File size: 25,371 Bytes
d8a1523
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
2023-10-11 16:06:44,778 ----------------------------------------------------------------------------------------------------
2023-10-11 16:06:44,780 Model: "SequenceTagger(
  (embeddings): ByT5Embeddings(
    (model): T5EncoderModel(
      (shared): Embedding(384, 1472)
      (encoder): T5Stack(
        (embed_tokens): Embedding(384, 1472)
        (block): ModuleList(
          (0): T5Block(
            (layer): ModuleList(
              (0): T5LayerSelfAttention(
                (SelfAttention): T5Attention(
                  (q): Linear(in_features=1472, out_features=384, bias=False)
                  (k): Linear(in_features=1472, out_features=384, bias=False)
                  (v): Linear(in_features=1472, out_features=384, bias=False)
                  (o): Linear(in_features=384, out_features=1472, bias=False)
                  (relative_attention_bias): Embedding(32, 6)
                )
                (layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
                (dropout): Dropout(p=0.1, inplace=False)
              )
              (1): T5LayerFF(
                (DenseReluDense): T5DenseGatedActDense(
                  (wi_0): Linear(in_features=1472, out_features=3584, bias=False)
                  (wi_1): Linear(in_features=1472, out_features=3584, bias=False)
                  (wo): Linear(in_features=3584, out_features=1472, bias=False)
                  (dropout): Dropout(p=0.1, inplace=False)
                  (act): NewGELUActivation()
                )
                (layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
                (dropout): Dropout(p=0.1, inplace=False)
              )
            )
          )
          (1-11): 11 x T5Block(
            (layer): ModuleList(
              (0): T5LayerSelfAttention(
                (SelfAttention): T5Attention(
                  (q): Linear(in_features=1472, out_features=384, bias=False)
                  (k): Linear(in_features=1472, out_features=384, bias=False)
                  (v): Linear(in_features=1472, out_features=384, bias=False)
                  (o): Linear(in_features=384, out_features=1472, bias=False)
                )
                (layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
                (dropout): Dropout(p=0.1, inplace=False)
              )
              (1): T5LayerFF(
                (DenseReluDense): T5DenseGatedActDense(
                  (wi_0): Linear(in_features=1472, out_features=3584, bias=False)
                  (wi_1): Linear(in_features=1472, out_features=3584, bias=False)
                  (wo): Linear(in_features=3584, out_features=1472, bias=False)
                  (dropout): Dropout(p=0.1, inplace=False)
                  (act): NewGELUActivation()
                )
                (layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
                (dropout): Dropout(p=0.1, inplace=False)
              )
            )
          )
        )
        (final_layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
        (dropout): Dropout(p=0.1, inplace=False)
      )
    )
  )
  (locked_dropout): LockedDropout(p=0.5)
  (linear): Linear(in_features=1472, out_features=17, bias=True)
  (loss_function): CrossEntropyLoss()
)"
2023-10-11 16:06:44,780 ----------------------------------------------------------------------------------------------------
2023-10-11 16:06:44,781 MultiCorpus: 7142 train + 698 dev + 2570 test sentences
 - NER_HIPE_2022 Corpus: 7142 train + 698 dev + 2570 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/fr/with_doc_seperator
2023-10-11 16:06:44,781 ----------------------------------------------------------------------------------------------------
2023-10-11 16:06:44,781 Train:  7142 sentences
2023-10-11 16:06:44,781         (train_with_dev=False, train_with_test=False)
2023-10-11 16:06:44,781 ----------------------------------------------------------------------------------------------------
2023-10-11 16:06:44,781 Training Params:
2023-10-11 16:06:44,781  - learning_rate: "0.00016" 
2023-10-11 16:06:44,781  - mini_batch_size: "8"
2023-10-11 16:06:44,781  - max_epochs: "10"
2023-10-11 16:06:44,781  - shuffle: "True"
2023-10-11 16:06:44,781 ----------------------------------------------------------------------------------------------------
2023-10-11 16:06:44,781 Plugins:
2023-10-11 16:06:44,782  - TensorboardLogger
2023-10-11 16:06:44,782  - LinearScheduler | warmup_fraction: '0.1'
2023-10-11 16:06:44,782 ----------------------------------------------------------------------------------------------------
2023-10-11 16:06:44,782 Final evaluation on model from best epoch (best-model.pt)
2023-10-11 16:06:44,782  - metric: "('micro avg', 'f1-score')"
2023-10-11 16:06:44,782 ----------------------------------------------------------------------------------------------------
2023-10-11 16:06:44,782 Computation:
2023-10-11 16:06:44,782  - compute on device: cuda:0
2023-10-11 16:06:44,782  - embedding storage: none
2023-10-11 16:06:44,782 ----------------------------------------------------------------------------------------------------
2023-10-11 16:06:44,782 Model training base path: "hmbench-newseye/fr-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs8-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-4"
2023-10-11 16:06:44,782 ----------------------------------------------------------------------------------------------------
2023-10-11 16:06:44,782 ----------------------------------------------------------------------------------------------------
2023-10-11 16:06:44,782 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-11 16:07:34,507 epoch 1 - iter 89/893 - loss 2.83150238 - time (sec): 49.72 - samples/sec: 469.42 - lr: 0.000016 - momentum: 0.000000
2023-10-11 16:08:24,856 epoch 1 - iter 178/893 - loss 2.73922034 - time (sec): 100.07 - samples/sec: 478.79 - lr: 0.000032 - momentum: 0.000000
2023-10-11 16:09:15,427 epoch 1 - iter 267/893 - loss 2.54002172 - time (sec): 150.64 - samples/sec: 478.44 - lr: 0.000048 - momentum: 0.000000
2023-10-11 16:10:07,265 epoch 1 - iter 356/893 - loss 2.31635472 - time (sec): 202.48 - samples/sec: 478.37 - lr: 0.000064 - momentum: 0.000000
2023-10-11 16:10:59,517 epoch 1 - iter 445/893 - loss 2.08548476 - time (sec): 254.73 - samples/sec: 472.32 - lr: 0.000080 - momentum: 0.000000
2023-10-11 16:11:54,697 epoch 1 - iter 534/893 - loss 1.85181034 - time (sec): 309.91 - samples/sec: 472.21 - lr: 0.000095 - momentum: 0.000000
2023-10-11 16:12:50,107 epoch 1 - iter 623/893 - loss 1.65665303 - time (sec): 365.32 - samples/sec: 473.67 - lr: 0.000111 - momentum: 0.000000
2023-10-11 16:13:44,323 epoch 1 - iter 712/893 - loss 1.50734839 - time (sec): 419.54 - samples/sec: 472.30 - lr: 0.000127 - momentum: 0.000000
2023-10-11 16:14:37,947 epoch 1 - iter 801/893 - loss 1.38574381 - time (sec): 473.16 - samples/sec: 469.95 - lr: 0.000143 - momentum: 0.000000
2023-10-11 16:15:34,092 epoch 1 - iter 890/893 - loss 1.27856971 - time (sec): 529.31 - samples/sec: 468.60 - lr: 0.000159 - momentum: 0.000000
2023-10-11 16:15:36,427 ----------------------------------------------------------------------------------------------------
2023-10-11 16:15:36,427 EPOCH 1 done: loss 1.2758 - lr: 0.000159
2023-10-11 16:15:58,475 DEV : loss 0.26407888531684875 - f1-score (micro avg)  0.2839
2023-10-11 16:15:58,513 saving best model
2023-10-11 16:15:59,455 ----------------------------------------------------------------------------------------------------
2023-10-11 16:16:53,512 epoch 2 - iter 89/893 - loss 0.30599282 - time (sec): 54.06 - samples/sec: 475.97 - lr: 0.000158 - momentum: 0.000000
2023-10-11 16:17:48,903 epoch 2 - iter 178/893 - loss 0.30164929 - time (sec): 109.45 - samples/sec: 465.09 - lr: 0.000156 - momentum: 0.000000
2023-10-11 16:18:45,107 epoch 2 - iter 267/893 - loss 0.27983412 - time (sec): 165.65 - samples/sec: 468.00 - lr: 0.000155 - momentum: 0.000000
2023-10-11 16:19:37,291 epoch 2 - iter 356/893 - loss 0.26596551 - time (sec): 217.83 - samples/sec: 462.78 - lr: 0.000153 - momentum: 0.000000
2023-10-11 16:20:32,275 epoch 2 - iter 445/893 - loss 0.24908676 - time (sec): 272.82 - samples/sec: 459.68 - lr: 0.000151 - momentum: 0.000000
2023-10-11 16:21:26,899 epoch 2 - iter 534/893 - loss 0.23391028 - time (sec): 327.44 - samples/sec: 457.37 - lr: 0.000149 - momentum: 0.000000
2023-10-11 16:22:17,982 epoch 2 - iter 623/893 - loss 0.22443918 - time (sec): 378.52 - samples/sec: 459.29 - lr: 0.000148 - momentum: 0.000000
2023-10-11 16:23:07,317 epoch 2 - iter 712/893 - loss 0.21432762 - time (sec): 427.86 - samples/sec: 463.46 - lr: 0.000146 - momentum: 0.000000
2023-10-11 16:23:54,170 epoch 2 - iter 801/893 - loss 0.20470001 - time (sec): 474.71 - samples/sec: 467.17 - lr: 0.000144 - momentum: 0.000000
2023-10-11 16:24:43,031 epoch 2 - iter 890/893 - loss 0.19693577 - time (sec): 523.57 - samples/sec: 472.97 - lr: 0.000142 - momentum: 0.000000
2023-10-11 16:24:44,752 ----------------------------------------------------------------------------------------------------
2023-10-11 16:24:44,752 EPOCH 2 done: loss 0.1964 - lr: 0.000142
2023-10-11 16:25:05,266 DEV : loss 0.10753299295902252 - f1-score (micro avg)  0.7415
2023-10-11 16:25:05,296 saving best model
2023-10-11 16:25:08,063 ----------------------------------------------------------------------------------------------------
2023-10-11 16:25:56,375 epoch 3 - iter 89/893 - loss 0.09369444 - time (sec): 48.31 - samples/sec: 513.41 - lr: 0.000140 - momentum: 0.000000
2023-10-11 16:26:45,041 epoch 3 - iter 178/893 - loss 0.09463543 - time (sec): 96.97 - samples/sec: 514.91 - lr: 0.000139 - momentum: 0.000000
2023-10-11 16:27:32,896 epoch 3 - iter 267/893 - loss 0.08742843 - time (sec): 144.83 - samples/sec: 511.93 - lr: 0.000137 - momentum: 0.000000
2023-10-11 16:28:21,237 epoch 3 - iter 356/893 - loss 0.08324272 - time (sec): 193.17 - samples/sec: 508.09 - lr: 0.000135 - momentum: 0.000000
2023-10-11 16:29:09,547 epoch 3 - iter 445/893 - loss 0.08475619 - time (sec): 241.48 - samples/sec: 508.37 - lr: 0.000133 - momentum: 0.000000
2023-10-11 16:29:59,366 epoch 3 - iter 534/893 - loss 0.08485703 - time (sec): 291.30 - samples/sec: 506.26 - lr: 0.000132 - momentum: 0.000000
2023-10-11 16:30:48,049 epoch 3 - iter 623/893 - loss 0.08297446 - time (sec): 339.98 - samples/sec: 505.75 - lr: 0.000130 - momentum: 0.000000
2023-10-11 16:31:36,824 epoch 3 - iter 712/893 - loss 0.08307353 - time (sec): 388.76 - samples/sec: 507.19 - lr: 0.000128 - momentum: 0.000000
2023-10-11 16:32:26,257 epoch 3 - iter 801/893 - loss 0.08179325 - time (sec): 438.19 - samples/sec: 507.93 - lr: 0.000126 - momentum: 0.000000
2023-10-11 16:33:14,792 epoch 3 - iter 890/893 - loss 0.08102671 - time (sec): 486.73 - samples/sec: 509.04 - lr: 0.000125 - momentum: 0.000000
2023-10-11 16:33:16,459 ----------------------------------------------------------------------------------------------------
2023-10-11 16:33:16,460 EPOCH 3 done: loss 0.0812 - lr: 0.000125
2023-10-11 16:33:38,017 DEV : loss 0.11330673843622208 - f1-score (micro avg)  0.7545
2023-10-11 16:33:38,056 saving best model
2023-10-11 16:33:40,713 ----------------------------------------------------------------------------------------------------
2023-10-11 16:34:31,047 epoch 4 - iter 89/893 - loss 0.05729569 - time (sec): 50.33 - samples/sec: 489.85 - lr: 0.000123 - momentum: 0.000000
2023-10-11 16:35:20,311 epoch 4 - iter 178/893 - loss 0.05027581 - time (sec): 99.59 - samples/sec: 498.88 - lr: 0.000121 - momentum: 0.000000
2023-10-11 16:36:10,722 epoch 4 - iter 267/893 - loss 0.05117128 - time (sec): 150.00 - samples/sec: 506.36 - lr: 0.000119 - momentum: 0.000000
2023-10-11 16:37:00,183 epoch 4 - iter 356/893 - loss 0.05260726 - time (sec): 199.47 - samples/sec: 504.60 - lr: 0.000117 - momentum: 0.000000
2023-10-11 16:37:48,423 epoch 4 - iter 445/893 - loss 0.05204046 - time (sec): 247.71 - samples/sec: 503.98 - lr: 0.000116 - momentum: 0.000000
2023-10-11 16:38:37,235 epoch 4 - iter 534/893 - loss 0.05076359 - time (sec): 296.52 - samples/sec: 505.45 - lr: 0.000114 - momentum: 0.000000
2023-10-11 16:39:27,083 epoch 4 - iter 623/893 - loss 0.05138865 - time (sec): 346.37 - samples/sec: 509.10 - lr: 0.000112 - momentum: 0.000000
2023-10-11 16:40:17,785 epoch 4 - iter 712/893 - loss 0.05173834 - time (sec): 397.07 - samples/sec: 502.80 - lr: 0.000110 - momentum: 0.000000
2023-10-11 16:41:07,370 epoch 4 - iter 801/893 - loss 0.05192190 - time (sec): 446.65 - samples/sec: 500.87 - lr: 0.000109 - momentum: 0.000000
2023-10-11 16:41:57,636 epoch 4 - iter 890/893 - loss 0.05095217 - time (sec): 496.92 - samples/sec: 499.12 - lr: 0.000107 - momentum: 0.000000
2023-10-11 16:41:59,142 ----------------------------------------------------------------------------------------------------
2023-10-11 16:41:59,143 EPOCH 4 done: loss 0.0509 - lr: 0.000107
2023-10-11 16:42:20,924 DEV : loss 0.13362975418567657 - f1-score (micro avg)  0.7709
2023-10-11 16:42:20,958 saving best model
2023-10-11 16:42:23,554 ----------------------------------------------------------------------------------------------------
2023-10-11 16:43:11,098 epoch 5 - iter 89/893 - loss 0.03653288 - time (sec): 47.54 - samples/sec: 500.14 - lr: 0.000105 - momentum: 0.000000
2023-10-11 16:44:00,117 epoch 5 - iter 178/893 - loss 0.03595217 - time (sec): 96.56 - samples/sec: 485.02 - lr: 0.000103 - momentum: 0.000000
2023-10-11 16:44:49,513 epoch 5 - iter 267/893 - loss 0.03646653 - time (sec): 145.95 - samples/sec: 502.95 - lr: 0.000101 - momentum: 0.000000
2023-10-11 16:45:39,118 epoch 5 - iter 356/893 - loss 0.03750037 - time (sec): 195.56 - samples/sec: 505.26 - lr: 0.000100 - momentum: 0.000000
2023-10-11 16:46:28,017 epoch 5 - iter 445/893 - loss 0.03695092 - time (sec): 244.46 - samples/sec: 503.19 - lr: 0.000098 - momentum: 0.000000
2023-10-11 16:47:16,434 epoch 5 - iter 534/893 - loss 0.03608794 - time (sec): 292.88 - samples/sec: 499.94 - lr: 0.000096 - momentum: 0.000000
2023-10-11 16:48:07,153 epoch 5 - iter 623/893 - loss 0.03592828 - time (sec): 343.59 - samples/sec: 502.61 - lr: 0.000094 - momentum: 0.000000
2023-10-11 16:48:56,551 epoch 5 - iter 712/893 - loss 0.03600119 - time (sec): 392.99 - samples/sec: 500.53 - lr: 0.000093 - momentum: 0.000000
2023-10-11 16:49:47,849 epoch 5 - iter 801/893 - loss 0.03656926 - time (sec): 444.29 - samples/sec: 500.02 - lr: 0.000091 - momentum: 0.000000
2023-10-11 16:50:39,277 epoch 5 - iter 890/893 - loss 0.03689169 - time (sec): 495.72 - samples/sec: 499.79 - lr: 0.000089 - momentum: 0.000000
2023-10-11 16:50:40,983 ----------------------------------------------------------------------------------------------------
2023-10-11 16:50:40,984 EPOCH 5 done: loss 0.0370 - lr: 0.000089
2023-10-11 16:51:02,760 DEV : loss 0.1409212052822113 - f1-score (micro avg)  0.7971
2023-10-11 16:51:02,797 saving best model
2023-10-11 16:51:05,388 ----------------------------------------------------------------------------------------------------
2023-10-11 16:51:56,629 epoch 6 - iter 89/893 - loss 0.03093811 - time (sec): 51.24 - samples/sec: 490.34 - lr: 0.000087 - momentum: 0.000000
2023-10-11 16:52:48,003 epoch 6 - iter 178/893 - loss 0.02884945 - time (sec): 102.61 - samples/sec: 483.47 - lr: 0.000085 - momentum: 0.000000
2023-10-11 16:53:37,672 epoch 6 - iter 267/893 - loss 0.02709581 - time (sec): 152.28 - samples/sec: 483.70 - lr: 0.000084 - momentum: 0.000000
2023-10-11 16:54:28,141 epoch 6 - iter 356/893 - loss 0.02801270 - time (sec): 202.75 - samples/sec: 486.61 - lr: 0.000082 - momentum: 0.000000
2023-10-11 16:55:18,447 epoch 6 - iter 445/893 - loss 0.02788463 - time (sec): 253.05 - samples/sec: 486.04 - lr: 0.000080 - momentum: 0.000000
2023-10-11 16:56:09,874 epoch 6 - iter 534/893 - loss 0.02959514 - time (sec): 304.48 - samples/sec: 487.32 - lr: 0.000078 - momentum: 0.000000
2023-10-11 16:57:00,389 epoch 6 - iter 623/893 - loss 0.02989629 - time (sec): 355.00 - samples/sec: 488.21 - lr: 0.000077 - momentum: 0.000000
2023-10-11 16:57:51,530 epoch 6 - iter 712/893 - loss 0.02978443 - time (sec): 406.14 - samples/sec: 487.85 - lr: 0.000075 - momentum: 0.000000
2023-10-11 16:58:41,657 epoch 6 - iter 801/893 - loss 0.02942788 - time (sec): 456.26 - samples/sec: 488.38 - lr: 0.000073 - momentum: 0.000000
2023-10-11 16:59:32,962 epoch 6 - iter 890/893 - loss 0.02909718 - time (sec): 507.57 - samples/sec: 488.53 - lr: 0.000071 - momentum: 0.000000
2023-10-11 16:59:34,505 ----------------------------------------------------------------------------------------------------
2023-10-11 16:59:34,505 EPOCH 6 done: loss 0.0292 - lr: 0.000071
2023-10-11 16:59:56,468 DEV : loss 0.16406482458114624 - f1-score (micro avg)  0.7936
2023-10-11 16:59:56,500 ----------------------------------------------------------------------------------------------------
2023-10-11 17:00:44,851 epoch 7 - iter 89/893 - loss 0.01961977 - time (sec): 48.35 - samples/sec: 500.18 - lr: 0.000069 - momentum: 0.000000
2023-10-11 17:01:34,295 epoch 7 - iter 178/893 - loss 0.02220619 - time (sec): 97.79 - samples/sec: 498.70 - lr: 0.000068 - momentum: 0.000000
2023-10-11 17:02:22,358 epoch 7 - iter 267/893 - loss 0.02475739 - time (sec): 145.86 - samples/sec: 499.63 - lr: 0.000066 - momentum: 0.000000
2023-10-11 17:03:11,495 epoch 7 - iter 356/893 - loss 0.02271444 - time (sec): 194.99 - samples/sec: 505.50 - lr: 0.000064 - momentum: 0.000000
2023-10-11 17:04:00,355 epoch 7 - iter 445/893 - loss 0.02294087 - time (sec): 243.85 - samples/sec: 508.07 - lr: 0.000062 - momentum: 0.000000
2023-10-11 17:04:49,325 epoch 7 - iter 534/893 - loss 0.02184486 - time (sec): 292.82 - samples/sec: 507.52 - lr: 0.000061 - momentum: 0.000000
2023-10-11 17:05:38,345 epoch 7 - iter 623/893 - loss 0.02120379 - time (sec): 341.84 - samples/sec: 506.40 - lr: 0.000059 - momentum: 0.000000
2023-10-11 17:06:29,178 epoch 7 - iter 712/893 - loss 0.02183594 - time (sec): 392.68 - samples/sec: 505.93 - lr: 0.000057 - momentum: 0.000000
2023-10-11 17:07:20,003 epoch 7 - iter 801/893 - loss 0.02155659 - time (sec): 443.50 - samples/sec: 504.30 - lr: 0.000055 - momentum: 0.000000
2023-10-11 17:08:08,594 epoch 7 - iter 890/893 - loss 0.02139749 - time (sec): 492.09 - samples/sec: 503.21 - lr: 0.000053 - momentum: 0.000000
2023-10-11 17:08:10,436 ----------------------------------------------------------------------------------------------------
2023-10-11 17:08:10,436 EPOCH 7 done: loss 0.0213 - lr: 0.000053
2023-10-11 17:08:31,770 DEV : loss 0.17885611951351166 - f1-score (micro avg)  0.8011
2023-10-11 17:08:31,801 saving best model
2023-10-11 17:08:34,366 ----------------------------------------------------------------------------------------------------
2023-10-11 17:09:24,407 epoch 8 - iter 89/893 - loss 0.01715620 - time (sec): 50.04 - samples/sec: 493.87 - lr: 0.000052 - momentum: 0.000000
2023-10-11 17:10:16,482 epoch 8 - iter 178/893 - loss 0.02006197 - time (sec): 102.11 - samples/sec: 494.07 - lr: 0.000050 - momentum: 0.000000
2023-10-11 17:11:06,885 epoch 8 - iter 267/893 - loss 0.01738780 - time (sec): 152.52 - samples/sec: 494.04 - lr: 0.000048 - momentum: 0.000000
2023-10-11 17:11:57,243 epoch 8 - iter 356/893 - loss 0.01811409 - time (sec): 202.87 - samples/sec: 494.75 - lr: 0.000046 - momentum: 0.000000
2023-10-11 17:12:46,260 epoch 8 - iter 445/893 - loss 0.01897415 - time (sec): 251.89 - samples/sec: 493.42 - lr: 0.000045 - momentum: 0.000000
2023-10-11 17:13:35,565 epoch 8 - iter 534/893 - loss 0.01805358 - time (sec): 301.20 - samples/sec: 493.49 - lr: 0.000043 - momentum: 0.000000
2023-10-11 17:14:26,683 epoch 8 - iter 623/893 - loss 0.01824633 - time (sec): 352.31 - samples/sec: 494.18 - lr: 0.000041 - momentum: 0.000000
2023-10-11 17:15:14,708 epoch 8 - iter 712/893 - loss 0.01811871 - time (sec): 400.34 - samples/sec: 491.30 - lr: 0.000039 - momentum: 0.000000
2023-10-11 17:16:04,826 epoch 8 - iter 801/893 - loss 0.01785414 - time (sec): 450.46 - samples/sec: 494.17 - lr: 0.000037 - momentum: 0.000000
2023-10-11 17:16:55,097 epoch 8 - iter 890/893 - loss 0.01770565 - time (sec): 500.73 - samples/sec: 495.42 - lr: 0.000036 - momentum: 0.000000
2023-10-11 17:16:56,588 ----------------------------------------------------------------------------------------------------
2023-10-11 17:16:56,588 EPOCH 8 done: loss 0.0177 - lr: 0.000036
2023-10-11 17:17:18,690 DEV : loss 0.19161181151866913 - f1-score (micro avg)  0.7955
2023-10-11 17:17:18,720 ----------------------------------------------------------------------------------------------------
2023-10-11 17:18:06,228 epoch 9 - iter 89/893 - loss 0.01298334 - time (sec): 47.51 - samples/sec: 497.93 - lr: 0.000034 - momentum: 0.000000
2023-10-11 17:18:55,387 epoch 9 - iter 178/893 - loss 0.01537760 - time (sec): 96.67 - samples/sec: 511.42 - lr: 0.000032 - momentum: 0.000000
2023-10-11 17:19:45,056 epoch 9 - iter 267/893 - loss 0.01528293 - time (sec): 146.33 - samples/sec: 517.74 - lr: 0.000030 - momentum: 0.000000
2023-10-11 17:20:34,247 epoch 9 - iter 356/893 - loss 0.01453369 - time (sec): 195.53 - samples/sec: 514.55 - lr: 0.000029 - momentum: 0.000000
2023-10-11 17:21:22,221 epoch 9 - iter 445/893 - loss 0.01412119 - time (sec): 243.50 - samples/sec: 512.37 - lr: 0.000027 - momentum: 0.000000
2023-10-11 17:22:10,404 epoch 9 - iter 534/893 - loss 0.01411537 - time (sec): 291.68 - samples/sec: 513.33 - lr: 0.000025 - momentum: 0.000000
2023-10-11 17:22:58,710 epoch 9 - iter 623/893 - loss 0.01468822 - time (sec): 339.99 - samples/sec: 511.65 - lr: 0.000023 - momentum: 0.000000
2023-10-11 17:23:46,894 epoch 9 - iter 712/893 - loss 0.01427033 - time (sec): 388.17 - samples/sec: 510.66 - lr: 0.000022 - momentum: 0.000000
2023-10-11 17:24:35,059 epoch 9 - iter 801/893 - loss 0.01423814 - time (sec): 436.34 - samples/sec: 509.86 - lr: 0.000020 - momentum: 0.000000
2023-10-11 17:25:24,567 epoch 9 - iter 890/893 - loss 0.01466708 - time (sec): 485.85 - samples/sec: 510.22 - lr: 0.000018 - momentum: 0.000000
2023-10-11 17:25:26,215 ----------------------------------------------------------------------------------------------------
2023-10-11 17:25:26,216 EPOCH 9 done: loss 0.0146 - lr: 0.000018
2023-10-11 17:25:46,751 DEV : loss 0.19483081996440887 - f1-score (micro avg)  0.7944
2023-10-11 17:25:46,781 ----------------------------------------------------------------------------------------------------
2023-10-11 17:26:35,157 epoch 10 - iter 89/893 - loss 0.01073313 - time (sec): 48.37 - samples/sec: 520.32 - lr: 0.000016 - momentum: 0.000000
2023-10-11 17:27:25,428 epoch 10 - iter 178/893 - loss 0.01013166 - time (sec): 98.65 - samples/sec: 519.71 - lr: 0.000014 - momentum: 0.000000
2023-10-11 17:28:12,901 epoch 10 - iter 267/893 - loss 0.01112463 - time (sec): 146.12 - samples/sec: 510.31 - lr: 0.000013 - momentum: 0.000000
2023-10-11 17:29:01,976 epoch 10 - iter 356/893 - loss 0.01098189 - time (sec): 195.19 - samples/sec: 508.58 - lr: 0.000011 - momentum: 0.000000
2023-10-11 17:29:49,819 epoch 10 - iter 445/893 - loss 0.01066311 - time (sec): 243.04 - samples/sec: 503.61 - lr: 0.000009 - momentum: 0.000000
2023-10-11 17:30:40,164 epoch 10 - iter 534/893 - loss 0.01119657 - time (sec): 293.38 - samples/sec: 508.28 - lr: 0.000007 - momentum: 0.000000
2023-10-11 17:31:28,238 epoch 10 - iter 623/893 - loss 0.01162212 - time (sec): 341.45 - samples/sec: 506.25 - lr: 0.000006 - momentum: 0.000000
2023-10-11 17:32:17,725 epoch 10 - iter 712/893 - loss 0.01182235 - time (sec): 390.94 - samples/sec: 506.41 - lr: 0.000004 - momentum: 0.000000
2023-10-11 17:33:08,012 epoch 10 - iter 801/893 - loss 0.01144574 - time (sec): 441.23 - samples/sec: 506.35 - lr: 0.000002 - momentum: 0.000000
2023-10-11 17:33:58,609 epoch 10 - iter 890/893 - loss 0.01158836 - time (sec): 491.83 - samples/sec: 504.65 - lr: 0.000000 - momentum: 0.000000
2023-10-11 17:34:00,008 ----------------------------------------------------------------------------------------------------
2023-10-11 17:34:00,008 EPOCH 10 done: loss 0.0116 - lr: 0.000000
2023-10-11 17:34:20,889 DEV : loss 0.19962604343891144 - f1-score (micro avg)  0.7897
2023-10-11 17:34:21,786 ----------------------------------------------------------------------------------------------------
2023-10-11 17:34:21,788 Loading model from best epoch ...
2023-10-11 17:34:26,562 SequenceTagger predicts: Dictionary with 17 tags: O, S-PER, B-PER, E-PER, I-PER, S-LOC, B-LOC, E-LOC, I-LOC, S-ORG, B-ORG, E-ORG, I-ORG, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd
2023-10-11 17:35:36,469 
Results:
- F-score (micro) 0.6917
- F-score (macro) 0.5961
- Accuracy 0.545

By class:
              precision    recall  f1-score   support

         LOC     0.7070    0.6986    0.7028      1095
         PER     0.7745    0.7806    0.7776      1012
         ORG     0.4204    0.5770    0.4864       357
   HumanProd     0.3276    0.5758    0.4176        33

   micro avg     0.6717    0.7129    0.6917      2497
   macro avg     0.5574    0.6580    0.5961      2497
weighted avg     0.6884    0.7129    0.6984      2497

2023-10-11 17:35:36,470 ----------------------------------------------------------------------------------------------------