File size: 26,535 Bytes
d360f74 |
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 |
2024-03-26 11:33:29,248 ----------------------------------------------------------------------------------------------------
2024-03-26 11:33:29,248 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): BertModel(
(embeddings): BertEmbeddings(
(word_embeddings): Embedding(30001, 768)
(position_embeddings): Embedding(512, 768)
(token_type_embeddings): Embedding(2, 768)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(encoder): BertEncoder(
(layer): ModuleList(
(0-11): 12 x BertLayer(
(attention): BertAttention(
(self): BertSelfAttention(
(query): Linear(in_features=768, out_features=768, bias=True)
(key): Linear(in_features=768, out_features=768, bias=True)
(value): Linear(in_features=768, out_features=768, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(output): BertSelfOutput(
(dense): Linear(in_features=768, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(intermediate): BertIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): BertOutput(
(dense): Linear(in_features=3072, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
(pooler): BertPooler(
(dense): Linear(in_features=768, out_features=768, bias=True)
(activation): Tanh()
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=768, out_features=17, bias=True)
(loss_function): CrossEntropyLoss()
)"
2024-03-26 11:33:29,248 ----------------------------------------------------------------------------------------------------
2024-03-26 11:33:29,248 Corpus: 758 train + 94 dev + 96 test sentences
2024-03-26 11:33:29,248 ----------------------------------------------------------------------------------------------------
2024-03-26 11:33:29,248 Train: 758 sentences
2024-03-26 11:33:29,248 (train_with_dev=False, train_with_test=False)
2024-03-26 11:33:29,248 ----------------------------------------------------------------------------------------------------
2024-03-26 11:33:29,248 Training Params:
2024-03-26 11:33:29,248 - learning_rate: "5e-05"
2024-03-26 11:33:29,248 - mini_batch_size: "16"
2024-03-26 11:33:29,248 - max_epochs: "10"
2024-03-26 11:33:29,248 - shuffle: "True"
2024-03-26 11:33:29,248 ----------------------------------------------------------------------------------------------------
2024-03-26 11:33:29,248 Plugins:
2024-03-26 11:33:29,248 - TensorboardLogger
2024-03-26 11:33:29,248 - LinearScheduler | warmup_fraction: '0.1'
2024-03-26 11:33:29,248 ----------------------------------------------------------------------------------------------------
2024-03-26 11:33:29,248 Final evaluation on model from best epoch (best-model.pt)
2024-03-26 11:33:29,248 - metric: "('micro avg', 'f1-score')"
2024-03-26 11:33:29,248 ----------------------------------------------------------------------------------------------------
2024-03-26 11:33:29,248 Computation:
2024-03-26 11:33:29,248 - compute on device: cuda:0
2024-03-26 11:33:29,248 - embedding storage: none
2024-03-26 11:33:29,248 ----------------------------------------------------------------------------------------------------
2024-03-26 11:33:29,248 Model training base path: "flair-co-funer-german_bert_base-bs16-e10-lr5e-05-3"
2024-03-26 11:33:29,248 ----------------------------------------------------------------------------------------------------
2024-03-26 11:33:29,249 ----------------------------------------------------------------------------------------------------
2024-03-26 11:33:29,249 Logging anything other than scalars to TensorBoard is currently not supported.
2024-03-26 11:33:30,579 epoch 1 - iter 4/48 - loss 3.07489042 - time (sec): 1.33 - samples/sec: 2069.01 - lr: 0.000003 - momentum: 0.000000
2024-03-26 11:33:32,704 epoch 1 - iter 8/48 - loss 3.08771911 - time (sec): 3.46 - samples/sec: 1685.38 - lr: 0.000007 - momentum: 0.000000
2024-03-26 11:33:34,228 epoch 1 - iter 12/48 - loss 2.94390144 - time (sec): 4.98 - samples/sec: 1681.86 - lr: 0.000011 - momentum: 0.000000
2024-03-26 11:33:37,270 epoch 1 - iter 16/48 - loss 2.77162015 - time (sec): 8.02 - samples/sec: 1445.94 - lr: 0.000016 - momentum: 0.000000
2024-03-26 11:33:39,010 epoch 1 - iter 20/48 - loss 2.60962695 - time (sec): 9.76 - samples/sec: 1475.60 - lr: 0.000020 - momentum: 0.000000
2024-03-26 11:33:40,472 epoch 1 - iter 24/48 - loss 2.51067364 - time (sec): 11.22 - samples/sec: 1529.72 - lr: 0.000024 - momentum: 0.000000
2024-03-26 11:33:41,866 epoch 1 - iter 28/48 - loss 2.39999582 - time (sec): 12.62 - samples/sec: 1547.60 - lr: 0.000028 - momentum: 0.000000
2024-03-26 11:33:44,008 epoch 1 - iter 32/48 - loss 2.27938260 - time (sec): 14.76 - samples/sec: 1539.92 - lr: 0.000032 - momentum: 0.000000
2024-03-26 11:33:45,593 epoch 1 - iter 36/48 - loss 2.16989687 - time (sec): 16.34 - samples/sec: 1561.40 - lr: 0.000036 - momentum: 0.000000
2024-03-26 11:33:47,867 epoch 1 - iter 40/48 - loss 2.03915890 - time (sec): 18.62 - samples/sec: 1556.37 - lr: 0.000041 - momentum: 0.000000
2024-03-26 11:33:49,800 epoch 1 - iter 44/48 - loss 1.93179330 - time (sec): 20.55 - samples/sec: 1560.33 - lr: 0.000045 - momentum: 0.000000
2024-03-26 11:33:51,477 epoch 1 - iter 48/48 - loss 1.84076884 - time (sec): 22.23 - samples/sec: 1550.84 - lr: 0.000049 - momentum: 0.000000
2024-03-26 11:33:51,477 ----------------------------------------------------------------------------------------------------
2024-03-26 11:33:51,477 EPOCH 1 done: loss 1.8408 - lr: 0.000049
2024-03-26 11:33:52,341 DEV : loss 0.5940399169921875 - f1-score (micro avg) 0.635
2024-03-26 11:33:52,342 saving best model
2024-03-26 11:33:52,617 ----------------------------------------------------------------------------------------------------
2024-03-26 11:33:54,062 epoch 2 - iter 4/48 - loss 0.72316569 - time (sec): 1.44 - samples/sec: 1727.06 - lr: 0.000050 - momentum: 0.000000
2024-03-26 11:33:55,558 epoch 2 - iter 8/48 - loss 0.57588712 - time (sec): 2.94 - samples/sec: 1659.65 - lr: 0.000049 - momentum: 0.000000
2024-03-26 11:33:57,043 epoch 2 - iter 12/48 - loss 0.56563452 - time (sec): 4.43 - samples/sec: 1737.03 - lr: 0.000049 - momentum: 0.000000
2024-03-26 11:33:59,008 epoch 2 - iter 16/48 - loss 0.52469522 - time (sec): 6.39 - samples/sec: 1683.10 - lr: 0.000048 - momentum: 0.000000
2024-03-26 11:34:01,353 epoch 2 - iter 20/48 - loss 0.50473970 - time (sec): 8.74 - samples/sec: 1623.17 - lr: 0.000048 - momentum: 0.000000
2024-03-26 11:34:03,440 epoch 2 - iter 24/48 - loss 0.47203328 - time (sec): 10.82 - samples/sec: 1602.62 - lr: 0.000047 - momentum: 0.000000
2024-03-26 11:34:06,292 epoch 2 - iter 28/48 - loss 0.46192324 - time (sec): 13.68 - samples/sec: 1530.16 - lr: 0.000047 - momentum: 0.000000
2024-03-26 11:34:08,529 epoch 2 - iter 32/48 - loss 0.44576921 - time (sec): 15.91 - samples/sec: 1500.06 - lr: 0.000046 - momentum: 0.000000
2024-03-26 11:34:10,295 epoch 2 - iter 36/48 - loss 0.43726579 - time (sec): 17.68 - samples/sec: 1494.39 - lr: 0.000046 - momentum: 0.000000
2024-03-26 11:34:12,061 epoch 2 - iter 40/48 - loss 0.43733009 - time (sec): 19.44 - samples/sec: 1500.28 - lr: 0.000046 - momentum: 0.000000
2024-03-26 11:34:14,381 epoch 2 - iter 44/48 - loss 0.42304062 - time (sec): 21.76 - samples/sec: 1488.86 - lr: 0.000045 - momentum: 0.000000
2024-03-26 11:34:15,946 epoch 2 - iter 48/48 - loss 0.41515240 - time (sec): 23.33 - samples/sec: 1477.66 - lr: 0.000045 - momentum: 0.000000
2024-03-26 11:34:15,946 ----------------------------------------------------------------------------------------------------
2024-03-26 11:34:15,946 EPOCH 2 done: loss 0.4152 - lr: 0.000045
2024-03-26 11:34:16,914 DEV : loss 0.26913678646087646 - f1-score (micro avg) 0.8164
2024-03-26 11:34:16,917 saving best model
2024-03-26 11:34:17,390 ----------------------------------------------------------------------------------------------------
2024-03-26 11:34:18,964 epoch 3 - iter 4/48 - loss 0.27840730 - time (sec): 1.57 - samples/sec: 1558.05 - lr: 0.000044 - momentum: 0.000000
2024-03-26 11:34:21,788 epoch 3 - iter 8/48 - loss 0.22660795 - time (sec): 4.40 - samples/sec: 1303.04 - lr: 0.000044 - momentum: 0.000000
2024-03-26 11:34:23,083 epoch 3 - iter 12/48 - loss 0.22777721 - time (sec): 5.69 - samples/sec: 1431.27 - lr: 0.000043 - momentum: 0.000000
2024-03-26 11:34:24,492 epoch 3 - iter 16/48 - loss 0.20623331 - time (sec): 7.10 - samples/sec: 1554.57 - lr: 0.000043 - momentum: 0.000000
2024-03-26 11:34:26,009 epoch 3 - iter 20/48 - loss 0.20655534 - time (sec): 8.62 - samples/sec: 1567.86 - lr: 0.000042 - momentum: 0.000000
2024-03-26 11:34:28,814 epoch 3 - iter 24/48 - loss 0.20077536 - time (sec): 11.42 - samples/sec: 1462.28 - lr: 0.000042 - momentum: 0.000000
2024-03-26 11:34:30,804 epoch 3 - iter 28/48 - loss 0.20515197 - time (sec): 13.41 - samples/sec: 1477.75 - lr: 0.000041 - momentum: 0.000000
2024-03-26 11:34:33,323 epoch 3 - iter 32/48 - loss 0.19788947 - time (sec): 15.93 - samples/sec: 1431.16 - lr: 0.000041 - momentum: 0.000000
2024-03-26 11:34:35,287 epoch 3 - iter 36/48 - loss 0.20165014 - time (sec): 17.90 - samples/sec: 1429.28 - lr: 0.000040 - momentum: 0.000000
2024-03-26 11:34:37,666 epoch 3 - iter 40/48 - loss 0.19334685 - time (sec): 20.27 - samples/sec: 1408.84 - lr: 0.000040 - momentum: 0.000000
2024-03-26 11:34:40,145 epoch 3 - iter 44/48 - loss 0.20288087 - time (sec): 22.75 - samples/sec: 1398.52 - lr: 0.000040 - momentum: 0.000000
2024-03-26 11:34:42,579 epoch 3 - iter 48/48 - loss 0.19661772 - time (sec): 25.19 - samples/sec: 1368.61 - lr: 0.000039 - momentum: 0.000000
2024-03-26 11:34:42,579 ----------------------------------------------------------------------------------------------------
2024-03-26 11:34:42,579 EPOCH 3 done: loss 0.1966 - lr: 0.000039
2024-03-26 11:34:43,613 DEV : loss 0.21700386703014374 - f1-score (micro avg) 0.8608
2024-03-26 11:34:43,614 saving best model
2024-03-26 11:34:44,036 ----------------------------------------------------------------------------------------------------
2024-03-26 11:34:45,449 epoch 4 - iter 4/48 - loss 0.14641328 - time (sec): 1.41 - samples/sec: 1773.85 - lr: 0.000039 - momentum: 0.000000
2024-03-26 11:34:47,395 epoch 4 - iter 8/48 - loss 0.13808574 - time (sec): 3.36 - samples/sec: 1595.81 - lr: 0.000038 - momentum: 0.000000
2024-03-26 11:34:50,115 epoch 4 - iter 12/48 - loss 0.12876595 - time (sec): 6.08 - samples/sec: 1388.37 - lr: 0.000038 - momentum: 0.000000
2024-03-26 11:34:52,081 epoch 4 - iter 16/48 - loss 0.13156323 - time (sec): 8.04 - samples/sec: 1407.85 - lr: 0.000037 - momentum: 0.000000
2024-03-26 11:34:54,572 epoch 4 - iter 20/48 - loss 0.12428123 - time (sec): 10.53 - samples/sec: 1394.75 - lr: 0.000037 - momentum: 0.000000
2024-03-26 11:34:57,507 epoch 4 - iter 24/48 - loss 0.11793333 - time (sec): 13.47 - samples/sec: 1353.70 - lr: 0.000036 - momentum: 0.000000
2024-03-26 11:34:58,691 epoch 4 - iter 28/48 - loss 0.11608881 - time (sec): 14.65 - samples/sec: 1386.24 - lr: 0.000036 - momentum: 0.000000
2024-03-26 11:35:01,804 epoch 4 - iter 32/48 - loss 0.11458934 - time (sec): 17.77 - samples/sec: 1329.30 - lr: 0.000035 - momentum: 0.000000
2024-03-26 11:35:03,594 epoch 4 - iter 36/48 - loss 0.11846800 - time (sec): 19.56 - samples/sec: 1360.43 - lr: 0.000035 - momentum: 0.000000
2024-03-26 11:35:06,500 epoch 4 - iter 40/48 - loss 0.12511856 - time (sec): 22.46 - samples/sec: 1330.40 - lr: 0.000034 - momentum: 0.000000
2024-03-26 11:35:07,417 epoch 4 - iter 44/48 - loss 0.12629946 - time (sec): 23.38 - samples/sec: 1376.32 - lr: 0.000034 - momentum: 0.000000
2024-03-26 11:35:08,913 epoch 4 - iter 48/48 - loss 0.12787118 - time (sec): 24.88 - samples/sec: 1385.74 - lr: 0.000034 - momentum: 0.000000
2024-03-26 11:35:08,914 ----------------------------------------------------------------------------------------------------
2024-03-26 11:35:08,914 EPOCH 4 done: loss 0.1279 - lr: 0.000034
2024-03-26 11:35:09,895 DEV : loss 0.18662486970424652 - f1-score (micro avg) 0.8817
2024-03-26 11:35:09,897 saving best model
2024-03-26 11:35:10,332 ----------------------------------------------------------------------------------------------------
2024-03-26 11:35:12,809 epoch 5 - iter 4/48 - loss 0.06830634 - time (sec): 2.48 - samples/sec: 1283.35 - lr: 0.000033 - momentum: 0.000000
2024-03-26 11:35:14,268 epoch 5 - iter 8/48 - loss 0.08573760 - time (sec): 3.94 - samples/sec: 1446.25 - lr: 0.000033 - momentum: 0.000000
2024-03-26 11:35:15,788 epoch 5 - iter 12/48 - loss 0.08684380 - time (sec): 5.46 - samples/sec: 1510.47 - lr: 0.000032 - momentum: 0.000000
2024-03-26 11:35:18,018 epoch 5 - iter 16/48 - loss 0.08965837 - time (sec): 7.69 - samples/sec: 1431.53 - lr: 0.000032 - momentum: 0.000000
2024-03-26 11:35:20,150 epoch 5 - iter 20/48 - loss 0.10094467 - time (sec): 9.82 - samples/sec: 1434.23 - lr: 0.000031 - momentum: 0.000000
2024-03-26 11:35:22,751 epoch 5 - iter 24/48 - loss 0.09379417 - time (sec): 12.42 - samples/sec: 1417.55 - lr: 0.000031 - momentum: 0.000000
2024-03-26 11:35:25,342 epoch 5 - iter 28/48 - loss 0.08737934 - time (sec): 15.01 - samples/sec: 1402.35 - lr: 0.000030 - momentum: 0.000000
2024-03-26 11:35:27,327 epoch 5 - iter 32/48 - loss 0.08652809 - time (sec): 16.99 - samples/sec: 1401.19 - lr: 0.000030 - momentum: 0.000000
2024-03-26 11:35:29,218 epoch 5 - iter 36/48 - loss 0.08405259 - time (sec): 18.89 - samples/sec: 1400.33 - lr: 0.000029 - momentum: 0.000000
2024-03-26 11:35:31,655 epoch 5 - iter 40/48 - loss 0.08424074 - time (sec): 21.32 - samples/sec: 1385.12 - lr: 0.000029 - momentum: 0.000000
2024-03-26 11:35:33,674 epoch 5 - iter 44/48 - loss 0.08675698 - time (sec): 23.34 - samples/sec: 1383.79 - lr: 0.000029 - momentum: 0.000000
2024-03-26 11:35:34,768 epoch 5 - iter 48/48 - loss 0.08578516 - time (sec): 24.44 - samples/sec: 1410.70 - lr: 0.000028 - momentum: 0.000000
2024-03-26 11:35:34,769 ----------------------------------------------------------------------------------------------------
2024-03-26 11:35:34,769 EPOCH 5 done: loss 0.0858 - lr: 0.000028
2024-03-26 11:35:35,716 DEV : loss 0.17283330857753754 - f1-score (micro avg) 0.8942
2024-03-26 11:35:35,717 saving best model
2024-03-26 11:35:36,174 ----------------------------------------------------------------------------------------------------
2024-03-26 11:35:38,850 epoch 6 - iter 4/48 - loss 0.05365840 - time (sec): 2.68 - samples/sec: 1188.40 - lr: 0.000028 - momentum: 0.000000
2024-03-26 11:35:40,878 epoch 6 - iter 8/48 - loss 0.05720723 - time (sec): 4.70 - samples/sec: 1248.47 - lr: 0.000027 - momentum: 0.000000
2024-03-26 11:35:42,486 epoch 6 - iter 12/48 - loss 0.06649040 - time (sec): 6.31 - samples/sec: 1397.93 - lr: 0.000027 - momentum: 0.000000
2024-03-26 11:35:44,519 epoch 6 - iter 16/48 - loss 0.06053967 - time (sec): 8.34 - samples/sec: 1393.76 - lr: 0.000026 - momentum: 0.000000
2024-03-26 11:35:45,652 epoch 6 - iter 20/48 - loss 0.05981345 - time (sec): 9.48 - samples/sec: 1474.40 - lr: 0.000026 - momentum: 0.000000
2024-03-26 11:35:47,585 epoch 6 - iter 24/48 - loss 0.05907532 - time (sec): 11.41 - samples/sec: 1464.86 - lr: 0.000025 - momentum: 0.000000
2024-03-26 11:35:48,768 epoch 6 - iter 28/48 - loss 0.06005105 - time (sec): 12.59 - samples/sec: 1510.48 - lr: 0.000025 - momentum: 0.000000
2024-03-26 11:35:50,567 epoch 6 - iter 32/48 - loss 0.05586702 - time (sec): 14.39 - samples/sec: 1530.72 - lr: 0.000024 - momentum: 0.000000
2024-03-26 11:35:53,101 epoch 6 - iter 36/48 - loss 0.06595337 - time (sec): 16.93 - samples/sec: 1500.21 - lr: 0.000024 - momentum: 0.000000
2024-03-26 11:35:55,243 epoch 6 - iter 40/48 - loss 0.06371281 - time (sec): 19.07 - samples/sec: 1490.16 - lr: 0.000023 - momentum: 0.000000
2024-03-26 11:35:57,163 epoch 6 - iter 44/48 - loss 0.06590059 - time (sec): 20.99 - samples/sec: 1498.80 - lr: 0.000023 - momentum: 0.000000
2024-03-26 11:35:58,837 epoch 6 - iter 48/48 - loss 0.06662131 - time (sec): 22.66 - samples/sec: 1521.17 - lr: 0.000023 - momentum: 0.000000
2024-03-26 11:35:58,837 ----------------------------------------------------------------------------------------------------
2024-03-26 11:35:58,837 EPOCH 6 done: loss 0.0666 - lr: 0.000023
2024-03-26 11:35:59,808 DEV : loss 0.15879587829113007 - f1-score (micro avg) 0.9341
2024-03-26 11:35:59,809 saving best model
2024-03-26 11:36:00,267 ----------------------------------------------------------------------------------------------------
2024-03-26 11:36:02,501 epoch 7 - iter 4/48 - loss 0.07284967 - time (sec): 2.23 - samples/sec: 1237.82 - lr: 0.000022 - momentum: 0.000000
2024-03-26 11:36:04,244 epoch 7 - iter 8/48 - loss 0.05810049 - time (sec): 3.98 - samples/sec: 1445.13 - lr: 0.000022 - momentum: 0.000000
2024-03-26 11:36:06,352 epoch 7 - iter 12/48 - loss 0.04687988 - time (sec): 6.08 - samples/sec: 1408.13 - lr: 0.000021 - momentum: 0.000000
2024-03-26 11:36:09,051 epoch 7 - iter 16/48 - loss 0.04388564 - time (sec): 8.78 - samples/sec: 1345.71 - lr: 0.000021 - momentum: 0.000000
2024-03-26 11:36:11,822 epoch 7 - iter 20/48 - loss 0.04451078 - time (sec): 11.55 - samples/sec: 1352.70 - lr: 0.000020 - momentum: 0.000000
2024-03-26 11:36:13,366 epoch 7 - iter 24/48 - loss 0.04438006 - time (sec): 13.10 - samples/sec: 1376.06 - lr: 0.000020 - momentum: 0.000000
2024-03-26 11:36:15,571 epoch 7 - iter 28/48 - loss 0.04184705 - time (sec): 15.30 - samples/sec: 1390.16 - lr: 0.000019 - momentum: 0.000000
2024-03-26 11:36:17,773 epoch 7 - iter 32/48 - loss 0.04525184 - time (sec): 17.51 - samples/sec: 1396.63 - lr: 0.000019 - momentum: 0.000000
2024-03-26 11:36:20,032 epoch 7 - iter 36/48 - loss 0.04947562 - time (sec): 19.76 - samples/sec: 1385.10 - lr: 0.000018 - momentum: 0.000000
2024-03-26 11:36:21,659 epoch 7 - iter 40/48 - loss 0.04694456 - time (sec): 21.39 - samples/sec: 1393.87 - lr: 0.000018 - momentum: 0.000000
2024-03-26 11:36:23,419 epoch 7 - iter 44/48 - loss 0.05016226 - time (sec): 23.15 - samples/sec: 1408.92 - lr: 0.000017 - momentum: 0.000000
2024-03-26 11:36:24,776 epoch 7 - iter 48/48 - loss 0.04985462 - time (sec): 24.51 - samples/sec: 1406.54 - lr: 0.000017 - momentum: 0.000000
2024-03-26 11:36:24,777 ----------------------------------------------------------------------------------------------------
2024-03-26 11:36:24,777 EPOCH 7 done: loss 0.0499 - lr: 0.000017
2024-03-26 11:36:25,828 DEV : loss 0.16802552342414856 - f1-score (micro avg) 0.9333
2024-03-26 11:36:25,830 ----------------------------------------------------------------------------------------------------
2024-03-26 11:36:28,176 epoch 8 - iter 4/48 - loss 0.02523109 - time (sec): 2.35 - samples/sec: 1253.12 - lr: 0.000017 - momentum: 0.000000
2024-03-26 11:36:30,775 epoch 8 - iter 8/48 - loss 0.02208519 - time (sec): 4.94 - samples/sec: 1337.73 - lr: 0.000016 - momentum: 0.000000
2024-03-26 11:36:32,833 epoch 8 - iter 12/48 - loss 0.02316165 - time (sec): 7.00 - samples/sec: 1314.10 - lr: 0.000016 - momentum: 0.000000
2024-03-26 11:36:34,948 epoch 8 - iter 16/48 - loss 0.02270118 - time (sec): 9.12 - samples/sec: 1315.04 - lr: 0.000015 - momentum: 0.000000
2024-03-26 11:36:36,478 epoch 8 - iter 20/48 - loss 0.02510760 - time (sec): 10.65 - samples/sec: 1342.76 - lr: 0.000015 - momentum: 0.000000
2024-03-26 11:36:38,968 epoch 8 - iter 24/48 - loss 0.02440371 - time (sec): 13.14 - samples/sec: 1319.99 - lr: 0.000014 - momentum: 0.000000
2024-03-26 11:36:41,209 epoch 8 - iter 28/48 - loss 0.02404743 - time (sec): 15.38 - samples/sec: 1310.63 - lr: 0.000014 - momentum: 0.000000
2024-03-26 11:36:43,580 epoch 8 - iter 32/48 - loss 0.03201898 - time (sec): 17.75 - samples/sec: 1322.64 - lr: 0.000013 - momentum: 0.000000
2024-03-26 11:36:46,883 epoch 8 - iter 36/48 - loss 0.03651999 - time (sec): 21.05 - samples/sec: 1274.74 - lr: 0.000013 - momentum: 0.000000
2024-03-26 11:36:48,978 epoch 8 - iter 40/48 - loss 0.04125936 - time (sec): 23.15 - samples/sec: 1278.60 - lr: 0.000012 - momentum: 0.000000
2024-03-26 11:36:49,806 epoch 8 - iter 44/48 - loss 0.04000375 - time (sec): 23.98 - samples/sec: 1325.12 - lr: 0.000012 - momentum: 0.000000
2024-03-26 11:36:51,680 epoch 8 - iter 48/48 - loss 0.03939075 - time (sec): 25.85 - samples/sec: 1333.53 - lr: 0.000011 - momentum: 0.000000
2024-03-26 11:36:51,681 ----------------------------------------------------------------------------------------------------
2024-03-26 11:36:51,681 EPOCH 8 done: loss 0.0394 - lr: 0.000011
2024-03-26 11:36:52,644 DEV : loss 0.18046870827674866 - f1-score (micro avg) 0.9294
2024-03-26 11:36:52,647 ----------------------------------------------------------------------------------------------------
2024-03-26 11:36:55,407 epoch 9 - iter 4/48 - loss 0.01368922 - time (sec): 2.76 - samples/sec: 1195.26 - lr: 0.000011 - momentum: 0.000000
2024-03-26 11:36:57,113 epoch 9 - iter 8/48 - loss 0.02194754 - time (sec): 4.47 - samples/sec: 1284.97 - lr: 0.000011 - momentum: 0.000000
2024-03-26 11:36:59,325 epoch 9 - iter 12/48 - loss 0.02259365 - time (sec): 6.68 - samples/sec: 1344.23 - lr: 0.000010 - momentum: 0.000000
2024-03-26 11:37:01,488 epoch 9 - iter 16/48 - loss 0.02747245 - time (sec): 8.84 - samples/sec: 1369.80 - lr: 0.000010 - momentum: 0.000000
2024-03-26 11:37:03,868 epoch 9 - iter 20/48 - loss 0.02337877 - time (sec): 11.22 - samples/sec: 1348.38 - lr: 0.000009 - momentum: 0.000000
2024-03-26 11:37:05,818 epoch 9 - iter 24/48 - loss 0.02369286 - time (sec): 13.17 - samples/sec: 1344.53 - lr: 0.000009 - momentum: 0.000000
2024-03-26 11:37:09,104 epoch 9 - iter 28/48 - loss 0.02677139 - time (sec): 16.46 - samples/sec: 1298.13 - lr: 0.000008 - momentum: 0.000000
2024-03-26 11:37:10,540 epoch 9 - iter 32/48 - loss 0.02760924 - time (sec): 17.89 - samples/sec: 1335.17 - lr: 0.000008 - momentum: 0.000000
2024-03-26 11:37:13,071 epoch 9 - iter 36/48 - loss 0.02778603 - time (sec): 20.42 - samples/sec: 1319.50 - lr: 0.000007 - momentum: 0.000000
2024-03-26 11:37:14,592 epoch 9 - iter 40/48 - loss 0.02879429 - time (sec): 21.94 - samples/sec: 1336.09 - lr: 0.000007 - momentum: 0.000000
2024-03-26 11:37:16,189 epoch 9 - iter 44/48 - loss 0.03224152 - time (sec): 23.54 - samples/sec: 1351.25 - lr: 0.000006 - momentum: 0.000000
2024-03-26 11:37:17,651 epoch 9 - iter 48/48 - loss 0.03182031 - time (sec): 25.00 - samples/sec: 1378.69 - lr: 0.000006 - momentum: 0.000000
2024-03-26 11:37:17,651 ----------------------------------------------------------------------------------------------------
2024-03-26 11:37:17,651 EPOCH 9 done: loss 0.0318 - lr: 0.000006
2024-03-26 11:37:18,590 DEV : loss 0.18695427477359772 - f1-score (micro avg) 0.9263
2024-03-26 11:37:18,591 ----------------------------------------------------------------------------------------------------
2024-03-26 11:37:21,015 epoch 10 - iter 4/48 - loss 0.02264444 - time (sec): 2.42 - samples/sec: 1358.51 - lr: 0.000006 - momentum: 0.000000
2024-03-26 11:37:22,976 epoch 10 - iter 8/48 - loss 0.01694052 - time (sec): 4.38 - samples/sec: 1333.02 - lr: 0.000005 - momentum: 0.000000
2024-03-26 11:37:24,212 epoch 10 - iter 12/48 - loss 0.02510073 - time (sec): 5.62 - samples/sec: 1484.16 - lr: 0.000005 - momentum: 0.000000
2024-03-26 11:37:25,824 epoch 10 - iter 16/48 - loss 0.02807504 - time (sec): 7.23 - samples/sec: 1551.36 - lr: 0.000004 - momentum: 0.000000
2024-03-26 11:37:27,540 epoch 10 - iter 20/48 - loss 0.02802998 - time (sec): 8.95 - samples/sec: 1595.24 - lr: 0.000004 - momentum: 0.000000
2024-03-26 11:37:29,677 epoch 10 - iter 24/48 - loss 0.02579472 - time (sec): 11.09 - samples/sec: 1537.39 - lr: 0.000003 - momentum: 0.000000
2024-03-26 11:37:31,826 epoch 10 - iter 28/48 - loss 0.02330368 - time (sec): 13.23 - samples/sec: 1507.27 - lr: 0.000003 - momentum: 0.000000
2024-03-26 11:37:34,002 epoch 10 - iter 32/48 - loss 0.02482044 - time (sec): 15.41 - samples/sec: 1513.99 - lr: 0.000002 - momentum: 0.000000
2024-03-26 11:37:35,409 epoch 10 - iter 36/48 - loss 0.02437032 - time (sec): 16.82 - samples/sec: 1516.24 - lr: 0.000002 - momentum: 0.000000
2024-03-26 11:37:38,098 epoch 10 - iter 40/48 - loss 0.02266370 - time (sec): 19.51 - samples/sec: 1479.35 - lr: 0.000001 - momentum: 0.000000
2024-03-26 11:37:40,623 epoch 10 - iter 44/48 - loss 0.02605707 - time (sec): 22.03 - samples/sec: 1459.52 - lr: 0.000001 - momentum: 0.000000
2024-03-26 11:37:42,293 epoch 10 - iter 48/48 - loss 0.02700708 - time (sec): 23.70 - samples/sec: 1454.42 - lr: 0.000000 - momentum: 0.000000
2024-03-26 11:37:42,293 ----------------------------------------------------------------------------------------------------
2024-03-26 11:37:42,293 EPOCH 10 done: loss 0.0270 - lr: 0.000000
2024-03-26 11:37:43,246 DEV : loss 0.17998817563056946 - f1-score (micro avg) 0.9326
2024-03-26 11:37:43,555 ----------------------------------------------------------------------------------------------------
2024-03-26 11:37:43,555 Loading model from best epoch ...
2024-03-26 11:37:44,460 SequenceTagger predicts: Dictionary with 17 tags: O, S-Unternehmen, B-Unternehmen, E-Unternehmen, I-Unternehmen, S-Auslagerung, B-Auslagerung, E-Auslagerung, I-Auslagerung, S-Ort, B-Ort, E-Ort, I-Ort, S-Software, B-Software, E-Software, I-Software
2024-03-26 11:37:45,303
Results:
- F-score (micro) 0.9052
- F-score (macro) 0.6892
- Accuracy 0.8279
By class:
precision recall f1-score support
Unternehmen 0.9027 0.8722 0.8872 266
Auslagerung 0.8775 0.8916 0.8845 249
Ort 0.9779 0.9925 0.9852 134
Software 0.0000 0.0000 0.0000 0
micro avg 0.9059 0.9045 0.9052 649
macro avg 0.6895 0.6891 0.6892 649
weighted avg 0.9086 0.9045 0.9064 649
2024-03-26 11:37:45,304 ----------------------------------------------------------------------------------------------------
|