File size: 24,022 Bytes
20910d6 |
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 |
2023-10-17 08:39:39,345 ----------------------------------------------------------------------------------------------------
2023-10-17 08:39:39,346 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): ElectraModel(
(embeddings): ElectraEmbeddings(
(word_embeddings): Embedding(32001, 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): ElectraEncoder(
(layer): ModuleList(
(0-11): 12 x ElectraLayer(
(attention): ElectraAttention(
(self): ElectraSelfAttention(
(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): ElectraSelfOutput(
(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): ElectraIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): ElectraOutput(
(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)
)
)
)
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=768, out_features=25, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-17 08:39:39,346 ----------------------------------------------------------------------------------------------------
2023-10-17 08:39:39,346 MultiCorpus: 1100 train + 206 dev + 240 test sentences
- NER_HIPE_2022 Corpus: 1100 train + 206 dev + 240 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/ajmc/de/with_doc_seperator
2023-10-17 08:39:39,346 ----------------------------------------------------------------------------------------------------
2023-10-17 08:39:39,346 Train: 1100 sentences
2023-10-17 08:39:39,346 (train_with_dev=False, train_with_test=False)
2023-10-17 08:39:39,347 ----------------------------------------------------------------------------------------------------
2023-10-17 08:39:39,347 Training Params:
2023-10-17 08:39:39,347 - learning_rate: "5e-05"
2023-10-17 08:39:39,347 - mini_batch_size: "4"
2023-10-17 08:39:39,347 - max_epochs: "10"
2023-10-17 08:39:39,347 - shuffle: "True"
2023-10-17 08:39:39,347 ----------------------------------------------------------------------------------------------------
2023-10-17 08:39:39,347 Plugins:
2023-10-17 08:39:39,347 - TensorboardLogger
2023-10-17 08:39:39,347 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 08:39:39,347 ----------------------------------------------------------------------------------------------------
2023-10-17 08:39:39,347 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 08:39:39,347 - metric: "('micro avg', 'f1-score')"
2023-10-17 08:39:39,347 ----------------------------------------------------------------------------------------------------
2023-10-17 08:39:39,347 Computation:
2023-10-17 08:39:39,347 - compute on device: cuda:0
2023-10-17 08:39:39,347 - embedding storage: none
2023-10-17 08:39:39,347 ----------------------------------------------------------------------------------------------------
2023-10-17 08:39:39,347 Model training base path: "hmbench-ajmc/de-hmteams/teams-base-historic-multilingual-discriminator-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-3"
2023-10-17 08:39:39,347 ----------------------------------------------------------------------------------------------------
2023-10-17 08:39:39,347 ----------------------------------------------------------------------------------------------------
2023-10-17 08:39:39,347 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 08:39:40,651 epoch 1 - iter 27/275 - loss 4.03878656 - time (sec): 1.30 - samples/sec: 1602.15 - lr: 0.000005 - momentum: 0.000000
2023-10-17 08:39:41,984 epoch 1 - iter 54/275 - loss 3.09313257 - time (sec): 2.64 - samples/sec: 1652.71 - lr: 0.000010 - momentum: 0.000000
2023-10-17 08:39:43,312 epoch 1 - iter 81/275 - loss 2.32201658 - time (sec): 3.96 - samples/sec: 1708.97 - lr: 0.000015 - momentum: 0.000000
2023-10-17 08:39:44,621 epoch 1 - iter 108/275 - loss 1.96172135 - time (sec): 5.27 - samples/sec: 1660.29 - lr: 0.000019 - momentum: 0.000000
2023-10-17 08:39:45,831 epoch 1 - iter 135/275 - loss 1.65705331 - time (sec): 6.48 - samples/sec: 1707.48 - lr: 0.000024 - momentum: 0.000000
2023-10-17 08:39:47,070 epoch 1 - iter 162/275 - loss 1.43741272 - time (sec): 7.72 - samples/sec: 1732.13 - lr: 0.000029 - momentum: 0.000000
2023-10-17 08:39:48,277 epoch 1 - iter 189/275 - loss 1.28970992 - time (sec): 8.93 - samples/sec: 1737.39 - lr: 0.000034 - momentum: 0.000000
2023-10-17 08:39:49,497 epoch 1 - iter 216/275 - loss 1.16239932 - time (sec): 10.15 - samples/sec: 1750.43 - lr: 0.000039 - momentum: 0.000000
2023-10-17 08:39:50,739 epoch 1 - iter 243/275 - loss 1.05732942 - time (sec): 11.39 - samples/sec: 1767.15 - lr: 0.000044 - momentum: 0.000000
2023-10-17 08:39:51,981 epoch 1 - iter 270/275 - loss 0.97265547 - time (sec): 12.63 - samples/sec: 1772.94 - lr: 0.000049 - momentum: 0.000000
2023-10-17 08:39:52,200 ----------------------------------------------------------------------------------------------------
2023-10-17 08:39:52,200 EPOCH 1 done: loss 0.9602 - lr: 0.000049
2023-10-17 08:39:52,734 DEV : loss 0.18448342382907867 - f1-score (micro avg) 0.7457
2023-10-17 08:39:52,738 saving best model
2023-10-17 08:39:53,085 ----------------------------------------------------------------------------------------------------
2023-10-17 08:39:54,298 epoch 2 - iter 27/275 - loss 0.27765461 - time (sec): 1.21 - samples/sec: 1847.98 - lr: 0.000049 - momentum: 0.000000
2023-10-17 08:39:55,529 epoch 2 - iter 54/275 - loss 0.20491030 - time (sec): 2.44 - samples/sec: 1847.56 - lr: 0.000049 - momentum: 0.000000
2023-10-17 08:39:56,740 epoch 2 - iter 81/275 - loss 0.18857691 - time (sec): 3.65 - samples/sec: 1881.68 - lr: 0.000048 - momentum: 0.000000
2023-10-17 08:39:57,971 epoch 2 - iter 108/275 - loss 0.18384314 - time (sec): 4.88 - samples/sec: 1863.64 - lr: 0.000048 - momentum: 0.000000
2023-10-17 08:39:59,237 epoch 2 - iter 135/275 - loss 0.19785182 - time (sec): 6.15 - samples/sec: 1883.70 - lr: 0.000047 - momentum: 0.000000
2023-10-17 08:40:00,460 epoch 2 - iter 162/275 - loss 0.18989803 - time (sec): 7.37 - samples/sec: 1867.37 - lr: 0.000047 - momentum: 0.000000
2023-10-17 08:40:01,679 epoch 2 - iter 189/275 - loss 0.17923154 - time (sec): 8.59 - samples/sec: 1851.17 - lr: 0.000046 - momentum: 0.000000
2023-10-17 08:40:02,886 epoch 2 - iter 216/275 - loss 0.16990514 - time (sec): 9.80 - samples/sec: 1844.09 - lr: 0.000046 - momentum: 0.000000
2023-10-17 08:40:04,108 epoch 2 - iter 243/275 - loss 0.16846483 - time (sec): 11.02 - samples/sec: 1828.04 - lr: 0.000045 - momentum: 0.000000
2023-10-17 08:40:05,323 epoch 2 - iter 270/275 - loss 0.16677599 - time (sec): 12.24 - samples/sec: 1833.14 - lr: 0.000045 - momentum: 0.000000
2023-10-17 08:40:05,547 ----------------------------------------------------------------------------------------------------
2023-10-17 08:40:05,547 EPOCH 2 done: loss 0.1650 - lr: 0.000045
2023-10-17 08:40:06,199 DEV : loss 0.18307897448539734 - f1-score (micro avg) 0.7929
2023-10-17 08:40:06,204 saving best model
2023-10-17 08:40:06,658 ----------------------------------------------------------------------------------------------------
2023-10-17 08:40:07,957 epoch 3 - iter 27/275 - loss 0.10069768 - time (sec): 1.30 - samples/sec: 1916.69 - lr: 0.000044 - momentum: 0.000000
2023-10-17 08:40:09,194 epoch 3 - iter 54/275 - loss 0.10050973 - time (sec): 2.53 - samples/sec: 1883.93 - lr: 0.000043 - momentum: 0.000000
2023-10-17 08:40:10,426 epoch 3 - iter 81/275 - loss 0.09155482 - time (sec): 3.77 - samples/sec: 1866.60 - lr: 0.000043 - momentum: 0.000000
2023-10-17 08:40:11,659 epoch 3 - iter 108/275 - loss 0.08904290 - time (sec): 5.00 - samples/sec: 1897.14 - lr: 0.000042 - momentum: 0.000000
2023-10-17 08:40:12,873 epoch 3 - iter 135/275 - loss 0.08844808 - time (sec): 6.21 - samples/sec: 1846.47 - lr: 0.000042 - momentum: 0.000000
2023-10-17 08:40:14,099 epoch 3 - iter 162/275 - loss 0.09007275 - time (sec): 7.44 - samples/sec: 1837.24 - lr: 0.000041 - momentum: 0.000000
2023-10-17 08:40:15,314 epoch 3 - iter 189/275 - loss 0.09894026 - time (sec): 8.65 - samples/sec: 1838.92 - lr: 0.000041 - momentum: 0.000000
2023-10-17 08:40:16,541 epoch 3 - iter 216/275 - loss 0.10190647 - time (sec): 9.88 - samples/sec: 1840.60 - lr: 0.000040 - momentum: 0.000000
2023-10-17 08:40:17,747 epoch 3 - iter 243/275 - loss 0.10769110 - time (sec): 11.09 - samples/sec: 1835.25 - lr: 0.000040 - momentum: 0.000000
2023-10-17 08:40:18,964 epoch 3 - iter 270/275 - loss 0.10706840 - time (sec): 12.30 - samples/sec: 1823.91 - lr: 0.000039 - momentum: 0.000000
2023-10-17 08:40:19,187 ----------------------------------------------------------------------------------------------------
2023-10-17 08:40:19,188 EPOCH 3 done: loss 0.1065 - lr: 0.000039
2023-10-17 08:40:19,998 DEV : loss 0.18310178816318512 - f1-score (micro avg) 0.8514
2023-10-17 08:40:20,002 saving best model
2023-10-17 08:40:20,437 ----------------------------------------------------------------------------------------------------
2023-10-17 08:40:21,645 epoch 4 - iter 27/275 - loss 0.05756256 - time (sec): 1.21 - samples/sec: 1794.76 - lr: 0.000038 - momentum: 0.000000
2023-10-17 08:40:22,858 epoch 4 - iter 54/275 - loss 0.05243097 - time (sec): 2.42 - samples/sec: 1833.72 - lr: 0.000038 - momentum: 0.000000
2023-10-17 08:40:24,061 epoch 4 - iter 81/275 - loss 0.05875408 - time (sec): 3.62 - samples/sec: 1831.26 - lr: 0.000037 - momentum: 0.000000
2023-10-17 08:40:25,273 epoch 4 - iter 108/275 - loss 0.05543076 - time (sec): 4.83 - samples/sec: 1783.41 - lr: 0.000037 - momentum: 0.000000
2023-10-17 08:40:26,480 epoch 4 - iter 135/275 - loss 0.06963521 - time (sec): 6.04 - samples/sec: 1806.84 - lr: 0.000036 - momentum: 0.000000
2023-10-17 08:40:27,712 epoch 4 - iter 162/275 - loss 0.07078516 - time (sec): 7.27 - samples/sec: 1816.65 - lr: 0.000036 - momentum: 0.000000
2023-10-17 08:40:28,939 epoch 4 - iter 189/275 - loss 0.07629535 - time (sec): 8.50 - samples/sec: 1793.65 - lr: 0.000035 - momentum: 0.000000
2023-10-17 08:40:30,163 epoch 4 - iter 216/275 - loss 0.08775086 - time (sec): 9.72 - samples/sec: 1822.76 - lr: 0.000035 - momentum: 0.000000
2023-10-17 08:40:31,360 epoch 4 - iter 243/275 - loss 0.08837665 - time (sec): 10.92 - samples/sec: 1825.54 - lr: 0.000034 - momentum: 0.000000
2023-10-17 08:40:32,593 epoch 4 - iter 270/275 - loss 0.08667772 - time (sec): 12.15 - samples/sec: 1836.66 - lr: 0.000034 - momentum: 0.000000
2023-10-17 08:40:32,816 ----------------------------------------------------------------------------------------------------
2023-10-17 08:40:32,816 EPOCH 4 done: loss 0.0867 - lr: 0.000034
2023-10-17 08:40:33,475 DEV : loss 0.21343372762203217 - f1-score (micro avg) 0.8759
2023-10-17 08:40:33,479 saving best model
2023-10-17 08:40:33,906 ----------------------------------------------------------------------------------------------------
2023-10-17 08:40:35,101 epoch 5 - iter 27/275 - loss 0.09904212 - time (sec): 1.19 - samples/sec: 1928.97 - lr: 0.000033 - momentum: 0.000000
2023-10-17 08:40:36,312 epoch 5 - iter 54/275 - loss 0.08332440 - time (sec): 2.40 - samples/sec: 1929.05 - lr: 0.000032 - momentum: 0.000000
2023-10-17 08:40:37,498 epoch 5 - iter 81/275 - loss 0.07371610 - time (sec): 3.59 - samples/sec: 1906.32 - lr: 0.000032 - momentum: 0.000000
2023-10-17 08:40:38,715 epoch 5 - iter 108/275 - loss 0.08587081 - time (sec): 4.81 - samples/sec: 1876.77 - lr: 0.000031 - momentum: 0.000000
2023-10-17 08:40:39,919 epoch 5 - iter 135/275 - loss 0.08650298 - time (sec): 6.01 - samples/sec: 1876.23 - lr: 0.000031 - momentum: 0.000000
2023-10-17 08:40:41,117 epoch 5 - iter 162/275 - loss 0.08560863 - time (sec): 7.21 - samples/sec: 1867.28 - lr: 0.000030 - momentum: 0.000000
2023-10-17 08:40:42,352 epoch 5 - iter 189/275 - loss 0.08186266 - time (sec): 8.44 - samples/sec: 1869.53 - lr: 0.000030 - momentum: 0.000000
2023-10-17 08:40:43,582 epoch 5 - iter 216/275 - loss 0.07453600 - time (sec): 9.67 - samples/sec: 1870.31 - lr: 0.000029 - momentum: 0.000000
2023-10-17 08:40:44,819 epoch 5 - iter 243/275 - loss 0.07028006 - time (sec): 10.91 - samples/sec: 1859.96 - lr: 0.000029 - momentum: 0.000000
2023-10-17 08:40:46,043 epoch 5 - iter 270/275 - loss 0.06854325 - time (sec): 12.14 - samples/sec: 1843.18 - lr: 0.000028 - momentum: 0.000000
2023-10-17 08:40:46,271 ----------------------------------------------------------------------------------------------------
2023-10-17 08:40:46,271 EPOCH 5 done: loss 0.0673 - lr: 0.000028
2023-10-17 08:40:46,938 DEV : loss 0.20947176218032837 - f1-score (micro avg) 0.8504
2023-10-17 08:40:46,943 ----------------------------------------------------------------------------------------------------
2023-10-17 08:40:48,187 epoch 6 - iter 27/275 - loss 0.05511466 - time (sec): 1.24 - samples/sec: 1842.84 - lr: 0.000027 - momentum: 0.000000
2023-10-17 08:40:49,440 epoch 6 - iter 54/275 - loss 0.05118589 - time (sec): 2.50 - samples/sec: 1787.99 - lr: 0.000027 - momentum: 0.000000
2023-10-17 08:40:50,650 epoch 6 - iter 81/275 - loss 0.05692708 - time (sec): 3.71 - samples/sec: 1763.66 - lr: 0.000026 - momentum: 0.000000
2023-10-17 08:40:51,852 epoch 6 - iter 108/275 - loss 0.06210658 - time (sec): 4.91 - samples/sec: 1755.23 - lr: 0.000026 - momentum: 0.000000
2023-10-17 08:40:53,063 epoch 6 - iter 135/275 - loss 0.05927054 - time (sec): 6.12 - samples/sec: 1813.61 - lr: 0.000025 - momentum: 0.000000
2023-10-17 08:40:54,220 epoch 6 - iter 162/275 - loss 0.05347616 - time (sec): 7.28 - samples/sec: 1801.60 - lr: 0.000025 - momentum: 0.000000
2023-10-17 08:40:55,399 epoch 6 - iter 189/275 - loss 0.05304045 - time (sec): 8.46 - samples/sec: 1844.48 - lr: 0.000024 - momentum: 0.000000
2023-10-17 08:40:56,562 epoch 6 - iter 216/275 - loss 0.05186256 - time (sec): 9.62 - samples/sec: 1849.76 - lr: 0.000024 - momentum: 0.000000
2023-10-17 08:40:57,712 epoch 6 - iter 243/275 - loss 0.04991913 - time (sec): 10.77 - samples/sec: 1848.15 - lr: 0.000023 - momentum: 0.000000
2023-10-17 08:40:58,869 epoch 6 - iter 270/275 - loss 0.04731578 - time (sec): 11.93 - samples/sec: 1866.98 - lr: 0.000022 - momentum: 0.000000
2023-10-17 08:40:59,090 ----------------------------------------------------------------------------------------------------
2023-10-17 08:40:59,090 EPOCH 6 done: loss 0.0469 - lr: 0.000022
2023-10-17 08:40:59,729 DEV : loss 0.17257489264011383 - f1-score (micro avg) 0.8764
2023-10-17 08:40:59,734 saving best model
2023-10-17 08:41:00,182 ----------------------------------------------------------------------------------------------------
2023-10-17 08:41:01,421 epoch 7 - iter 27/275 - loss 0.01858912 - time (sec): 1.24 - samples/sec: 1872.12 - lr: 0.000022 - momentum: 0.000000
2023-10-17 08:41:02,633 epoch 7 - iter 54/275 - loss 0.01466161 - time (sec): 2.45 - samples/sec: 1827.37 - lr: 0.000021 - momentum: 0.000000
2023-10-17 08:41:03,852 epoch 7 - iter 81/275 - loss 0.01630909 - time (sec): 3.67 - samples/sec: 1800.60 - lr: 0.000021 - momentum: 0.000000
2023-10-17 08:41:05,148 epoch 7 - iter 108/275 - loss 0.02794978 - time (sec): 4.96 - samples/sec: 1740.67 - lr: 0.000020 - momentum: 0.000000
2023-10-17 08:41:06,405 epoch 7 - iter 135/275 - loss 0.02556503 - time (sec): 6.22 - samples/sec: 1794.75 - lr: 0.000020 - momentum: 0.000000
2023-10-17 08:41:07,676 epoch 7 - iter 162/275 - loss 0.03038199 - time (sec): 7.49 - samples/sec: 1806.56 - lr: 0.000019 - momentum: 0.000000
2023-10-17 08:41:08,890 epoch 7 - iter 189/275 - loss 0.03380410 - time (sec): 8.71 - samples/sec: 1799.57 - lr: 0.000019 - momentum: 0.000000
2023-10-17 08:41:10,121 epoch 7 - iter 216/275 - loss 0.03086095 - time (sec): 9.94 - samples/sec: 1810.37 - lr: 0.000018 - momentum: 0.000000
2023-10-17 08:41:11,366 epoch 7 - iter 243/275 - loss 0.03220375 - time (sec): 11.18 - samples/sec: 1806.21 - lr: 0.000017 - momentum: 0.000000
2023-10-17 08:41:12,567 epoch 7 - iter 270/275 - loss 0.03117750 - time (sec): 12.38 - samples/sec: 1815.30 - lr: 0.000017 - momentum: 0.000000
2023-10-17 08:41:12,788 ----------------------------------------------------------------------------------------------------
2023-10-17 08:41:12,788 EPOCH 7 done: loss 0.0328 - lr: 0.000017
2023-10-17 08:41:13,421 DEV : loss 0.16961951553821564 - f1-score (micro avg) 0.8696
2023-10-17 08:41:13,426 ----------------------------------------------------------------------------------------------------
2023-10-17 08:41:14,692 epoch 8 - iter 27/275 - loss 0.03404871 - time (sec): 1.26 - samples/sec: 1755.31 - lr: 0.000016 - momentum: 0.000000
2023-10-17 08:41:15,915 epoch 8 - iter 54/275 - loss 0.04007446 - time (sec): 2.49 - samples/sec: 1762.18 - lr: 0.000016 - momentum: 0.000000
2023-10-17 08:41:17,187 epoch 8 - iter 81/275 - loss 0.03058524 - time (sec): 3.76 - samples/sec: 1822.23 - lr: 0.000015 - momentum: 0.000000
2023-10-17 08:41:18,407 epoch 8 - iter 108/275 - loss 0.03511864 - time (sec): 4.98 - samples/sec: 1835.28 - lr: 0.000015 - momentum: 0.000000
2023-10-17 08:41:19,619 epoch 8 - iter 135/275 - loss 0.03051247 - time (sec): 6.19 - samples/sec: 1832.90 - lr: 0.000014 - momentum: 0.000000
2023-10-17 08:41:20,817 epoch 8 - iter 162/275 - loss 0.02693145 - time (sec): 7.39 - samples/sec: 1798.00 - lr: 0.000014 - momentum: 0.000000
2023-10-17 08:41:22,089 epoch 8 - iter 189/275 - loss 0.02624564 - time (sec): 8.66 - samples/sec: 1798.56 - lr: 0.000013 - momentum: 0.000000
2023-10-17 08:41:23,318 epoch 8 - iter 216/275 - loss 0.02675046 - time (sec): 9.89 - samples/sec: 1803.44 - lr: 0.000012 - momentum: 0.000000
2023-10-17 08:41:24,550 epoch 8 - iter 243/275 - loss 0.02456897 - time (sec): 11.12 - samples/sec: 1803.77 - lr: 0.000012 - momentum: 0.000000
2023-10-17 08:41:25,712 epoch 8 - iter 270/275 - loss 0.02792934 - time (sec): 12.29 - samples/sec: 1822.32 - lr: 0.000011 - momentum: 0.000000
2023-10-17 08:41:25,926 ----------------------------------------------------------------------------------------------------
2023-10-17 08:41:25,926 EPOCH 8 done: loss 0.0275 - lr: 0.000011
2023-10-17 08:41:26,560 DEV : loss 0.18767307698726654 - f1-score (micro avg) 0.8819
2023-10-17 08:41:26,565 saving best model
2023-10-17 08:41:26,999 ----------------------------------------------------------------------------------------------------
2023-10-17 08:41:28,308 epoch 9 - iter 27/275 - loss 0.01847685 - time (sec): 1.31 - samples/sec: 1706.16 - lr: 0.000011 - momentum: 0.000000
2023-10-17 08:41:29,553 epoch 9 - iter 54/275 - loss 0.02177958 - time (sec): 2.55 - samples/sec: 1832.70 - lr: 0.000010 - momentum: 0.000000
2023-10-17 08:41:30,790 epoch 9 - iter 81/275 - loss 0.01693967 - time (sec): 3.79 - samples/sec: 1789.02 - lr: 0.000010 - momentum: 0.000000
2023-10-17 08:41:32,050 epoch 9 - iter 108/275 - loss 0.01773225 - time (sec): 5.05 - samples/sec: 1776.39 - lr: 0.000009 - momentum: 0.000000
2023-10-17 08:41:33,286 epoch 9 - iter 135/275 - loss 0.01737827 - time (sec): 6.28 - samples/sec: 1794.29 - lr: 0.000009 - momentum: 0.000000
2023-10-17 08:41:34,507 epoch 9 - iter 162/275 - loss 0.01537384 - time (sec): 7.50 - samples/sec: 1796.61 - lr: 0.000008 - momentum: 0.000000
2023-10-17 08:41:35,743 epoch 9 - iter 189/275 - loss 0.01914939 - time (sec): 8.74 - samples/sec: 1794.42 - lr: 0.000007 - momentum: 0.000000
2023-10-17 08:41:36,975 epoch 9 - iter 216/275 - loss 0.01928300 - time (sec): 9.97 - samples/sec: 1800.75 - lr: 0.000007 - momentum: 0.000000
2023-10-17 08:41:38,189 epoch 9 - iter 243/275 - loss 0.01786580 - time (sec): 11.19 - samples/sec: 1788.11 - lr: 0.000006 - momentum: 0.000000
2023-10-17 08:41:39,426 epoch 9 - iter 270/275 - loss 0.01731168 - time (sec): 12.42 - samples/sec: 1796.45 - lr: 0.000006 - momentum: 0.000000
2023-10-17 08:41:39,653 ----------------------------------------------------------------------------------------------------
2023-10-17 08:41:39,653 EPOCH 9 done: loss 0.0170 - lr: 0.000006
2023-10-17 08:41:40,301 DEV : loss 0.1904018521308899 - f1-score (micro avg) 0.8759
2023-10-17 08:41:40,306 ----------------------------------------------------------------------------------------------------
2023-10-17 08:41:41,562 epoch 10 - iter 27/275 - loss 0.03813907 - time (sec): 1.25 - samples/sec: 2061.13 - lr: 0.000005 - momentum: 0.000000
2023-10-17 08:41:42,792 epoch 10 - iter 54/275 - loss 0.02883733 - time (sec): 2.48 - samples/sec: 2002.02 - lr: 0.000005 - momentum: 0.000000
2023-10-17 08:41:44,014 epoch 10 - iter 81/275 - loss 0.02220392 - time (sec): 3.71 - samples/sec: 1992.70 - lr: 0.000004 - momentum: 0.000000
2023-10-17 08:41:45,249 epoch 10 - iter 108/275 - loss 0.02022173 - time (sec): 4.94 - samples/sec: 1911.67 - lr: 0.000004 - momentum: 0.000000
2023-10-17 08:41:46,451 epoch 10 - iter 135/275 - loss 0.01751670 - time (sec): 6.14 - samples/sec: 1884.10 - lr: 0.000003 - momentum: 0.000000
2023-10-17 08:41:47,677 epoch 10 - iter 162/275 - loss 0.01868292 - time (sec): 7.37 - samples/sec: 1861.62 - lr: 0.000002 - momentum: 0.000000
2023-10-17 08:41:48,901 epoch 10 - iter 189/275 - loss 0.01721044 - time (sec): 8.59 - samples/sec: 1841.39 - lr: 0.000002 - momentum: 0.000000
2023-10-17 08:41:50,124 epoch 10 - iter 216/275 - loss 0.01701337 - time (sec): 9.82 - samples/sec: 1826.79 - lr: 0.000001 - momentum: 0.000000
2023-10-17 08:41:51,282 epoch 10 - iter 243/275 - loss 0.01725403 - time (sec): 10.97 - samples/sec: 1844.54 - lr: 0.000001 - momentum: 0.000000
2023-10-17 08:41:52,480 epoch 10 - iter 270/275 - loss 0.01637470 - time (sec): 12.17 - samples/sec: 1837.19 - lr: 0.000000 - momentum: 0.000000
2023-10-17 08:41:52,705 ----------------------------------------------------------------------------------------------------
2023-10-17 08:41:52,705 EPOCH 10 done: loss 0.0164 - lr: 0.000000
2023-10-17 08:41:53,342 DEV : loss 0.1948593109846115 - f1-score (micro avg) 0.8703
2023-10-17 08:41:53,711 ----------------------------------------------------------------------------------------------------
2023-10-17 08:41:53,712 Loading model from best epoch ...
2023-10-17 08:41:55,054 SequenceTagger predicts: Dictionary with 25 tags: O, S-scope, B-scope, E-scope, I-scope, S-pers, B-pers, E-pers, I-pers, S-work, B-work, E-work, I-work, S-loc, B-loc, E-loc, I-loc, S-object, B-object, E-object, I-object, S-date, B-date, E-date, I-date
2023-10-17 08:41:55,885
Results:
- F-score (micro) 0.8924
- F-score (macro) 0.8645
- Accuracy 0.8175
By class:
precision recall f1-score support
scope 0.8895 0.8693 0.8793 176
pers 0.9754 0.9297 0.9520 128
work 0.8243 0.8243 0.8243 74
object 1.0000 1.0000 1.0000 2
loc 1.0000 0.5000 0.6667 2
micro avg 0.9057 0.8796 0.8924 382
macro avg 0.9379 0.8247 0.8645 382
weighted avg 0.9068 0.8796 0.8925 382
2023-10-17 08:41:55,885 ----------------------------------------------------------------------------------------------------
|