File size: 26,652 Bytes
e85e689 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 |
2024-03-26 10:20:21,938 ----------------------------------------------------------------------------------------------------
2024-03-26 10:20:21,938 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): BertModel(
(embeddings): BertEmbeddings(
(word_embeddings): Embedding(31103, 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 10:20:21,938 ----------------------------------------------------------------------------------------------------
2024-03-26 10:20:21,938 Corpus: 758 train + 94 dev + 96 test sentences
2024-03-26 10:20:21,938 ----------------------------------------------------------------------------------------------------
2024-03-26 10:20:21,938 Train: 758 sentences
2024-03-26 10:20:21,938 (train_with_dev=False, train_with_test=False)
2024-03-26 10:20:21,938 ----------------------------------------------------------------------------------------------------
2024-03-26 10:20:21,938 Training Params:
2024-03-26 10:20:21,938 - learning_rate: "5e-05"
2024-03-26 10:20:21,938 - mini_batch_size: "16"
2024-03-26 10:20:21,938 - max_epochs: "10"
2024-03-26 10:20:21,938 - shuffle: "True"
2024-03-26 10:20:21,938 ----------------------------------------------------------------------------------------------------
2024-03-26 10:20:21,938 Plugins:
2024-03-26 10:20:21,938 - TensorboardLogger
2024-03-26 10:20:21,939 - LinearScheduler | warmup_fraction: '0.1'
2024-03-26 10:20:21,939 ----------------------------------------------------------------------------------------------------
2024-03-26 10:20:21,939 Final evaluation on model from best epoch (best-model.pt)
2024-03-26 10:20:21,939 - metric: "('micro avg', 'f1-score')"
2024-03-26 10:20:21,939 ----------------------------------------------------------------------------------------------------
2024-03-26 10:20:21,939 Computation:
2024-03-26 10:20:21,939 - compute on device: cuda:0
2024-03-26 10:20:21,939 - embedding storage: none
2024-03-26 10:20:21,939 ----------------------------------------------------------------------------------------------------
2024-03-26 10:20:21,939 Model training base path: "flair-co-funer-gbert_base-bs16-e10-lr5e-05-4"
2024-03-26 10:20:21,939 ----------------------------------------------------------------------------------------------------
2024-03-26 10:20:21,939 ----------------------------------------------------------------------------------------------------
2024-03-26 10:20:21,939 Logging anything other than scalars to TensorBoard is currently not supported.
2024-03-26 10:20:23,389 epoch 1 - iter 4/48 - loss 3.33143156 - time (sec): 1.45 - samples/sec: 1800.62 - lr: 0.000003 - momentum: 0.000000
2024-03-26 10:20:25,188 epoch 1 - iter 8/48 - loss 3.24253277 - time (sec): 3.25 - samples/sec: 1576.87 - lr: 0.000007 - momentum: 0.000000
2024-03-26 10:20:26,510 epoch 1 - iter 12/48 - loss 3.09296223 - time (sec): 4.57 - samples/sec: 1597.62 - lr: 0.000011 - momentum: 0.000000
2024-03-26 10:20:29,021 epoch 1 - iter 16/48 - loss 2.83159528 - time (sec): 7.08 - samples/sec: 1510.67 - lr: 0.000016 - momentum: 0.000000
2024-03-26 10:20:31,108 epoch 1 - iter 20/48 - loss 2.68393125 - time (sec): 9.17 - samples/sec: 1493.92 - lr: 0.000020 - momentum: 0.000000
2024-03-26 10:20:33,765 epoch 1 - iter 24/48 - loss 2.52199527 - time (sec): 11.83 - samples/sec: 1431.14 - lr: 0.000024 - momentum: 0.000000
2024-03-26 10:20:36,248 epoch 1 - iter 28/48 - loss 2.39681153 - time (sec): 14.31 - samples/sec: 1417.79 - lr: 0.000028 - momentum: 0.000000
2024-03-26 10:20:38,112 epoch 1 - iter 32/48 - loss 2.30150701 - time (sec): 16.17 - samples/sec: 1414.35 - lr: 0.000032 - momentum: 0.000000
2024-03-26 10:20:38,990 epoch 1 - iter 36/48 - loss 2.22901668 - time (sec): 17.05 - samples/sec: 1465.06 - lr: 0.000036 - momentum: 0.000000
2024-03-26 10:20:40,843 epoch 1 - iter 40/48 - loss 2.11985964 - time (sec): 18.90 - samples/sec: 1473.35 - lr: 0.000041 - momentum: 0.000000
2024-03-26 10:20:42,869 epoch 1 - iter 44/48 - loss 1.99600231 - time (sec): 20.93 - samples/sec: 1492.07 - lr: 0.000045 - momentum: 0.000000
2024-03-26 10:20:44,580 epoch 1 - iter 48/48 - loss 1.90143416 - time (sec): 22.64 - samples/sec: 1522.55 - lr: 0.000049 - momentum: 0.000000
2024-03-26 10:20:44,580 ----------------------------------------------------------------------------------------------------
2024-03-26 10:20:44,580 EPOCH 1 done: loss 1.9014 - lr: 0.000049
2024-03-26 10:20:45,388 DEV : loss 0.5843583941459656 - f1-score (micro avg) 0.6105
2024-03-26 10:20:45,389 saving best model
2024-03-26 10:20:45,671 ----------------------------------------------------------------------------------------------------
2024-03-26 10:20:46,897 epoch 2 - iter 4/48 - loss 0.87169473 - time (sec): 1.23 - samples/sec: 1931.82 - lr: 0.000050 - momentum: 0.000000
2024-03-26 10:20:49,128 epoch 2 - iter 8/48 - loss 0.67248507 - time (sec): 3.46 - samples/sec: 1578.54 - lr: 0.000049 - momentum: 0.000000
2024-03-26 10:20:50,900 epoch 2 - iter 12/48 - loss 0.64179862 - time (sec): 5.23 - samples/sec: 1629.89 - lr: 0.000049 - momentum: 0.000000
2024-03-26 10:20:53,272 epoch 2 - iter 16/48 - loss 0.57875086 - time (sec): 7.60 - samples/sec: 1484.27 - lr: 0.000048 - momentum: 0.000000
2024-03-26 10:20:56,664 epoch 2 - iter 20/48 - loss 0.52549505 - time (sec): 10.99 - samples/sec: 1342.28 - lr: 0.000048 - momentum: 0.000000
2024-03-26 10:20:58,142 epoch 2 - iter 24/48 - loss 0.52600785 - time (sec): 12.47 - samples/sec: 1398.75 - lr: 0.000047 - momentum: 0.000000
2024-03-26 10:21:00,781 epoch 2 - iter 28/48 - loss 0.51174459 - time (sec): 15.11 - samples/sec: 1370.32 - lr: 0.000047 - momentum: 0.000000
2024-03-26 10:21:03,461 epoch 2 - iter 32/48 - loss 0.49143427 - time (sec): 17.79 - samples/sec: 1372.11 - lr: 0.000046 - momentum: 0.000000
2024-03-26 10:21:05,537 epoch 2 - iter 36/48 - loss 0.48842538 - time (sec): 19.87 - samples/sec: 1361.31 - lr: 0.000046 - momentum: 0.000000
2024-03-26 10:21:08,002 epoch 2 - iter 40/48 - loss 0.47106482 - time (sec): 22.33 - samples/sec: 1351.68 - lr: 0.000046 - momentum: 0.000000
2024-03-26 10:21:09,051 epoch 2 - iter 44/48 - loss 0.46284650 - time (sec): 23.38 - samples/sec: 1386.82 - lr: 0.000045 - momentum: 0.000000
2024-03-26 10:21:10,205 epoch 2 - iter 48/48 - loss 0.45432071 - time (sec): 24.53 - samples/sec: 1405.07 - lr: 0.000045 - momentum: 0.000000
2024-03-26 10:21:10,206 ----------------------------------------------------------------------------------------------------
2024-03-26 10:21:10,206 EPOCH 2 done: loss 0.4543 - lr: 0.000045
2024-03-26 10:21:11,095 DEV : loss 0.27210864424705505 - f1-score (micro avg) 0.8439
2024-03-26 10:21:11,096 saving best model
2024-03-26 10:21:11,525 ----------------------------------------------------------------------------------------------------
2024-03-26 10:21:13,505 epoch 3 - iter 4/48 - loss 0.28056207 - time (sec): 1.98 - samples/sec: 1240.70 - lr: 0.000044 - momentum: 0.000000
2024-03-26 10:21:15,052 epoch 3 - iter 8/48 - loss 0.23255003 - time (sec): 3.53 - samples/sec: 1358.59 - lr: 0.000044 - momentum: 0.000000
2024-03-26 10:21:17,622 epoch 3 - iter 12/48 - loss 0.23329462 - time (sec): 6.10 - samples/sec: 1276.30 - lr: 0.000043 - momentum: 0.000000
2024-03-26 10:21:19,629 epoch 3 - iter 16/48 - loss 0.23716737 - time (sec): 8.10 - samples/sec: 1316.13 - lr: 0.000043 - momentum: 0.000000
2024-03-26 10:21:21,510 epoch 3 - iter 20/48 - loss 0.23579224 - time (sec): 9.98 - samples/sec: 1386.82 - lr: 0.000042 - momentum: 0.000000
2024-03-26 10:21:23,730 epoch 3 - iter 24/48 - loss 0.23278104 - time (sec): 12.20 - samples/sec: 1400.92 - lr: 0.000042 - momentum: 0.000000
2024-03-26 10:21:26,190 epoch 3 - iter 28/48 - loss 0.22345605 - time (sec): 14.66 - samples/sec: 1360.25 - lr: 0.000041 - momentum: 0.000000
2024-03-26 10:21:28,736 epoch 3 - iter 32/48 - loss 0.21820840 - time (sec): 17.21 - samples/sec: 1336.04 - lr: 0.000041 - momentum: 0.000000
2024-03-26 10:21:30,845 epoch 3 - iter 36/48 - loss 0.21675927 - time (sec): 19.32 - samples/sec: 1340.02 - lr: 0.000040 - momentum: 0.000000
2024-03-26 10:21:33,145 epoch 3 - iter 40/48 - loss 0.22341811 - time (sec): 21.62 - samples/sec: 1355.58 - lr: 0.000040 - momentum: 0.000000
2024-03-26 10:21:35,667 epoch 3 - iter 44/48 - loss 0.21735862 - time (sec): 24.14 - samples/sec: 1338.36 - lr: 0.000040 - momentum: 0.000000
2024-03-26 10:21:37,169 epoch 3 - iter 48/48 - loss 0.21861097 - time (sec): 25.64 - samples/sec: 1344.30 - lr: 0.000039 - momentum: 0.000000
2024-03-26 10:21:37,170 ----------------------------------------------------------------------------------------------------
2024-03-26 10:21:37,170 EPOCH 3 done: loss 0.2186 - lr: 0.000039
2024-03-26 10:21:38,088 DEV : loss 0.1997971087694168 - f1-score (micro avg) 0.8749
2024-03-26 10:21:38,089 saving best model
2024-03-26 10:21:38,536 ----------------------------------------------------------------------------------------------------
2024-03-26 10:21:41,519 epoch 4 - iter 4/48 - loss 0.09551912 - time (sec): 2.98 - samples/sec: 1223.05 - lr: 0.000039 - momentum: 0.000000
2024-03-26 10:21:42,899 epoch 4 - iter 8/48 - loss 0.11646094 - time (sec): 4.36 - samples/sec: 1348.52 - lr: 0.000038 - momentum: 0.000000
2024-03-26 10:21:44,963 epoch 4 - iter 12/48 - loss 0.12918594 - time (sec): 6.42 - samples/sec: 1435.71 - lr: 0.000038 - momentum: 0.000000
2024-03-26 10:21:47,488 epoch 4 - iter 16/48 - loss 0.13063851 - time (sec): 8.95 - samples/sec: 1361.30 - lr: 0.000037 - momentum: 0.000000
2024-03-26 10:21:48,468 epoch 4 - iter 20/48 - loss 0.12827349 - time (sec): 9.93 - samples/sec: 1446.70 - lr: 0.000037 - momentum: 0.000000
2024-03-26 10:21:49,848 epoch 4 - iter 24/48 - loss 0.13092356 - time (sec): 11.31 - samples/sec: 1494.57 - lr: 0.000036 - momentum: 0.000000
2024-03-26 10:21:52,936 epoch 4 - iter 28/48 - loss 0.12716420 - time (sec): 14.40 - samples/sec: 1401.82 - lr: 0.000036 - momentum: 0.000000
2024-03-26 10:21:55,392 epoch 4 - iter 32/48 - loss 0.13483254 - time (sec): 16.85 - samples/sec: 1395.38 - lr: 0.000035 - momentum: 0.000000
2024-03-26 10:21:56,884 epoch 4 - iter 36/48 - loss 0.13516354 - time (sec): 18.35 - samples/sec: 1432.40 - lr: 0.000035 - momentum: 0.000000
2024-03-26 10:21:58,842 epoch 4 - iter 40/48 - loss 0.13302960 - time (sec): 20.30 - samples/sec: 1448.04 - lr: 0.000034 - momentum: 0.000000
2024-03-26 10:22:00,728 epoch 4 - iter 44/48 - loss 0.13158699 - time (sec): 22.19 - samples/sec: 1461.65 - lr: 0.000034 - momentum: 0.000000
2024-03-26 10:22:01,761 epoch 4 - iter 48/48 - loss 0.13451918 - time (sec): 23.22 - samples/sec: 1484.41 - lr: 0.000034 - momentum: 0.000000
2024-03-26 10:22:01,761 ----------------------------------------------------------------------------------------------------
2024-03-26 10:22:01,761 EPOCH 4 done: loss 0.1345 - lr: 0.000034
2024-03-26 10:22:02,660 DEV : loss 0.1830204278230667 - f1-score (micro avg) 0.8992
2024-03-26 10:22:02,661 saving best model
2024-03-26 10:22:03,103 ----------------------------------------------------------------------------------------------------
2024-03-26 10:22:04,153 epoch 5 - iter 4/48 - loss 0.15872146 - time (sec): 1.05 - samples/sec: 2426.35 - lr: 0.000033 - momentum: 0.000000
2024-03-26 10:22:06,022 epoch 5 - iter 8/48 - loss 0.14255133 - time (sec): 2.92 - samples/sec: 1776.12 - lr: 0.000033 - momentum: 0.000000
2024-03-26 10:22:08,118 epoch 5 - iter 12/48 - loss 0.12951855 - time (sec): 5.01 - samples/sec: 1596.60 - lr: 0.000032 - momentum: 0.000000
2024-03-26 10:22:10,380 epoch 5 - iter 16/48 - loss 0.12352511 - time (sec): 7.27 - samples/sec: 1524.70 - lr: 0.000032 - momentum: 0.000000
2024-03-26 10:22:12,623 epoch 5 - iter 20/48 - loss 0.11945075 - time (sec): 9.52 - samples/sec: 1437.80 - lr: 0.000031 - momentum: 0.000000
2024-03-26 10:22:14,779 epoch 5 - iter 24/48 - loss 0.11337964 - time (sec): 11.67 - samples/sec: 1455.39 - lr: 0.000031 - momentum: 0.000000
2024-03-26 10:22:16,378 epoch 5 - iter 28/48 - loss 0.11046502 - time (sec): 13.27 - samples/sec: 1482.93 - lr: 0.000030 - momentum: 0.000000
2024-03-26 10:22:18,466 epoch 5 - iter 32/48 - loss 0.10254048 - time (sec): 15.36 - samples/sec: 1503.92 - lr: 0.000030 - momentum: 0.000000
2024-03-26 10:22:19,855 epoch 5 - iter 36/48 - loss 0.09999268 - time (sec): 16.75 - samples/sec: 1528.35 - lr: 0.000029 - momentum: 0.000000
2024-03-26 10:22:22,382 epoch 5 - iter 40/48 - loss 0.09503797 - time (sec): 19.28 - samples/sec: 1495.46 - lr: 0.000029 - momentum: 0.000000
2024-03-26 10:22:25,295 epoch 5 - iter 44/48 - loss 0.09353316 - time (sec): 22.19 - samples/sec: 1443.35 - lr: 0.000029 - momentum: 0.000000
2024-03-26 10:22:26,793 epoch 5 - iter 48/48 - loss 0.09494866 - time (sec): 23.69 - samples/sec: 1455.28 - lr: 0.000028 - momentum: 0.000000
2024-03-26 10:22:26,793 ----------------------------------------------------------------------------------------------------
2024-03-26 10:22:26,793 EPOCH 5 done: loss 0.0949 - lr: 0.000028
2024-03-26 10:22:27,682 DEV : loss 0.15144632756710052 - f1-score (micro avg) 0.9025
2024-03-26 10:22:27,683 saving best model
2024-03-26 10:22:28,118 ----------------------------------------------------------------------------------------------------
2024-03-26 10:22:29,963 epoch 6 - iter 4/48 - loss 0.10851113 - time (sec): 1.84 - samples/sec: 1594.45 - lr: 0.000028 - momentum: 0.000000
2024-03-26 10:22:31,670 epoch 6 - iter 8/48 - loss 0.09307528 - time (sec): 3.55 - samples/sec: 1633.20 - lr: 0.000027 - momentum: 0.000000
2024-03-26 10:22:33,956 epoch 6 - iter 12/48 - loss 0.08665395 - time (sec): 5.84 - samples/sec: 1511.12 - lr: 0.000027 - momentum: 0.000000
2024-03-26 10:22:35,512 epoch 6 - iter 16/48 - loss 0.08378454 - time (sec): 7.39 - samples/sec: 1533.01 - lr: 0.000026 - momentum: 0.000000
2024-03-26 10:22:38,049 epoch 6 - iter 20/48 - loss 0.07627659 - time (sec): 9.93 - samples/sec: 1446.92 - lr: 0.000026 - momentum: 0.000000
2024-03-26 10:22:40,084 epoch 6 - iter 24/48 - loss 0.07713601 - time (sec): 11.96 - samples/sec: 1461.94 - lr: 0.000025 - momentum: 0.000000
2024-03-26 10:22:42,691 epoch 6 - iter 28/48 - loss 0.07848976 - time (sec): 14.57 - samples/sec: 1436.72 - lr: 0.000025 - momentum: 0.000000
2024-03-26 10:22:44,719 epoch 6 - iter 32/48 - loss 0.07641593 - time (sec): 16.60 - samples/sec: 1415.80 - lr: 0.000024 - momentum: 0.000000
2024-03-26 10:22:45,816 epoch 6 - iter 36/48 - loss 0.07605365 - time (sec): 17.70 - samples/sec: 1465.91 - lr: 0.000024 - momentum: 0.000000
2024-03-26 10:22:47,991 epoch 6 - iter 40/48 - loss 0.07646544 - time (sec): 19.87 - samples/sec: 1455.23 - lr: 0.000023 - momentum: 0.000000
2024-03-26 10:22:49,584 epoch 6 - iter 44/48 - loss 0.08001823 - time (sec): 21.46 - samples/sec: 1479.31 - lr: 0.000023 - momentum: 0.000000
2024-03-26 10:22:51,349 epoch 6 - iter 48/48 - loss 0.07755461 - time (sec): 23.23 - samples/sec: 1484.01 - lr: 0.000023 - momentum: 0.000000
2024-03-26 10:22:51,349 ----------------------------------------------------------------------------------------------------
2024-03-26 10:22:51,349 EPOCH 6 done: loss 0.0776 - lr: 0.000023
2024-03-26 10:22:52,261 DEV : loss 0.1642305999994278 - f1-score (micro avg) 0.9209
2024-03-26 10:22:52,262 saving best model
2024-03-26 10:22:52,690 ----------------------------------------------------------------------------------------------------
2024-03-26 10:22:54,223 epoch 7 - iter 4/48 - loss 0.05032160 - time (sec): 1.53 - samples/sec: 1827.49 - lr: 0.000022 - momentum: 0.000000
2024-03-26 10:22:56,327 epoch 7 - iter 8/48 - loss 0.04567333 - time (sec): 3.64 - samples/sec: 1683.33 - lr: 0.000022 - momentum: 0.000000
2024-03-26 10:22:58,559 epoch 7 - iter 12/48 - loss 0.04588195 - time (sec): 5.87 - samples/sec: 1501.40 - lr: 0.000021 - momentum: 0.000000
2024-03-26 10:22:59,732 epoch 7 - iter 16/48 - loss 0.05196772 - time (sec): 7.04 - samples/sec: 1598.72 - lr: 0.000021 - momentum: 0.000000
2024-03-26 10:23:01,851 epoch 7 - iter 20/48 - loss 0.05230221 - time (sec): 9.16 - samples/sec: 1566.06 - lr: 0.000020 - momentum: 0.000000
2024-03-26 10:23:03,370 epoch 7 - iter 24/48 - loss 0.05050299 - time (sec): 10.68 - samples/sec: 1611.85 - lr: 0.000020 - momentum: 0.000000
2024-03-26 10:23:05,473 epoch 7 - iter 28/48 - loss 0.05048660 - time (sec): 12.78 - samples/sec: 1570.56 - lr: 0.000019 - momentum: 0.000000
2024-03-26 10:23:08,249 epoch 7 - iter 32/48 - loss 0.05200003 - time (sec): 15.56 - samples/sec: 1497.01 - lr: 0.000019 - momentum: 0.000000
2024-03-26 10:23:10,207 epoch 7 - iter 36/48 - loss 0.05120253 - time (sec): 17.52 - samples/sec: 1498.16 - lr: 0.000018 - momentum: 0.000000
2024-03-26 10:23:11,324 epoch 7 - iter 40/48 - loss 0.05496877 - time (sec): 18.63 - samples/sec: 1529.53 - lr: 0.000018 - momentum: 0.000000
2024-03-26 10:23:13,921 epoch 7 - iter 44/48 - loss 0.05649099 - time (sec): 21.23 - samples/sec: 1510.23 - lr: 0.000017 - momentum: 0.000000
2024-03-26 10:23:15,025 epoch 7 - iter 48/48 - loss 0.05715487 - time (sec): 22.33 - samples/sec: 1543.52 - lr: 0.000017 - momentum: 0.000000
2024-03-26 10:23:15,025 ----------------------------------------------------------------------------------------------------
2024-03-26 10:23:15,025 EPOCH 7 done: loss 0.0572 - lr: 0.000017
2024-03-26 10:23:15,918 DEV : loss 0.16066581010818481 - f1-score (micro avg) 0.9321
2024-03-26 10:23:15,919 saving best model
2024-03-26 10:23:16,376 ----------------------------------------------------------------------------------------------------
2024-03-26 10:23:18,477 epoch 8 - iter 4/48 - loss 0.03381401 - time (sec): 2.10 - samples/sec: 1320.91 - lr: 0.000017 - momentum: 0.000000
2024-03-26 10:23:21,156 epoch 8 - iter 8/48 - loss 0.03430206 - time (sec): 4.78 - samples/sec: 1264.11 - lr: 0.000016 - momentum: 0.000000
2024-03-26 10:23:22,808 epoch 8 - iter 12/48 - loss 0.03362778 - time (sec): 6.43 - samples/sec: 1318.90 - lr: 0.000016 - momentum: 0.000000
2024-03-26 10:23:25,407 epoch 8 - iter 16/48 - loss 0.04013356 - time (sec): 9.03 - samples/sec: 1275.00 - lr: 0.000015 - momentum: 0.000000
2024-03-26 10:23:27,037 epoch 8 - iter 20/48 - loss 0.04031454 - time (sec): 10.66 - samples/sec: 1332.95 - lr: 0.000015 - momentum: 0.000000
2024-03-26 10:23:28,493 epoch 8 - iter 24/48 - loss 0.04286096 - time (sec): 12.12 - samples/sec: 1403.05 - lr: 0.000014 - momentum: 0.000000
2024-03-26 10:23:30,353 epoch 8 - iter 28/48 - loss 0.04539810 - time (sec): 13.97 - samples/sec: 1427.05 - lr: 0.000014 - momentum: 0.000000
2024-03-26 10:23:32,986 epoch 8 - iter 32/48 - loss 0.04655029 - time (sec): 16.61 - samples/sec: 1414.48 - lr: 0.000013 - momentum: 0.000000
2024-03-26 10:23:35,375 epoch 8 - iter 36/48 - loss 0.04613594 - time (sec): 19.00 - samples/sec: 1407.20 - lr: 0.000013 - momentum: 0.000000
2024-03-26 10:23:37,561 epoch 8 - iter 40/48 - loss 0.04545780 - time (sec): 21.18 - samples/sec: 1389.03 - lr: 0.000012 - momentum: 0.000000
2024-03-26 10:23:39,791 epoch 8 - iter 44/48 - loss 0.04427927 - time (sec): 23.41 - samples/sec: 1379.75 - lr: 0.000012 - momentum: 0.000000
2024-03-26 10:23:41,345 epoch 8 - iter 48/48 - loss 0.04501008 - time (sec): 24.97 - samples/sec: 1380.70 - lr: 0.000011 - momentum: 0.000000
2024-03-26 10:23:41,345 ----------------------------------------------------------------------------------------------------
2024-03-26 10:23:41,346 EPOCH 8 done: loss 0.0450 - lr: 0.000011
2024-03-26 10:23:42,258 DEV : loss 0.15505997836589813 - f1-score (micro avg) 0.9301
2024-03-26 10:23:42,259 ----------------------------------------------------------------------------------------------------
2024-03-26 10:23:44,109 epoch 9 - iter 4/48 - loss 0.04194533 - time (sec): 1.85 - samples/sec: 1561.57 - lr: 0.000011 - momentum: 0.000000
2024-03-26 10:23:47,244 epoch 9 - iter 8/48 - loss 0.03637572 - time (sec): 4.98 - samples/sec: 1262.88 - lr: 0.000011 - momentum: 0.000000
2024-03-26 10:23:48,869 epoch 9 - iter 12/48 - loss 0.03048679 - time (sec): 6.61 - samples/sec: 1308.50 - lr: 0.000010 - momentum: 0.000000
2024-03-26 10:23:50,738 epoch 9 - iter 16/48 - loss 0.03637230 - time (sec): 8.48 - samples/sec: 1348.29 - lr: 0.000010 - momentum: 0.000000
2024-03-26 10:23:53,600 epoch 9 - iter 20/48 - loss 0.03169081 - time (sec): 11.34 - samples/sec: 1309.60 - lr: 0.000009 - momentum: 0.000000
2024-03-26 10:23:55,125 epoch 9 - iter 24/48 - loss 0.03128243 - time (sec): 12.87 - samples/sec: 1356.14 - lr: 0.000009 - momentum: 0.000000
2024-03-26 10:23:57,053 epoch 9 - iter 28/48 - loss 0.03451015 - time (sec): 14.79 - samples/sec: 1380.89 - lr: 0.000008 - momentum: 0.000000
2024-03-26 10:23:59,368 epoch 9 - iter 32/48 - loss 0.03338486 - time (sec): 17.11 - samples/sec: 1357.83 - lr: 0.000008 - momentum: 0.000000
2024-03-26 10:24:00,662 epoch 9 - iter 36/48 - loss 0.03742122 - time (sec): 18.40 - samples/sec: 1389.10 - lr: 0.000007 - momentum: 0.000000
2024-03-26 10:24:03,850 epoch 9 - iter 40/48 - loss 0.03719738 - time (sec): 21.59 - samples/sec: 1340.32 - lr: 0.000007 - momentum: 0.000000
2024-03-26 10:24:05,961 epoch 9 - iter 44/48 - loss 0.03525908 - time (sec): 23.70 - samples/sec: 1362.76 - lr: 0.000006 - momentum: 0.000000
2024-03-26 10:24:06,946 epoch 9 - iter 48/48 - loss 0.03580414 - time (sec): 24.69 - samples/sec: 1396.37 - lr: 0.000006 - momentum: 0.000000
2024-03-26 10:24:06,946 ----------------------------------------------------------------------------------------------------
2024-03-26 10:24:06,946 EPOCH 9 done: loss 0.0358 - lr: 0.000006
2024-03-26 10:24:07,845 DEV : loss 0.15631996095180511 - f1-score (micro avg) 0.9391
2024-03-26 10:24:07,847 saving best model
2024-03-26 10:24:08,273 ----------------------------------------------------------------------------------------------------
2024-03-26 10:24:10,138 epoch 10 - iter 4/48 - loss 0.04267402 - time (sec): 1.86 - samples/sec: 1387.06 - lr: 0.000006 - momentum: 0.000000
2024-03-26 10:24:12,900 epoch 10 - iter 8/48 - loss 0.02765691 - time (sec): 4.63 - samples/sec: 1250.69 - lr: 0.000005 - momentum: 0.000000
2024-03-26 10:24:14,919 epoch 10 - iter 12/48 - loss 0.03229017 - time (sec): 6.64 - samples/sec: 1311.33 - lr: 0.000005 - momentum: 0.000000
2024-03-26 10:24:16,936 epoch 10 - iter 16/48 - loss 0.02870081 - time (sec): 8.66 - samples/sec: 1404.55 - lr: 0.000004 - momentum: 0.000000
2024-03-26 10:24:17,802 epoch 10 - iter 20/48 - loss 0.02796793 - time (sec): 9.53 - samples/sec: 1481.67 - lr: 0.000004 - momentum: 0.000000
2024-03-26 10:24:19,480 epoch 10 - iter 24/48 - loss 0.02776836 - time (sec): 11.21 - samples/sec: 1509.57 - lr: 0.000003 - momentum: 0.000000
2024-03-26 10:24:20,416 epoch 10 - iter 28/48 - loss 0.02689986 - time (sec): 12.14 - samples/sec: 1573.80 - lr: 0.000003 - momentum: 0.000000
2024-03-26 10:24:22,721 epoch 10 - iter 32/48 - loss 0.02474031 - time (sec): 14.45 - samples/sec: 1540.01 - lr: 0.000002 - momentum: 0.000000
2024-03-26 10:24:25,204 epoch 10 - iter 36/48 - loss 0.02836269 - time (sec): 16.93 - samples/sec: 1505.96 - lr: 0.000002 - momentum: 0.000000
2024-03-26 10:24:27,087 epoch 10 - iter 40/48 - loss 0.02902092 - time (sec): 18.81 - samples/sec: 1499.79 - lr: 0.000001 - momentum: 0.000000
2024-03-26 10:24:29,653 epoch 10 - iter 44/48 - loss 0.02860789 - time (sec): 21.38 - samples/sec: 1487.45 - lr: 0.000001 - momentum: 0.000000
2024-03-26 10:24:31,253 epoch 10 - iter 48/48 - loss 0.02837633 - time (sec): 22.98 - samples/sec: 1500.21 - lr: 0.000000 - momentum: 0.000000
2024-03-26 10:24:31,253 ----------------------------------------------------------------------------------------------------
2024-03-26 10:24:31,253 EPOCH 10 done: loss 0.0284 - lr: 0.000000
2024-03-26 10:24:32,155 DEV : loss 0.16044245660305023 - f1-score (micro avg) 0.9489
2024-03-26 10:24:32,156 saving best model
2024-03-26 10:24:32,848 ----------------------------------------------------------------------------------------------------
2024-03-26 10:24:32,849 Loading model from best epoch ...
2024-03-26 10:24:33,749 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 10:24:34,587
Results:
- F-score (micro) 0.9038
- F-score (macro) 0.6866
- Accuracy 0.8268
By class:
precision recall f1-score support
Unternehmen 0.9109 0.8835 0.8969 266
Auslagerung 0.8555 0.9036 0.8789 249
Ort 0.9565 0.9851 0.9706 134
Software 0.0000 0.0000 0.0000 0
micro avg 0.8956 0.9122 0.9038 649
macro avg 0.6807 0.6930 0.6866 649
weighted avg 0.8991 0.9122 0.9052 649
2024-03-26 10:24:34,587 ----------------------------------------------------------------------------------------------------
|