456
1299
+ att_loss = 0.0
1300
+ cls_loss = 0.2510506063008253
1301
+ eval_loss = 0.3505384589617069
1302
+ f1 = 0.9026845637583892
1303
+ global_step = 14499
1304
+ loss = 0.2510506063008253
1305
+ rep_loss = 0.0
1306
+ acc = 0.8529411764705882
1307
+ acc_and_f1 = 0.8766367011921048
1308
+ att_loss = 0.0
1309
+ cls_loss = 0.2510734923927573
1310
+ eval_loss = 0.35494573758198666
1311
+ f1 = 0.9003322259136213
1312
+ global_step = 14599
1313
+ loss = 0.2510734923927573
1314
+ rep_loss = 0.0
1315
+ acc = 0.8676470588235294
1316
+ acc_and_f1 = 0.8882154213036566
1317
+ att_loss = 0.0
1318
+ cls_loss = 0.2504184823689861
1319
+ eval_loss = 0.3464414981695322
1320
+ f1 = 0.9087837837837838
1321
+ global_step = 14699
1322
+ loss = 0.2504184823689861
1323
+ rep_loss = 0.0
1324
+ acc = 0.8602941176470589
1325
+ acc_and_f1 = 0.8822478991596638
1326
+ att_loss = 0.0
1327
+ cls_loss = 0.25041205412548967
1328
+ eval_loss = 0.34841770506822145
1329
+ f1 = 0.9042016806722689
1330
+ global_step = 14799
1331
+ loss = 0.25041205412548967
1332
+ rep_loss = 0.0
1333
+ acc = 0.8700980392156863
1334
+ acc_and_f1 = 0.8900575085721896
1335
+ att_loss = 0.0
1336
+ cls_loss = 0.25021783050001745
1337
+ eval_loss = 0.3431699975178792
1338
+ f1 = 0.9100169779286927
1339
+ global_step = 14899
1340
+ loss = 0.25021783050001745
1341
+ rep_loss = 0.0
1342
+ acc = 0.8504901960784313
1343
+ acc_and_f1 = 0.873984593837535
1344
+ att_loss = 0.0
1345
+ cls_loss = 0.25025104604077264
1346
+ eval_loss = 0.3495776573052773
1347
+ f1 = 0.8974789915966387
1348
+ global_step = 14999
1349
+ loss = 0.25025104604077264
1350
+ rep_loss = 0.0
1351
+ acc = 0.8651960784313726
1352
+ acc_and_f1 = 0.8862236715934266
1353
+ att_loss = 0.0
1354
+ cls_loss = 0.25033993448133496
1355
+ eval_loss = 0.34707575004834396
1356
+ f1 = 0.9072512647554806
1357
+ global_step = 15099
1358
+ loss = 0.25033993448133496
1359
+ rep_loss = 0.0
1360
+ acc = 0.8725490196078431
1361
+ acc_and_f1 = 0.8920568227290917
1362
+ att_loss = 0.0
1363
+ cls_loss = 0.2503684941089649
1364
+ eval_loss = 0.34068380066981685
1365
+ f1 = 0.9115646258503401
1366
+ global_step = 15199
1367
+ loss = 0.2503684941089649
1368
+ rep_loss = 0.0
1369
+ acc = 0.8799019607843137
1370
+ acc_and_f1 = 0.8982133313291245
1371
+ att_loss = 0.0
1372
+ cls_loss = 0.25035796194033927
1373
+ eval_loss = 0.341502499121886
1374
+ f1 = 0.9165247018739353
1375
+ global_step = 15299
1376
+ loss = 0.25035796194033927
1377
+ rep_loss = 0.0
1378
+ acc = 0.875
1379
+ acc_and_f1 = 0.8942062818336163
1380
+ att_loss = 0.0
1381
+ cls_loss = 0.2503947774822957
1382
+ eval_loss = 0.3458839265199808
1383
+ f1 = 0.9134125636672326
1384
+ global_step = 15399
1385
+ loss = 0.2503947774822957
1386
+ rep_loss = 0.0
1387
+ acc = 0.8700980392156863
1388
+ acc_and_f1 = 0.8899042155192571
1389
+ att_loss = 0.0
1390
+ cls_loss = 0.2503597757705519
1391
+ eval_loss = 0.34732022881507874
1392
+ f1 = 0.909710391822828
1393
+ global_step = 15499
1394
+ loss = 0.2503597757705519
1395
+ rep_loss = 0.0
1396
+ acc = 0.8676470588235294
1397
+ acc_and_f1 = 0.8882154213036566
1398
+ att_loss = 0.0
1399
+ cls_loss = 0.2503518128243246
1400
+ eval_loss = 0.35214132414414334
1401
+ f1 = 0.9087837837837838
1402
+ global_step = 15599
1403
+ loss = 0.2503518128243246
1404
+ rep_loss = 0.0
1405
+ acc = 0.8553921568627451
1406
+ acc_and_f1 = 0.8786112198623209
1407
+ att_loss = 0.0
1408
+ cls_loss = 0.250352953718821
1409
+ eval_loss = 0.3568104081428968
1410
+ f1 = 0.9018302828618968
1411
+ global_step = 15699
1412
+ loss = 0.250352953718821
1413
+ rep_loss = 0.0
1414
+ acc = 0.8627450980392157
1415
+ acc_and_f1 = 0.8840752517223105
1416
+ att_loss = 0.0
1417
+ cls_loss = 0.2503842528089494
1418
+ eval_loss = 0.3509543159833321
1419
+ f1 = 0.9054054054054054
1420
+ global_step = 15799
1421
+ loss = 0.2503842528089494
1422
+ rep_loss = 0.0
1423
+ acc = 0.8700980392156863
1424
+ acc_and_f1 = 0.8900575085721896
1425
+ att_loss = 0.0
1426
+ cls_loss = 0.2502596350900467
1427
+ eval_loss = 0.3475727037741588
1428
+ f1 = 0.9100169779286927
1429
+ global_step = 15899
1430
+ loss = 0.2502596350900467
1431
+ rep_loss = 0.0
1432
+ acc = 0.8602941176470589
1433
+ acc_and_f1 = 0.8822478991596638
1434
+ att_loss = 0.0
1435
+ cls_loss = 0.2502096541677344
1436
+ eval_loss = 0.3506972056168776
1437
+ f1 = 0.9042016806722689
1438
+ global_step = 15999
1439
+ loss = 0.2502096541677344
1440
+ rep_loss = 0.0
1441
+ acc = 0.8602941176470589
1442
+ acc_and_f1 = 0.8822478991596638
1443
+ att_loss = 0.0
1444
+ cls_loss = 0.25021608849571936
1445
+ eval_loss = 0.3495794569070523
1446
+ f1 = 0.9042016806722689
1447
+ global_step = 16099
1448
+ loss = 0.25021608849571936
1449
+ rep_loss = 0.0
1450
+ acc = 0.875
1451
+ acc_and_f1 = 0.8940587734241908
1452
+ att_loss = 0.0
1453
+ cls_loss = 0.2501752441587274
1454
+ eval_loss = 0.3453184182827289
1455
+ f1 = 0.9131175468483816
1456
+ global_step = 16199
1457
+ loss = 0.2501752441587274
1458
+ rep_loss = 0.0
1459
+ acc = 0.8676470588235294
1460
+ acc_and_f1 = 0.8882154213036566
1461
+ att_loss = 0.0
1462
+ cls_loss = 0.25031862577238945
1463
+ eval_loss = 0.344640382207357
1464
+ f1 = 0.9087837837837838
1465
+ global_step = 16299
1466
+ loss = 0.25031862577238945
1467
+ rep_loss = 0.0
1468
+ acc = 0.8578431372549019
1469
+ acc_and_f1 = 0.8802638505066456
1470
+ att_loss = 0.0
1471
+ cls_loss = 0.2503466642066574
1472
+ eval_loss = 0.3490766011751615
1473
+ f1 = 0.9026845637583892
1474
+ global_step = 16399
1475
+ loss = 0.2503466642066574
1476
+ rep_loss = 0.0
1477
+ acc = 0.8676470588235294
1478
+ acc_and_f1 = 0.8880608175473579
1479
+ att_loss = 0.0
1480
+ cls_loss = 0.25038913499789195
1481
+ eval_loss = 0.34382400260521817
1482
+ f1 = 0.9084745762711864
1483
+ global_step = 16499
1484
+ loss = 0.25038913499789195
1485
+ rep_loss = 0.0
1486
+ acc = 0.8651960784313726
1487
+ acc_and_f1 = 0.8860667363392057
1488
+ att_loss = 0.0
1489
+ cls_loss = 0.250366505928682
1490
+ eval_loss = 0.3444054745710813
1491
+ f1 = 0.9069373942470389
1492
+ global_step = 16599
1493
+ loss = 0.250366505928682
1494
+ rep_loss = 0.0
1495
+ acc = 0.8676470588235294
1496
+ acc_and_f1 = 0.8882154213036566
1497
+ att_loss = 0.0
1498
+ cls_loss = 0.2503603221866306
1499
+ eval_loss = 0.3445770419560946
1500
+ f1 = 0.9087837837837838
1501
+ global_step = 16699
1502
+ loss = 0.2503603221866306
1503
+ rep_loss = 0.0
1504
+ acc = 0.8602941176470589
1505
+ acc_and_f1 = 0.8820863505604604
1506
+ att_loss = 0.0
1507
+ cls_loss = 0.250318055251543
1508
+ eval_loss = 0.3461114030617934
1509
+ f1 = 0.9038785834738617
1510
+ global_step = 16799
1511
+ loss = 0.250318055251543
1512
+ rep_loss = 0.0
1513
+ acc = 0.8578431372549019
1514
+ acc_and_f1 = 0.8801000198059021
1515
+ att_loss = 0.0
1516
+ cls_loss = 0.25031335944528926
1517
+ eval_loss = 0.3489151998208119
1518
+ f1 = 0.9023569023569024
1519
+ global_step = 16899
1520
+ loss = 0.25031335944528926
1521
+ rep_loss = 0.0
1522
+ acc = 0.8700980392156863
1523
+ acc_and_f1 = 0.8900575085721896
1524
+ att_loss = 0.0
1525
+ cls_loss = 0.25023560517275956
1526
+ eval_loss = 0.344931518802276
1527
+ f1 = 0.9100169779286927
1528
+ global_step = 16999
1529
+ loss = 0.25023560517275956
1530
+ rep_loss = 0.0
1531
+ acc = 0.8676470588235294
1532
+ acc_and_f1 = 0.8880608175473579
1533
+ att_loss = 0.0
1534
+ cls_loss = 0.25021038493261366
1535
+ eval_loss = 0.3449639528989792
1536
+ f1 = 0.9084745762711864
1537
+ global_step = 17099
1538
+ loss = 0.25021038493261366
1539
+ rep_loss = 0.0
1540
+ acc = 0.8676470588235294
1541
+ acc_and_f1 = 0.8880608175473579
1542
+ att_loss = 0.0
1543
+ cls_loss = 0.25012775577892704
1544
+ eval_loss = 0.34533809240047747
1545
+ f1 = 0.9084745762711864
1546
+ global_step = 17199
1547
+ loss = 0.25012775577892704
1548
+ rep_loss = 0.0
1549
+ acc = 0.8700980392156863
1550
+ acc_and_f1 = 0.8899042155192571
1551
+ att_loss = 0.0
1552
+ cls_loss = 0.25011522198140307
1553
+ eval_loss = 0.3427927837922023
1554
+ f1 = 0.909710391822828
1555
+ global_step = 17299
1556
+ loss = 0.25011522198140307
1557
+ rep_loss = 0.0
1558
+ acc = 0.8651960784313726
1559
+ acc_and_f1 = 0.8863795518207283
1560
+ att_loss = 0.0
1561
+ cls_loss = 0.2501817453027452
1562
+ eval_loss = 0.34944057579223925
1563
+ f1 = 0.907563025210084
1564
+ global_step = 17399
1565
+ loss = 0.2501817453027452
1566
+ rep_loss = 0.0
1567
+ acc = 0.8504901960784313
1568
+ acc_and_f1 = 0.8743269010442241
1569
+ att_loss = 0.0
1570
+ cls_loss = 0.25017694739776325
1571
+ eval_loss = 0.3516889890799156
1572
+ f1 = 0.8981636060100167
1573
+ global_step = 17499
1574
+ loss = 0.25017694739776325
1575
+ rep_loss = 0.0
1576
+ acc = 0.8651960784313726
1577
+ acc_and_f1 = 0.8862236715934266
1578
+ att_loss = 0.0
1579
+ cls_loss = 0.2501919267958131
1580
+ eval_loss = 0.3467919322160574
1581
+ f1 = 0.9072512647554806
1582
+ global_step = 17599
1583
+ loss = 0.2501919267958131
1584
+ rep_loss = 0.0
1585
+ acc = 0.8799019607843137
1586
+ acc_and_f1 = 0.8979269666700299
1587
+ att_loss = 0.0
1588
+ cls_loss = 0.25025702385871545
1589
+ eval_loss = 0.3421338934164781
1590
+ f1 = 0.9159519725557461
1591
+ global_step = 17699
1592
+ loss = 0.25025702385871545
1593
+ rep_loss = 0.0
1594
+ acc = 0.8676470588235294
1595
+ acc_and_f1 = 0.8882154213036566
1596
+ att_loss = 0.0
1597
+ cls_loss = 0.2502066430170713
1598
+ eval_loss = 0.3479071053174826
1599
+ f1 = 0.9087837837837838
1600
+ global_step = 17799
1601
+ loss = 0.2502066430170713
1602
+ rep_loss = 0.0
1603
+ acc = 0.8774509803921569
1604
+ acc_and_f1 = 0.8962084833933573
1605
+ att_loss = 0.0
1606
+ cls_loss = 0.2501907047070028
1607
+ eval_loss = 0.3449676117071739
1608
+ f1 = 0.9149659863945578
1609
+ global_step = 17899
1610
+ loss = 0.2501907047070028
1611
+ rep_loss = 0.0
1612
+ acc = 0.8651960784313726
1613
+ acc_and_f1 = 0.8862236715934266
1614
+ att_loss = 0.0
1615
+ cls_loss = 0.2501619358882744
1616
+ eval_loss = 0.3473269148514821
1617
+ f1 = 0.9072512647554806
1618
+ global_step = 17999
1619
+ loss = 0.2501619358882744
1620
+ rep_loss = 0.0
1621
+ acc = 0.8651960784313726
1622
+ acc_and_f1 = 0.8862236715934266
1623
+ att_loss = 0.0
1624
+ cls_loss = 0.25011761821076345
1625
+ eval_loss = 0.3486064626620366
1626
+ f1 = 0.9072512647554806
1627
+ global_step = 18099
1628
+ loss = 0.25011761821076345
1629
+ rep_loss = 0.0
1630
+ acc = 0.8725490196078431
1631
+ acc_and_f1 = 0.8922067131937521
1632
+ att_loss = 0.0
1633
+ cls_loss = 0.25002052384076756
1634
+ eval_loss = 0.3449101081261268
1635
+ f1 = 0.911864406779661
1636
+ global_step = 18199
1637
+ loss = 0.25002052384076756
1638
+ rep_loss = 0.0
1639
+ acc = 0.8651960784313726
1640
+ acc_and_f1 = 0.8862236715934266
1641
+ att_loss = 0.0
1642
+ cls_loss = 0.24997850480555248
1643
+ eval_loss = 0.34695371985435486
1644
+ f1 = 0.9072512647554806
1645
+ global_step = 18299
1646
+ loss = 0.24997850480555248
1647
+ rep_loss = 0.0
1648
+ acc = 0.8700980392156863
1649
+ acc_and_f1 = 0.8902097641086892
1650
+ att_loss = 0.0
1651
+ cls_loss = 0.24999612231943796
1652
+ eval_loss = 0.3450571711246784
1653
+ f1 = 0.9103214890016921
1654
+ global_step = 18399
1655
+ loss = 0.24999612231943796
1656
+ rep_loss = 0.0
1657
+ acc = 0.8651960784313726
1658
+ acc_and_f1 = 0.8862236715934266
1659
+ att_loss = 0.0
1660
+ cls_loss = 0.2500270959406022
1661
+ eval_loss = 0.34581805765628815
1662
+ f1 = 0.9072512647554806
1663
+ global_step = 18499
1664
+ loss = 0.2500270959406022
1665
+ rep_loss = 0.0
1666
+ acc = 0.8676470588235294
1667
+ acc_and_f1 = 0.8882154213036566
1668
+ att_loss = 0.0
1669
+ cls_loss = 0.25004593747139925
1670
+ eval_loss = 0.34492750695118535
1671
+ f1 = 0.9087837837837838
1672
+ global_step = 18599
1673
+ loss = 0.25004593747139925
1674
+ rep_loss = 0.0
1675
+ acc = 0.8725490196078431
1676
+ acc_and_f1 = 0.8922067131937521
1677
+ att_loss = 0.0
1678
+ cls_loss = 0.2500000122594308
1679
+ eval_loss = 0.3441122449361361
1680
+ f1 = 0.911864406779661
1681
+ global_step = 18699
1682
+ loss = 0.2500000122594308
1683
+ rep_loss = 0.0
1684
+ acc = 0.8627450980392157
1685
+ acc_and_f1 = 0.8842345018815607
1686
+ att_loss = 0.0
1687
+ cls_loss = 0.24999139947244667
1688
+ eval_loss = 0.34669444308831143
1689
+ f1 = 0.9057239057239057
1690
+ global_step = 18799
1691
+ loss = 0.24999139947244667
1692
+ rep_loss = 0.0
1693
+ acc = 0.8725490196078431
1694
+ acc_and_f1 = 0.8922067131937521
1695
+ att_loss = 0.0
1696
+ cls_loss = 0.2499992541553959
1697
+ eval_loss = 0.34382077707694125
1698
+ f1 = 0.911864406779661
1699
+ global_step = 18899
1700
+ loss = 0.2499992541553959
1701
+ rep_loss = 0.0
1702
+ acc = 0.8627450980392157
1703
+ acc_and_f1 = 0.8842345018815607
1704
+ att_loss = 0.0
1705
+ cls_loss = 0.2500313110929115
1706
+ eval_loss = 0.3455964017372865
1707
+ f1 = 0.9057239057239057
1708
+ global_step = 18999
1709
+ loss = 0.2500313110929115
1710
+ rep_loss = 0.0
1711
+ acc = 0.8627450980392157
1712
+ acc_and_f1 = 0.8842345018815607
1713
+ att_loss = 0.0
1714
+ cls_loss = 0.25004326058557225
1715
+ eval_loss = 0.3483260709505815
1716
+ f1 = 0.9057239057239057
1717
+ global_step = 19099
1718
+ loss = 0.25004326058557225
1719
+ rep_loss = 0.0
1720
+ acc = 0.8627450980392157
1721
+ acc_and_f1 = 0.8842345018815607
1722
+ att_loss = 0.0
1723
+ cls_loss = 0.25000432810630513
1724
+ eval_loss = 0.3474111511157109
1725
+ f1 = 0.9057239057239057
1726
+ global_step = 19199
1727
+ loss = 0.25000432810630513
1728
+ rep_loss = 0.0
1729
+ acc = 0.8627450980392157
1730
+ acc_and_f1 = 0.8842345018815607
1731
+ att_loss = 0.0
1732
+ cls_loss = 0.25002709521923255
1733
+ eval_loss = 0.3503383317819008
1734
+ f1 = 0.9057239057239057
1735
+ global_step = 19299
1736
+ loss = 0.25002709521923255
1737
+ rep_loss = 0.0
1738
+ acc = 0.8676470588235294
1739
+ acc_and_f1 = 0.8882154213036566
1740
+ att_loss = 0.0
1741
+ cls_loss = 0.2499719287391631
1742
+ eval_loss = 0.3482268200470851
1743
+ f1 = 0.9087837837837838
1744
+ global_step = 19399
1745
+ loss = 0.2499719287391631
1746
+ rep_loss = 0.0
1747
+ acc = 0.8676470588235294
1748
+ acc_and_f1 = 0.8882154213036566
1749
+ att_loss = 0.0
1750
+ cls_loss = 0.24998016587675134
1751
+ eval_loss = 0.34600579050871044
1752
+ f1 = 0.9087837837837838
1753
+ global_step = 19499
1754
+ loss = 0.24998016587675134
1755
+ rep_loss = 0.0
1756
+ acc = 0.8725490196078431
1757
+ acc_and_f1 = 0.8922067131937521
1758
+ att_loss = 0.0
1759
+ cls_loss = 0.2499536421803966
1760
+ eval_loss = 0.3457164294444598
1761
+ f1 = 0.911864406779661
1762
+ global_step = 19599
1763
+ loss = 0.2499536421803966
1764
+ rep_loss = 0.0
1765
+ acc = 0.8676470588235294
1766
+ acc_and_f1 = 0.8882154213036566
1767
+ att_loss = 0.0
1768
+ cls_loss = 0.2499844801852513
1769
+ eval_loss = 0.34607594402936787
1770
+ f1 = 0.9087837837837838
1771
+ global_step = 19699
1772
+ loss = 0.2499844801852513
1773
+ rep_loss = 0.0
1774
+ acc = 0.8676470588235294
1775
+ acc_and_f1 = 0.8882154213036566
1776
+ att_loss = 0.0
1777
+ cls_loss = 0.24993976037491056
1778
+ eval_loss = 0.3465753828103726
1779
+ f1 = 0.9087837837837838
1780
+ global_step = 19799
1781
+ loss = 0.24993976037491056
1782
+ rep_loss = 0.0
1783
+ acc = 0.8700980392156863
1784
+ acc_and_f1 = 0.8902097641086892
1785
+ att_loss = 0.0
1786
+ cls_loss = 0.24996107566767584
1787
+ eval_loss = 0.3468713015317917
1788
+ f1 = 0.9103214890016921
1789
+ global_step = 19899
1790
+ loss = 0.24996107566767584
1791
+ rep_loss = 0.0
1792
+ acc = 0.8700980392156863
1793
+ acc_and_f1 = 0.8902097641086892
1794
+ att_loss = 0.0
1795
+ cls_loss = 0.24992618599381783
1796
+ eval_loss = 0.34673953514832717
1797
+ f1 = 0.9103214890016921
1798
+ global_step = 19999
1799
+ loss = 0.24992618599381783
1800
+ rep_loss = 0.0
1801
+ acc = 0.8700980392156863
1802
+ acc_and_f1 = 0.8902097641086892
1803
+ att_loss = 0.0
1804
+ cls_loss = 0.24993879183610465
1805
+ eval_loss = 0.34482155396388126
1806
+ f1 = 0.9103214890016921
1807
+ global_step = 20099
1808
+ loss = 0.24993879183610465
1809
+ rep_loss = 0.0
1810
+ acc = 0.8700980392156863
1811
+ acc_and_f1 = 0.8902097641086892
1812
+ att_loss = 0.0
1813
+ cls_loss = 0.24991122533691354
1814
+ eval_loss = 0.3457224632685001
1815
+ f1 = 0.9103214890016921
1816
+ global_step = 20199
1817
+ loss = 0.24991122533691354
1818
+ rep_loss = 0.0
1819
+ acc = 0.8725490196078431
1820
+ acc_and_f1 = 0.8922067131937521
1821
+ att_loss = 0.0
1822
+ cls_loss = 0.2499278331071717
1823
+ eval_loss = 0.3458889149702512
1824
+ f1 = 0.911864406779661
1825
+ global_step = 20299
1826
+ loss = 0.2499278331071717
1827
+ rep_loss = 0.0
1828
+ acc = 0.8725490196078431
1829
+ acc_and_f1 = 0.8922067131937521
1830
+ att_loss = 0.0
1831
+ cls_loss = 0.2499310848526886
1832
+ eval_loss = 0.345659288076254
1833
+ f1 = 0.911864406779661
1834
+ global_step = 20399
1835
+ loss = 0.2499310848526886
1836
+ rep_loss = 0.0
1837
+ acc = 0.8676470588235294
1838
+ acc_and_f1 = 0.8882154213036566
1839
+ att_loss = 0.0
1840
+ cls_loss = 0.24990426886509892
1841
+ eval_loss = 0.3469357112279305
1842
+ f1 = 0.9087837837837838
1843
+ global_step = 20499
1844
+ loss = 0.24990426886509892
1845
+ rep_loss = 0.0
1846
+ acc = 0.8725490196078431
1847
+ acc_and_f1 = 0.8922067131937521
1848
+ att_loss = 0.0
1849
+ cls_loss = 0.2499004627529904
1850
+ eval_loss = 0.34655847343114704
1851
+ f1 = 0.911864406779661
1852
+ global_step = 20599
1853
+ loss = 0.2499004627529904
1854
+ rep_loss = 0.0
1855
+ acc = 0.8725490196078431
1856
+ acc_and_f1 = 0.8922067131937521
1857
+ att_loss = 0.0
1858
+ cls_loss = 0.2498836103085748
1859
+ eval_loss = 0.34516667288083297
1860
+ f1 = 0.911864406779661
1861
+ global_step = 20699
1862
+ loss = 0.2498836103085748
1863
+ rep_loss = 0.0
1864
+ acc = 0.8725490196078431
1865
+ acc_and_f1 = 0.8922067131937521
1866
+ att_loss = 0.0
1867
+ cls_loss = 0.2498801494756933
1868
+ eval_loss = 0.34544021578935474
1869
+ f1 = 0.911864406779661
1870
+ global_step = 20799
1871
+ loss = 0.2498801494756933
1872
+ rep_loss = 0.0
1873
+ acc = 0.8725490196078431
1874
+ acc_and_f1 = 0.8922067131937521
1875
+ att_loss = 0.0
1876
+ cls_loss = 0.24983511446527848
1877
+ eval_loss = 0.3453839478584436
1878
+ f1 = 0.911864406779661
1879
+ global_step = 20899
1880
+ loss = 0.24983511446527848
1881
+ rep_loss = 0.0
1882
+ acc = 0.8725490196078431
1883
+ acc_and_f1 = 0.8922067131937521
1884
+ att_loss = 0.0
1885
+ cls_loss = 0.24985123587934302
1886
+ eval_loss = 0.3454859256744385
1887
+ f1 = 0.911864406779661
1888
+ global_step = 20999
1889
+ loss = 0.24985123587934302
1890
+ rep_loss = 0.0
1891
+ acc = 0.8725490196078431
1892
+ acc_and_f1 = 0.8922067131937521
1893
+ att_loss = 0.0
1894
+ cls_loss = 0.24985949066298696
1895
+ eval_loss = 0.3453449228635201
1896
+ f1 = 0.911864406779661
1897
+ global_step = 21099
1898
+ loss = 0.24985949066298696
1899
+ rep_loss = 0.0
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:51096df6c46fbf9fd273231ee36f02fe1d8a0b67ca29755d0a07fc8e4e1bf826
3
+ size 270232335
vocab.txt ADDED
The diff for this file is too large to render. See raw diff