Hugging Face's logo

ledas
/
RR-RE_extreme

Model card Files
xet
 Community
ledas commited on
Commit
2b193a2
·
verified ·
1 Parent(s): 631c600

Upload log_seed_62.log with huggingface_hub

Browse files
Files changed (1) hide show
  1. log_seed_62.log +452 -0
log_seed_62.log ADDED
@@ -0,0 +1,452 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [2026-04-16 11:36:17,409][__main__][INFO] - seed: 62
2
+ result_path: ${hydra:runtime.output_dir}
3
+ misc:
4
+ bfloat16: false
5
+ wandb: true
6
+ compile_model: false
7
+ save_cloud: false
8
+ encoder_lazy_load: false
9
+ dataset:
10
+ name: ReDocRED
11
+ path: data/redocred
12
+ rel2id: ${dataset.path}/rel2id.json
13
+ sets:
14
+ train: ${dataset.path}/train_ext.json
15
+ dev: ${dataset.path}/dev_ext.json
16
+ test: ${dataset.path}/test_revised.json
17
+ num_class: 97
18
+ max_seq_length: 1024
19
+ cached_location: ${dataset.path}/cached_ext.pkl
20
+ encoder:
21
+ name: roberta-large
22
+ transformer_type: roberta
23
+ attn_impl: null
24
+ lazy: ${misc.encoder_lazy_load}
25
+ trainer:
26
+ use_bfloat16: ${misc.bfloat16}
27
+ test_best_dev: false
28
+ epochs: 30
29
+ batch_size: 3
30
+ eval_freq: -1
31
+ print_freq: 100
32
+ max_grad_norm: -1
33
+ model_save: ${result_path}/best.pt
34
+ ema: true
35
+ ema_rate: 0.999
36
+ pretrain_lr: 1.0e-05
37
+ new_lr: 8.0e-05
38
+ warmup_steps: 1000
39
+ optimizer_cfg:
40
+ name: AdamW
41
+ kwargs:
42
+ betas:
43
+ - 0.9
44
+ - 0.98
45
+ eps: 1.0e-09
46
+ wandb:
47
+ used: ${misc.wandb}
48
+ entity: triyn01052004-vietnam-national-university-hanoi
49
+ project: RR-RE
50
+ tester:
51
+ batch_size: 4
52
+ rel2id: ${dataset.path}/rel2id.json
53
+ data_dir: ${dataset.path}
54
+ sets: ${dataset.sets}
55
+ loss:
56
+ tag: S_PU
57
+ num_class: ${dataset.num_class}
58
+ S_PU_e: 12.0
59
+ S_PU_margin: 0.25
60
+ S_PU_beta: 0.0
61
+ model:
62
+ is_compiled: ${misc.compile_model}
63
+ num_class: ${dataset.num_class}
64
+ H_layers: 2
65
+ L_layers: 1
66
+ H_cycles: 2
67
+ L_cycles: 1
68
+ num_heads: 8
69
+ expansion: 4
70
+ num_tokens: 64
71
+ rope_theta: 10000.0
72
+ single_net: true
73
+ use_mlp: true
74
+ dropout_rate: 0.1
75
+ topk: 0
76
+ halt_max_steps: 1
77
+ saver:
78
+ is_saved: ${misc.save_cloud}
79
+ repo_id: ledas/RR-RE
80
+ model_save: ${trainer.model_save}
81
+
82
+ [2026-04-16 11:36:20,499][datasets.redocred][INFO] - use cached dataset.
83
+ [2026-04-16 11:36:33,622][trainer][INFO] - Use float32.
84
+ [2026-04-16 11:36:33,641][trainer][INFO] - Total update steps: 30510
85
+ [2026-04-16 11:36:33,642][trainer][INFO] - Total warmup steps: 1000
86
+ [2026-04-16 11:36:33,642][trainer][INFO] - epoch 1/30 ====================================================================================================
87
+ [2026-04-16 11:37:00,842][trainer][INFO] - step: 100, batch loss: 14.372481344308643
88
+ [2026-04-16 11:37:28,549][trainer][INFO] - step: 200, batch loss: 5.1498959946189595
89
+ [2026-04-16 11:37:55,999][trainer][INFO] - step: 300, batch loss: 2.358692680218233
90
+ [2026-04-16 11:38:24,222][trainer][INFO] - step: 400, batch loss: 1.3536494553108755
91
+ [2026-04-16 11:38:52,788][trainer][INFO] - step: 500, batch loss: 0.8644837588153416
92
+ [2026-04-16 11:39:20,355][trainer][INFO] - step: 600, batch loss: 0.6187835970507837
93
+ [2026-04-16 11:39:49,886][trainer][INFO] - step: 700, batch loss: 0.4778190415339417
94
+ [2026-04-16 11:40:18,451][trainer][INFO] - step: 800, batch loss: 0.39078212589186007
95
+ [2026-04-16 11:40:46,723][trainer][INFO] - step: 900, batch loss: 0.3211043474984199
96
+ [2026-04-16 11:41:15,222][trainer][INFO] - step: 1000, batch loss: 0.26276470400777957
97
+ [2026-04-16 11:41:20,443][trainer][INFO] - epoch: 1, loss=2621.005068729941 .
98
+ [2026-04-16 11:41:20,444][trainer][INFO] - epoch 2/30 ====================================================================================================
99
+ [2026-04-16 11:41:44,209][trainer][INFO] - step: 1100, batch loss: 0.21486287607140436
100
+ [2026-04-16 11:42:13,384][trainer][INFO] - step: 1200, batch loss: 0.181058871229292
101
+ [2026-04-16 11:42:42,225][trainer][INFO] - step: 1300, batch loss: 0.15291562010620402
102
+ [2026-04-16 11:43:10,077][trainer][INFO] - step: 1400, batch loss: 0.12953796291767286
103
+ [2026-04-16 11:43:39,328][trainer][INFO] - step: 1500, batch loss: 0.11205477635276098
104
+ [2026-04-16 11:44:07,061][trainer][INFO] - step: 1600, batch loss: 0.09823677934871813
105
+ [2026-04-16 11:44:35,320][trainer][INFO] - step: 1700, batch loss: 0.0841140382858507
106
+ [2026-04-16 11:45:03,652][trainer][INFO] - step: 1800, batch loss: 0.0769994029906792
107
+ [2026-04-16 11:45:31,735][trainer][INFO] - step: 1900, batch loss: 0.06676817794979696
108
+ [2026-04-16 11:46:00,082][trainer][INFO] - step: 2000, batch loss: 0.057951557045442706
109
+ [2026-04-16 11:46:09,975][trainer][INFO] - epoch: 2, loss=115.2898217951255 .
110
+ [2026-04-16 11:46:09,976][trainer][INFO] - epoch 3/30 ====================================================================================================
111
+ [2026-04-16 11:46:28,986][trainer][INFO] - step: 2100, batch loss: 0.050982800820382035
112
+ [2026-04-16 11:46:57,208][trainer][INFO] - step: 2200, batch loss: 0.045417678612340494
113
+ [2026-04-16 11:47:25,859][trainer][INFO] - step: 2300, batch loss: 0.04243810363895698
114
+ [2026-04-16 11:47:53,963][trainer][INFO] - step: 2400, batch loss: 0.0385742968778258
115
+ [2026-04-16 11:48:21,053][trainer][INFO] - step: 2500, batch loss: 0.03630768215636435
116
+ [2026-04-16 11:48:49,066][trainer][INFO] - step: 2600, batch loss: 0.03260645932821213
117
+ [2026-04-16 11:49:18,302][trainer][INFO] - step: 2700, batch loss: 0.030585853820789752
118
+ [2026-04-16 11:49:47,602][trainer][INFO] - step: 2800, batch loss: 0.027651222872184137
119
+ [2026-04-16 11:50:15,904][trainer][INFO] - step: 2900, batch loss: 0.026540609397900802
120
+ [2026-04-16 11:50:44,276][trainer][INFO] - step: 3000, batch loss: 0.024220742362400367
121
+ [2026-04-16 11:50:58,047][trainer][INFO] - epoch: 3, loss=34.9365116394468 .
122
+ [2026-04-16 11:50:58,048][trainer][INFO] - epoch 4/30 ====================================================================================================
123
+ [2026-04-16 11:51:12,517][trainer][INFO] - step: 3100, batch loss: 0.022548734056421885
124
+ [2026-04-16 11:51:40,708][trainer][INFO] - step: 3200, batch loss: 0.021157247288277815
125
+ [2026-04-16 11:52:07,692][trainer][INFO] - step: 3300, batch loss: 0.02041594522093551
126
+ [2026-04-16 11:52:35,336][trainer][INFO] - step: 3400, batch loss: 0.018750325838973875
127
+ [2026-04-16 11:53:03,783][trainer][INFO] - step: 3500, batch loss: 0.018171414423171516
128
+ [2026-04-16 11:53:31,781][trainer][INFO] - step: 3600, batch loss: 0.017046153149936096
129
+ [2026-04-16 11:53:59,948][trainer][INFO] - step: 3700, batch loss: 0.01710568391571758
130
+ [2026-04-16 11:54:27,822][trainer][INFO] - step: 3800, batch loss: 0.016864234499532848
131
+ [2026-04-16 11:54:56,298][trainer][INFO] - step: 3900, batch loss: 0.015904570179371844
132
+ [2026-04-16 11:55:25,675][trainer][INFO] - step: 4000, batch loss: 0.015590111643653632
133
+ [2026-04-16 11:55:44,412][trainer][INFO] - epoch: 4, loss=18.069890483185354 .
134
+ [2026-04-16 11:55:44,413][trainer][INFO] - epoch 5/30 ====================================================================================================
135
+ [2026-04-16 11:55:53,479][trainer][INFO] - step: 4100, batch loss: 0.013582912205513673
136
+ [2026-04-16 11:56:21,179][trainer][INFO] - step: 4200, batch loss: 0.012360292306482343
137
+ [2026-04-16 11:56:49,456][trainer][INFO] - step: 4300, batch loss: 0.01252288248137216
138
+ [2026-04-16 11:57:18,121][trainer][INFO] - step: 4400, batch loss: 0.012558968219882853
139
+ [2026-04-16 11:57:47,282][trainer][INFO] - step: 4500, batch loss: 0.012186244129059432
140
+ [2026-04-16 11:58:15,454][trainer][INFO] - step: 4600, batch loss: 0.011611059731775085
141
+ [2026-04-16 11:58:43,690][trainer][INFO] - step: 4700, batch loss: 0.010872988982613956
142
+ [2026-04-16 11:59:12,487][trainer][INFO] - step: 4800, batch loss: 0.011392692114693464
143
+ [2026-04-16 11:59:40,680][trainer][INFO] - step: 4900, batch loss: 0.011081511022592186
144
+ [2026-04-16 12:00:07,846][trainer][INFO] - step: 5000, batch loss: 0.010127249668654312
145
+ [2026-04-16 12:00:31,974][trainer][INFO] - epoch: 5, loss=11.802104033883893 .
146
+ [2026-04-16 12:00:31,974][trainer][INFO] - epoch 6/30 ====================================================================================================
147
+ [2026-04-16 12:00:36,499][trainer][INFO] - step: 5100, batch loss: 0.010405419038693913
148
+ [2026-04-16 12:01:04,648][trainer][INFO] - step: 5200, batch loss: 0.009348925272687797
149
+ [2026-04-16 12:01:32,588][trainer][INFO] - step: 5300, batch loss: 0.008846599160556013
150
+ [2026-04-16 12:02:00,627][trainer][INFO] - step: 5400, batch loss: 0.008969164042525795
151
+ [2026-04-16 12:02:28,169][trainer][INFO] - step: 5500, batch loss: 0.008601303553669531
152
+ [2026-04-16 12:02:57,027][trainer][INFO] - step: 5600, batch loss: 0.008682515079731511
153
+ [2026-04-16 12:03:24,524][trainer][INFO] - step: 5700, batch loss: 0.00813845869651484
154
+ [2026-04-16 12:03:51,904][trainer][INFO] - step: 5800, batch loss: 0.008420788731231211
155
+ [2026-04-16 12:04:20,428][trainer][INFO] - step: 5900, batch loss: 0.008579030010389977
156
+ [2026-04-16 12:04:48,956][trainer][INFO] - step: 6000, batch loss: 0.008367034351147403
157
+ [2026-04-16 12:05:17,731][trainer][INFO] - step: 6100, batch loss: 0.008628539295575392
158
+ [2026-04-16 12:05:18,249][trainer][INFO] - epoch: 6, loss=8.823075594215288 .
159
+ [2026-04-16 12:05:18,250][trainer][INFO] - epoch 7/30 ====================================================================================================
160
+ [2026-04-16 12:05:46,627][trainer][INFO] - step: 6200, batch loss: 0.006900199136435453
161
+ [2026-04-16 12:06:14,778][trainer][INFO] - step: 6300, batch loss: 0.007158803172909635
162
+ [2026-04-16 12:06:42,463][trainer][INFO] - step: 6400, batch loss: 0.0075046361084367685
163
+ [2026-04-16 12:07:11,146][trainer][INFO] - step: 6500, batch loss: 0.007108374178161721
164
+ [2026-04-16 12:07:38,380][trainer][INFO] - step: 6600, batch loss: 0.006918792361191551
165
+ [2026-04-16 12:08:06,878][trainer][INFO] - step: 6700, batch loss: 0.007106824280571874
166
+ [2026-04-16 12:08:35,383][trainer][INFO] - step: 6800, batch loss: 0.006816616881940002
167
+ [2026-04-16 12:09:02,073][trainer][INFO] - step: 6900, batch loss: 0.006850832490765389
168
+ [2026-04-16 12:09:30,347][trainer][INFO] - step: 7000, batch loss: 0.007581198519951158
169
+ [2026-04-16 12:09:58,432][trainer][INFO] - step: 7100, batch loss: 0.0070646863528994255
170
+ [2026-04-16 12:10:03,970][trainer][INFO] - epoch: 7, loss=7.20227986491609 .
171
+ [2026-04-16 12:10:03,971][trainer][INFO] - epoch 8/30 ====================================================================================================
172
+ [2026-04-16 12:10:26,115][trainer][INFO] - step: 7200, batch loss: 0.006089581779804737
173
+ [2026-04-16 12:10:55,918][trainer][INFO] - step: 7300, batch loss: 0.006184211708841302
174
+ [2026-04-16 12:11:24,172][trainer][INFO] - step: 7400, batch loss: 0.005752493865626846
175
+ [2026-04-16 12:11:52,535][trainer][INFO] - step: 7500, batch loss: 0.00622657166860751
176
+ [2026-04-16 12:12:21,348][trainer][INFO] - step: 7600, batch loss: 0.005901896766494691
177
+ [2026-04-16 12:12:48,833][trainer][INFO] - step: 7700, batch loss: 0.006314166031512643
178
+ [2026-04-16 12:13:16,999][trainer][INFO] - step: 7800, batch loss: 0.005745719468274202
179
+ [2026-04-16 12:13:44,548][trainer][INFO] - step: 7900, batch loss: 0.0059718955279174
180
+ [2026-04-16 12:14:12,829][trainer][INFO] - step: 8000, batch loss: 0.005816076011283907
181
+ [2026-04-16 12:14:40,414][trainer][INFO] - step: 8100, batch loss: 0.005677689694884452
182
+ [2026-04-16 12:14:50,531][trainer][INFO] - epoch: 8, loss=6.064758144511986 .
183
+ [2026-04-16 12:14:50,531][trainer][INFO] - epoch 9/30 ====================================================================================================
184
+ [2026-04-16 12:15:08,585][trainer][INFO] - step: 8200, batch loss: 0.0052527969340894045
185
+ [2026-04-16 12:15:37,736][trainer][INFO] - step: 8300, batch loss: 0.005221572600016492
186
+ [2026-04-16 12:16:07,265][trainer][INFO] - step: 8400, batch loss: 0.005388583083517337
187
+ [2026-04-16 12:16:36,167][trainer][INFO] - step: 8500, batch loss: 0.00546124652781845
188
+ [2026-04-16 12:17:03,768][trainer][INFO] - step: 8600, batch loss: 0.0050852070422046335
189
+ [2026-04-16 12:17:31,299][trainer][INFO] - step: 8700, batch loss: 0.005416722316862699
190
+ [2026-04-16 12:17:59,042][trainer][INFO] - step: 8800, batch loss: 0.005446841319548537
191
+ [2026-04-16 12:18:27,450][trainer][INFO] - step: 8900, batch loss: 0.0052915976807138745
192
+ [2026-04-16 12:18:55,058][trainer][INFO] - step: 9000, batch loss: 0.005285817508796772
193
+ [2026-04-16 12:19:22,322][trainer][INFO] - step: 9100, batch loss: 0.005201099921331648
194
+ [2026-04-16 12:19:36,402][trainer][INFO] - epoch: 9, loss=5.358912876333564 .
195
+ [2026-04-16 12:19:36,402][trainer][INFO] - epoch 10/30 ====================================================================================================
196
+ [2026-04-16 12:19:49,845][trainer][INFO] - step: 9200, batch loss: 0.0048850688126760335
197
+ [2026-04-16 12:20:19,120][trainer][INFO] - step: 9300, batch loss: 0.004662054820596329
198
+ [2026-04-16 12:20:46,877][trainer][INFO] - step: 9400, batch loss: 0.004582623031008094
199
+ [2026-04-16 12:21:15,542][trainer][INFO] - step: 9500, batch loss: 0.00453476719328621
200
+ [2026-04-16 12:21:42,962][trainer][INFO] - step: 9600, batch loss: 0.004831115418722085
201
+ [2026-04-16 12:22:10,603][trainer][INFO] - step: 9700, batch loss: 0.004727381135881814
202
+ [2026-04-16 12:22:38,375][trainer][INFO] - step: 9800, batch loss: 0.004616486235481662
203
+ [2026-04-16 12:23:06,246][trainer][INFO] - step: 9900, batch loss: 0.004763632137494476
204
+ [2026-04-16 12:23:34,430][trainer][INFO] - step: 10000, batch loss: 0.004307499012140327
205
+ [2026-04-16 12:24:02,819][trainer][INFO] - step: 10100, batch loss: 0.004360870717091268
206
+ [2026-04-16 12:24:22,258][trainer][INFO] - epoch: 10, loss=4.6841030612525865 .
207
+ [2026-04-16 12:24:22,259][trainer][INFO] - epoch 11/30 ====================================================================================================
208
+ [2026-04-16 12:24:30,866][trainer][INFO] - step: 10200, batch loss: 0.004558738150148841
209
+ [2026-04-16 12:24:58,486][trainer][INFO] - step: 10300, batch loss: 0.0038788321300182384
210
+ [2026-04-16 12:25:26,204][trainer][INFO] - step: 10400, batch loss: 0.004316426535096676
211
+ [2026-04-16 12:25:54,567][trainer][INFO] - step: 10500, batch loss: 0.004219791887116666
212
+ [2026-04-16 12:26:23,067][trainer][INFO] - step: 10600, batch loss: 0.0043255675376139245
213
+ [2026-04-16 12:26:50,383][trainer][INFO] - step: 10700, batch loss: 0.0037705287373310248
214
+ [2026-04-16 12:27:18,701][trainer][INFO] - step: 10800, batch loss: 0.003691437426028097
215
+ [2026-04-16 12:27:47,704][trainer][INFO] - step: 10900, batch loss: 0.004407755648677198
216
+ [2026-04-16 12:28:15,039][trainer][INFO] - step: 11000, batch loss: 0.004628108930216442
217
+ [2026-04-16 12:28:43,569][trainer][INFO] - step: 11100, batch loss: 0.00403286316948356
218
+ [2026-04-16 12:29:09,379][trainer][INFO] - epoch: 11, loss=4.236317333201349 .
219
+ [2026-04-16 12:29:09,380][trainer][INFO] - epoch 12/30 ====================================================================================================
220
+ [2026-04-16 12:29:13,034][trainer][INFO] - step: 11200, batch loss: 0.00425695829016126
221
+ [2026-04-16 12:29:41,746][trainer][INFO] - step: 11300, batch loss: 0.003620378879285937
222
+ [2026-04-16 12:30:08,779][trainer][INFO] - step: 11400, batch loss: 0.0038416854120986105
223
+ [2026-04-16 12:30:37,499][trainer][INFO] - step: 11500, batch loss: 0.0038706240336318348
224
+ [2026-04-16 12:31:04,808][trainer][INFO] - step: 11600, batch loss: 0.0037893279249211153
225
+ [2026-04-16 12:31:33,165][trainer][INFO] - step: 11700, batch loss: 0.0037928532021543403
226
+ [2026-04-16 12:32:01,194][trainer][INFO] - step: 11800, batch loss: 0.0033946389710242035
227
+ [2026-04-16 12:32:28,462][trainer][INFO] - step: 11900, batch loss: 0.0033888662810403164
228
+ [2026-04-16 12:32:56,085][trainer][INFO] - step: 12000, batch loss: 0.003881935564982511
229
+ [2026-04-16 12:33:24,482][trainer][INFO] - step: 12100, batch loss: 0.0038536999676220774
230
+ [2026-04-16 12:33:53,269][trainer][INFO] - step: 12200, batch loss: 0.0034482428594833523
231
+ [2026-04-16 12:33:54,328][trainer][INFO] - epoch: 12, loss=3.7551927965674743 .
232
+ [2026-04-16 12:33:54,328][trainer][INFO] - epoch 13/30 ====================================================================================================
233
+ [2026-04-16 12:34:21,083][trainer][INFO] - step: 12300, batch loss: 0.0032137125137372295
234
+ [2026-04-16 12:34:49,536][trainer][INFO] - step: 12400, batch loss: 0.003414331361550378
235
+ [2026-04-16 12:35:17,832][trainer][INFO] - step: 12500, batch loss: 0.00336960159916176
236
+ [2026-04-16 12:35:46,218][trainer][INFO] - step: 12600, batch loss: 0.002982874007172602
237
+ [2026-04-16 12:36:14,226][trainer][INFO] - step: 12700, batch loss: 0.003320795512622638
238
+ [2026-04-16 12:36:42,877][trainer][INFO] - step: 12800, batch loss: 0.003334473202608793
239
+ [2026-04-16 12:37:11,416][trainer][INFO] - step: 12900, batch loss: 0.003207056526424326
240
+ [2026-04-16 12:37:39,637][trainer][INFO] - step: 13000, batch loss: 0.0033718823354620623
241
+ [2026-04-16 12:38:06,886][trainer][INFO] - step: 13100, batch loss: 0.0033878284576790325
242
+ [2026-04-16 12:38:34,635][trainer][INFO] - step: 13200, batch loss: 0.0036032341676812533
243
+ [2026-04-16 12:38:40,638][trainer][INFO] - epoch: 13, loss=3.387888896616788 .
244
+ [2026-04-16 12:38:40,638][trainer][INFO] - epoch 14/30 ====================================================================================================
245
+ [2026-04-16 12:39:02,024][trainer][INFO] - step: 13300, batch loss: 0.0032738539911636583
246
+ [2026-04-16 12:39:31,034][trainer][INFO] - step: 13400, batch loss: 0.003105927012780422
247
+ [2026-04-16 12:39:58,642][trainer][INFO] - step: 13500, batch loss: 0.002920743068900013
248
+ [2026-04-16 12:40:26,180][trainer][INFO] - step: 13600, batch loss: 0.00319378525104619
249
+ [2026-04-16 12:40:54,007][trainer][INFO] - step: 13700, batch loss: 0.0034836822066549424
250
+ [2026-04-16 12:41:21,637][trainer][INFO] - step: 13800, batch loss: 0.0031375463833607413
251
+ [2026-04-16 12:41:50,227][trainer][INFO] - step: 13900, batch loss: 0.00298911757929088
252
+ [2026-04-16 12:42:19,874][trainer][INFO] - step: 14000, batch loss: 0.0028512716249663606
253
+ [2026-04-16 12:42:47,549][trainer][INFO] - step: 14100, batch loss: 0.003044298499846189
254
+ [2026-04-16 12:43:16,766][trainer][INFO] - step: 14200, batch loss: 0.003158952667977941
255
+ [2026-04-16 12:43:27,733][trainer][INFO] - epoch: 14, loss=3.1393872521262223 .
256
+ [2026-04-16 12:43:27,733][trainer][INFO] - epoch 15/30 ====================================================================================================
257
+ [2026-04-16 12:43:45,094][trainer][INFO] - step: 14300, batch loss: 0.0029099242453157355
258
+ [2026-04-16 12:44:13,278][trainer][INFO] - step: 14400, batch loss: 0.002729736065685065
259
+ [2026-04-16 12:44:42,121][trainer][INFO] - step: 14500, batch loss: 0.002790610685053156
260
+ [2026-04-16 12:45:11,275][trainer][INFO] - step: 14600, batch loss: 0.002918851664491029
261
+ [2026-04-16 12:45:39,335][trainer][INFO] - step: 14700, batch loss: 0.002843388581868006
262
+ [2026-04-16 12:46:08,447][trainer][INFO] - step: 14800, batch loss: 0.0029374525214382104
263
+ [2026-04-16 12:46:36,885][trainer][INFO] - step: 14900, batch loss: 0.002572607961164028
264
+ [2026-04-16 12:47:04,772][trainer][INFO] - step: 15000, batch loss: 0.0030270419171856532
265
+ [2026-04-16 12:47:32,481][trainer][INFO] - step: 15100, batch loss: 0.0025315387855101056
266
+ [2026-04-16 12:47:59,553][trainer][INFO] - step: 15200, batch loss: 0.002772726206468118
267
+ [2026-04-16 12:48:15,105][trainer][INFO] - epoch: 15, loss=2.864250061225439 .
268
+ [2026-04-16 12:48:15,105][trainer][INFO] - epoch 16/30 ====================================================================================================
269
+ [2026-04-16 12:48:27,586][trainer][INFO] - step: 15300, batch loss: 0.0026817279519553944
270
+ [2026-04-16 12:48:55,662][trainer][INFO] - step: 15400, batch loss: 0.0028904729455053674
271
+ [2026-04-16 12:49:23,920][trainer][INFO] - step: 15500, batch loss: 0.0024433053788965333
272
+ [2026-04-16 12:49:52,620][trainer][INFO] - step: 15600, batch loss: 0.002546149720766183
273
+ [2026-04-16 12:50:20,102][trainer][INFO] - step: 15700, batch loss: 0.0027873495440495196
274
+ [2026-04-16 12:50:47,742][trainer][INFO] - step: 15800, batch loss: 0.0026471394108989505
275
+ [2026-04-16 12:51:17,223][trainer][INFO] - step: 15900, batch loss: 0.0025099273080376834
276
+ [2026-04-16 12:51:45,422][trainer][INFO] - step: 16000, batch loss: 0.0027630068418019156
277
+ [2026-04-16 12:52:14,294][trainer][INFO] - step: 16100, batch loss: 0.002566703433731202
278
+ [2026-04-16 12:52:42,156][trainer][INFO] - step: 16200, batch loss: 0.0025821640793670135
279
+ [2026-04-16 12:53:02,552][trainer][INFO] - epoch: 16, loss=2.678656037314361 .
280
+ [2026-04-16 12:53:02,553][trainer][INFO] - epoch 17/30 ====================================================================================================
281
+ [2026-04-16 12:53:09,948][trainer][INFO] - step: 16300, batch loss: 0.0027434779168308086
282
+ [2026-04-16 12:53:38,480][trainer][INFO] - step: 16400, batch loss: 0.0024423383224517576
283
+ [2026-04-16 12:54:07,174][trainer][INFO] - step: 16500, batch loss: 0.002616920436043859
284
+ [2026-04-16 12:54:35,165][trainer][INFO] - step: 16600, batch loss: 0.002767015770112725
285
+ [2026-04-16 12:55:04,199][trainer][INFO] - step: 16700, batch loss: 0.0024118812957647103
286
+ [2026-04-16 12:55:33,232][trainer][INFO] - step: 16800, batch loss: 0.0025690640781465463
287
+ [2026-04-16 12:56:01,941][trainer][INFO] - step: 16900, batch loss: 0.0024862230091483805
288
+ [2026-04-16 12:56:30,794][trainer][INFO] - step: 17000, batch loss: 0.002387646912743262
289
+ [2026-04-16 12:56:59,564][trainer][INFO] - step: 17100, batch loss: 0.0025128672178466177
290
+ [2026-04-16 12:57:27,126][trainer][INFO] - step: 17200, batch loss: 0.0022417695855722636
291
+ [2026-04-16 12:57:52,425][trainer][INFO] - epoch: 17, loss=2.5205702751160404 .
292
+ [2026-04-16 12:57:52,425][trainer][INFO] - epoch 18/30 ====================================================================================================
293
+ [2026-04-16 12:57:55,964][trainer][INFO] - step: 17300, batch loss: 0.002421379157073753
294
+ [2026-04-16 12:58:23,301][trainer][INFO] - step: 17400, batch loss: 0.0023909204848946985
295
+ [2026-04-16 12:58:51,840][trainer][INFO] - step: 17500, batch loss: 0.0024254048386660675
296
+ [2026-04-16 12:59:18,968][trainer][INFO] - step: 17600, batch loss: 0.002241210455042669
297
+ [2026-04-16 12:59:48,091][trainer][INFO] - step: 17700, batch loss: 0.00209224737248592
298
+ [2026-04-16 13:00:16,137][trainer][INFO] - step: 17800, batch loss: 0.002401935899829873
299
+ [2026-04-16 13:00:45,239][trainer][INFO] - step: 17900, batch loss: 0.002368715156284538
300
+ [2026-04-16 13:01:13,121][trainer][INFO] - step: 18000, batch loss: 0.0024371431917691285
301
+ [2026-04-16 13:01:41,955][trainer][INFO] - step: 18100, batch loss: 0.0021528245797330604
302
+ [2026-04-16 13:02:10,431][trainer][INFO] - step: 18200, batch loss: 0.002323571648299183
303
+ [2026-04-16 13:02:38,126][trainer][INFO] - step: 18300, batch loss: 0.0022393950587594833
304
+ [2026-04-16 13:02:40,163][trainer][INFO] - epoch: 18, loss=2.355786796982774 .
305
+ [2026-04-16 13:02:40,163][trainer][INFO] - epoch 19/30 ====================================================================================================
306
+ [2026-04-16 13:03:05,752][trainer][INFO] - step: 18400, batch loss: 0.002219963581215771
307
+ [2026-04-16 13:03:34,620][trainer][INFO] - step: 18500, batch loss: 0.0018696615507874412
308
+ [2026-04-16 13:04:03,474][trainer][INFO] - step: 18600, batch loss: 0.0019033543081467643
309
+ [2026-04-16 13:04:31,599][trainer][INFO] - step: 18700, batch loss: 0.0021161602678558696
310
+ [2026-04-16 13:04:59,869][trainer][INFO] - step: 18800, batch loss: 0.001997172368420452
311
+ [2026-04-16 13:05:28,149][trainer][INFO] - step: 18900, batch loss: 0.0022058641285584778
312
+ [2026-04-16 13:05:56,658][trainer][INFO] - step: 19000, batch loss: 0.002269889635464539
313
+ [2026-04-16 13:06:24,136][trainer][INFO] - step: 19100, batch loss: 0.0020763865953965063
314
+ [2026-04-16 13:06:52,688][trainer][INFO] - step: 19200, batch loss: 0.0021475402238218984
315
+ [2026-04-16 13:07:20,470][trainer][INFO] - step: 19300, batch loss: 0.002360360966073781
316
+ [2026-04-16 13:07:27,230][trainer][INFO] - epoch: 19, loss=2.146043473818992 .
317
+ [2026-04-16 13:07:27,230][trainer][INFO] - epoch 20/30 ====================================================================================================
318
+ [2026-04-16 13:07:50,017][trainer][INFO] - step: 19400, batch loss: 0.001961981183797819
319
+ [2026-04-16 13:08:17,801][trainer][INFO] - step: 19500, batch loss: 0.0019488741380568988
320
+ [2026-04-16 13:08:46,263][trainer][INFO] - step: 19600, batch loss: 0.0019090044843522807
321
+ [2026-04-16 13:09:15,567][trainer][INFO] - step: 19700, batch loss: 0.002340233057591772
322
+ [2026-04-16 13:09:43,529][trainer][INFO] - step: 19800, batch loss: 0.002077641087458179
323
+ [2026-04-16 13:10:10,794][trainer][INFO] - step: 19900, batch loss: 0.002163110761674728
324
+ [2026-04-16 13:10:39,648][trainer][INFO] - step: 20000, batch loss: 0.001775486892803784
325
+ [2026-04-16 13:11:07,519][trainer][INFO] - step: 20100, batch loss: 0.0019226342059011664
326
+ [2026-04-16 13:11:35,879][trainer][INFO] - step: 20200, batch loss: 0.0023017652530302217
327
+ [2026-04-16 13:12:04,012][trainer][INFO] - step: 20300, batch loss: 0.0021294051857368994
328
+ [2026-04-16 13:12:15,351][trainer][INFO] - epoch: 20, loss=2.083137390276559 .
329
+ [2026-04-16 13:12:15,351][trainer][INFO] - epoch 21/30 ====================================================================================================
330
+ [2026-04-16 13:12:31,928][trainer][INFO] - step: 20400, batch loss: 0.001951870011393638
331
+ [2026-04-16 13:13:00,342][trainer][INFO] - step: 20500, batch loss: 0.0020286565882502296
332
+ [2026-04-16 13:13:28,099][trainer][INFO] - step: 20600, batch loss: 0.0018413494807879363
333
+ [2026-04-16 13:13:55,628][trainer][INFO] - step: 20700, batch loss: 0.0021078637093467816
334
+ [2026-04-16 13:14:23,636][trainer][INFO] - step: 20800, batch loss: 0.0018862042788233286
335
+ [2026-04-16 13:14:51,400][trainer][INFO] - step: 20900, batch loss: 0.001978682629401495
336
+ [2026-04-16 13:15:19,446][trainer][INFO] - step: 21000, batch loss: 0.0017329989096811385
337
+ [2026-04-16 13:15:48,447][trainer][INFO] - step: 21100, batch loss: 0.0020449507752006288
338
+ [2026-04-16 13:16:18,280][trainer][INFO] - step: 21200, batch loss: 0.0021740554559772298
339
+ [2026-04-16 13:16:47,802][trainer][INFO] - step: 21300, batch loss: 0.0016781868489209465
340
+ [2026-04-16 13:17:04,775][trainer][INFO] - epoch: 21, loss=1.9708028829450803 .
341
+ [2026-04-16 13:17:04,775][trainer][INFO] - epoch 22/30 ====================================================================================================
342
+ [2026-04-16 13:17:16,186][trainer][INFO] - step: 21400, batch loss: 0.0018367681823929843
343
+ [2026-04-16 13:17:44,795][trainer][INFO] - step: 21500, batch loss: 0.0019972127183728494
344
+ [2026-04-16 13:18:13,695][trainer][INFO] - step: 21600, batch loss: 0.0021870278152413686
345
+ [2026-04-16 13:18:42,216][trainer][INFO] - step: 21700, batch loss: 0.0019397866133346055
346
+ [2026-04-16 13:19:10,280][trainer][INFO] - step: 21800, batch loss: 0.001541316848569508
347
+ [2026-04-16 13:19:38,870][trainer][INFO] - step: 21900, batch loss: 0.001696447314300431
348
+ [2026-04-16 13:20:06,601][trainer][INFO] - step: 22000, batch loss: 0.0018095931130988274
349
+ [2026-04-16 13:20:34,378][trainer][INFO] - step: 22100, batch loss: 0.0017913717386478236
350
+ [2026-04-16 13:21:03,300][trainer][INFO] - step: 22200, batch loss: 0.0020080075325978834
351
+ [2026-04-16 13:21:32,359][trainer][INFO] - step: 22300, batch loss: 0.0016641279931326855
352
+ [2026-04-16 13:21:52,577][trainer][INFO] - epoch: 22, loss=1.8975751850201992 .
353
+ [2026-04-16 13:21:52,577][trainer][INFO] - epoch 23/30 ====================================================================================================
354
+ [2026-04-16 13:22:00,081][trainer][INFO] - step: 22400, batch loss: 0.0019321582445929325
355
+ [2026-04-16 13:22:28,635][trainer][INFO] - step: 22500, batch loss: 0.0017830777999615812
356
+ [2026-04-16 13:22:57,180][trainer][INFO] - step: 22600, batch loss: 0.0015882500955075231
357
+ [2026-04-16 13:23:25,492][trainer][INFO] - step: 22700, batch loss: 0.0018369562826873978
358
+ [2026-04-16 13:23:53,710][trainer][INFO] - step: 22800, batch loss: 0.001600465566474544
359
+ [2026-04-16 13:24:23,312][trainer][INFO] - step: 22900, batch loss: 0.001816364677410936
360
+ [2026-04-16 13:24:50,710][trainer][INFO] - step: 23000, batch loss: 0.001727650601465811
361
+ [2026-04-16 13:25:19,323][trainer][INFO] - step: 23100, batch loss: 0.0018357900294103042
362
+ [2026-04-16 13:25:47,164][trainer][INFO] - step: 23200, batch loss: 0.0016840357080021085
363
+ [2026-04-16 13:26:15,642][trainer][INFO] - step: 23300, batch loss: 0.0017996952061494196
364
+ [2026-04-16 13:26:40,377][trainer][INFO] - epoch: 23, loss=1.776736755123287 .
365
+ [2026-04-16 13:26:40,377][trainer][INFO] - epoch 24/30 ====================================================================================================
366
+ [2026-04-16 13:26:42,900][trainer][INFO] - step: 23400, batch loss: 0.0018819219776265797
367
+ [2026-04-16 13:27:10,705][trainer][INFO] - step: 23500, batch loss: 0.0018132978210692716
368
+ [2026-04-16 13:27:38,613][trainer][INFO] - step: 23600, batch loss: 0.001503919148924308
369
+ [2026-04-16 13:28:07,115][trainer][INFO] - step: 23700, batch loss: 0.0017438039147652058
370
+ [2026-04-16 13:28:36,789][trainer][INFO] - step: 23800, batch loss: 0.001839133310394442
371
+ [2026-04-16 13:29:05,209][trainer][INFO] - step: 23900, batch loss: 0.0017078353521901377
372
+ [2026-04-16 13:29:33,298][trainer][INFO] - step: 24000, batch loss: 0.0016456692155035865
373
+ [2026-04-16 13:29:59,944][trainer][INFO] - step: 24100, batch loss: 0.0016331123365545762
374
+ [2026-04-16 13:30:28,223][trainer][INFO] - step: 24200, batch loss: 0.0015735684771110326
375
+ [2026-04-16 13:30:56,459][trainer][INFO] - step: 24300, batch loss: 0.0016113364352728144
376
+ [2026-04-16 13:31:25,886][trainer][INFO] - step: 24400, batch loss: 0.0017065834537295383
377
+ [2026-04-16 13:31:28,199][trainer][INFO] - epoch: 24, loss=1.709218715279162 .
378
+ [2026-04-16 13:31:28,200][trainer][INFO] - epoch 25/30 ====================================================================================================
379
+ [2026-04-16 13:31:53,567][trainer][INFO] - step: 24500, batch loss: 0.001562461601380471
380
+ [2026-04-16 13:32:22,682][trainer][INFO] - step: 24600, batch loss: 0.0015762986043305024
381
+ [2026-04-16 13:32:50,770][trainer][INFO] - step: 24700, batch loss: 0.0016680720107341584
382
+ [2026-04-16 13:33:19,937][trainer][INFO] - step: 24800, batch loss: 0.001567825402190578
383
+ [2026-04-16 13:33:48,558][trainer][INFO] - step: 24900, batch loss: 0.0015669424517056891
384
+ [2026-04-16 13:34:16,231][trainer][INFO] - step: 25000, batch loss: 0.0016323063785341882
385
+ [2026-04-16 13:34:44,579][trainer][INFO] - step: 25100, batch loss: 0.001575590355658057
386
+ [2026-04-16 13:35:11,993][trainer][INFO] - step: 25200, batch loss: 0.0016560732517946946
387
+ [2026-04-16 13:35:40,305][trainer][INFO] - step: 25300, batch loss: 0.0017458187306147641
388
+ [2026-04-16 13:36:09,580][trainer][INFO] - step: 25400, batch loss: 0.0016586954741193722
389
+ [2026-04-16 13:36:16,907][trainer][INFO] - epoch: 25, loss=1.6429951496106299 .
390
+ [2026-04-16 13:36:16,907][trainer][INFO] - epoch 26/30 ====================================================================================================
391
+ [2026-04-16 13:36:38,498][trainer][INFO] - step: 25500, batch loss: 0.0015957185101252102
392
+ [2026-04-16 13:37:05,945][trainer][INFO] - step: 25600, batch loss: 0.0015428501537910477
393
+ [2026-04-16 13:37:34,777][trainer][INFO] - step: 25700, batch loss: 0.0015625915928823184
394
+ [2026-04-16 13:38:02,613][trainer][INFO] - step: 25800, batch loss: 0.00150969700639935
395
+ [2026-04-16 13:38:31,489][trainer][INFO] - step: 25900, batch loss: 0.0014703967567863672
396
+ [2026-04-16 13:38:59,194][trainer][INFO] - step: 26000, batch loss: 0.001548027046171977
397
+ [2026-04-16 13:39:26,781][trainer][INFO] - step: 26100, batch loss: 0.0013223594415649785
398
+ [2026-04-16 13:39:53,783][trainer][INFO] - step: 26200, batch loss: 0.0015450125840879235
399
+ [2026-04-16 13:40:21,763][trainer][INFO] - step: 26300, batch loss: 0.0017352549924605498
400
+ [2026-04-16 13:40:48,573][trainer][INFO] - step: 26400, batch loss: 0.001641774173918348
401
+ [2026-04-16 13:41:00,233][trainer][INFO] - epoch: 26, loss=1.577395426137004 .
402
+ [2026-04-16 13:41:00,234][trainer][INFO] - epoch 27/30 ====================================================================================================
403
+ [2026-04-16 13:41:16,143][trainer][INFO] - step: 26500, batch loss: 0.0014178042029794405
404
+ [2026-04-16 13:41:44,274][trainer][INFO] - step: 26600, batch loss: 0.0014566411440965548
405
+ [2026-04-16 13:42:12,178][trainer][INFO] - step: 26700, batch loss: 0.0015253939264680367
406
+ [2026-04-16 13:42:39,040][trainer][INFO] - step: 26800, batch loss: 0.0012138379613354136
407
+ [2026-04-16 13:43:06,366][trainer][INFO] - step: 26900, batch loss: 0.0018657652148001685
408
+ [2026-04-16 13:43:32,566][trainer][INFO] - step: 27000, batch loss: 0.0015860109423216876
409
+ [2026-04-16 13:43:59,297][trainer][INFO] - step: 27100, batch loss: 0.0016634922683699139
410
+ [2026-04-16 13:44:26,691][trainer][INFO] - step: 27200, batch loss: 0.0012321010620718225
411
+ [2026-04-16 13:44:54,433][trainer][INFO] - step: 27300, batch loss: 0.0017192253651209996
412
+ [2026-04-16 13:45:21,746][trainer][INFO] - step: 27400, batch loss: 0.0013890920564133488
413
+ [2026-04-16 13:45:37,708][trainer][INFO] - epoch: 27, loss=1.52693572255509 .
414
+ [2026-04-16 13:45:37,709][trainer][INFO] - epoch 28/30 ====================================================================================================
415
+ [2026-04-16 13:45:49,028][trainer][INFO] - step: 27500, batch loss: 0.0013282260757696393
416
+ [2026-04-16 13:46:16,101][trainer][INFO] - step: 27600, batch loss: 0.0016133229770784472
417
+ [2026-04-16 13:46:43,306][trainer][INFO] - step: 27700, batch loss: 0.0016010305725081943
418
+ [2026-04-16 13:47:10,032][trainer][INFO] - step: 27800, batch loss: 0.0013438603023247409
419
+ [2026-04-16 13:47:37,349][trainer][INFO] - step: 27900, batch loss: 0.0012717020705763319
420
+ [2026-04-16 13:48:04,776][trainer][INFO] - step: 28000, batch loss: 0.0012529942276512834
421
+ [2026-04-16 13:48:32,252][trainer][INFO] - step: 28100, batch loss: 0.001682373622607305
422
+ [2026-04-16 13:49:00,043][trainer][INFO] - step: 28200, batch loss: 0.0015611862711005149
423
+ [2026-04-16 13:49:26,101][trainer][INFO] - step: 28300, batch loss: 0.0015575882056425344
424
+ [2026-04-16 13:49:53,722][trainer][INFO] - step: 28400, batch loss: 0.0014441255623003644
425
+ [2026-04-16 13:50:13,997][trainer][INFO] - epoch: 28, loss=1.4890802098536768 .
426
+ [2026-04-16 13:50:13,998][trainer][INFO] - epoch 29/30 ====================================================================================================
427
+ [2026-04-16 13:50:20,292][trainer][INFO] - step: 28500, batch loss: 0.00130912150250459
428
+ [2026-04-16 13:50:47,631][trainer][INFO] - step: 28600, batch loss: 0.0014248949373997421
429
+ [2026-04-16 13:51:15,391][trainer][INFO] - step: 28700, batch loss: 0.0014834315834324743
430
+ [2026-04-16 13:51:42,076][trainer][INFO] - step: 28800, batch loss: 0.0012627411501471405
431
+ [2026-04-16 13:52:08,884][trainer][INFO] - step: 28900, batch loss: 0.0017058910431501544
432
+ [2026-04-16 13:52:36,417][trainer][INFO] - step: 29000, batch loss: 0.0014680034484676292
433
+ [2026-04-16 13:53:04,348][trainer][INFO] - step: 29100, batch loss: 0.0015802718792215105
434
+ [2026-04-16 13:53:31,122][trainer][INFO] - step: 29200, batch loss: 0.0014035104402983995
435
+ [2026-04-16 13:53:58,181][trainer][INFO] - step: 29300, batch loss: 0.0014280556279115664
436
+ [2026-04-16 13:54:25,086][trainer][INFO] - step: 29400, batch loss: 0.0013249082937964068
437
+ [2026-04-16 13:54:51,315][trainer][INFO] - epoch: 29, loss=1.448775170778226 .
438
+ [2026-04-16 13:54:51,315][trainer][INFO] - epoch 30/30 ====================================================================================================
439
+ [2026-04-16 13:54:53,650][trainer][INFO] - step: 29500, batch loss: 0.001292500724798964
440
+ [2026-04-16 13:55:21,049][trainer][INFO] - step: 29600, batch loss: 0.0015078988298161203
441
+ [2026-04-16 13:55:48,924][trainer][INFO] - step: 29700, batch loss: 0.0013672888715987538
442
+ [2026-04-16 13:56:17,073][trainer][INFO] - step: 29800, batch loss: 0.001459387114330153
443
+ [2026-04-16 13:56:44,143][trainer][INFO] - step: 29900, batch loss: 0.0014157829589697614
444
+ [2026-04-16 13:57:11,178][trainer][INFO] - step: 30000, batch loss: 0.0014095398042151027
445
+ [2026-04-16 13:57:37,722][trainer][INFO] - step: 30100, batch loss: 0.0012149674265985657
446
+ [2026-04-16 13:58:03,959][trainer][INFO] - step: 30200, batch loss: 0.0013031805306945874
447
+ [2026-04-16 13:58:30,958][trainer][INFO] - step: 30300, batch loss: 0.0013667242398454614
448
+ [2026-04-16 13:58:58,329][trainer][INFO] - step: 30400, batch loss: 0.0013467074709867148
449
+ [2026-04-16 13:59:26,392][trainer][INFO] - step: 30500, batch loss: 0.0016136745981880955
450
+ [2026-04-16 13:59:29,368][trainer][INFO] - epoch: 30, loss=1.423759139578092 .
451
+ [2026-04-16 13:59:30,863][trainer][INFO] - Best dev result: 0
452
+ [2026-04-16 14:00:07,924][trainer][INFO] - Test result: {'test_F1': 51.159889274435656, 'test_F1_ign': 50.81235543694389, 'test_precision': 80.00243872698451, 'test_recall': 37.603163686382395} .