File size: 26,694 Bytes
8eeb39c |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 |
2024-03-26 10:15:43,560 ----------------------------------------------------------------------------------------------------
2024-03-26 10:15:43,560 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 10:15:43,560 ----------------------------------------------------------------------------------------------------
2024-03-26 10:15:43,560 Corpus: 758 train + 94 dev + 96 test sentences
2024-03-26 10:15:43,560 ----------------------------------------------------------------------------------------------------
2024-03-26 10:15:43,560 Train: 758 sentences
2024-03-26 10:15:43,560 (train_with_dev=False, train_with_test=False)
2024-03-26 10:15:43,560 ----------------------------------------------------------------------------------------------------
2024-03-26 10:15:43,560 Training Params:
2024-03-26 10:15:43,560 - learning_rate: "3e-05"
2024-03-26 10:15:43,560 - mini_batch_size: "16"
2024-03-26 10:15:43,560 - max_epochs: "10"
2024-03-26 10:15:43,560 - shuffle: "True"
2024-03-26 10:15:43,560 ----------------------------------------------------------------------------------------------------
2024-03-26 10:15:43,560 Plugins:
2024-03-26 10:15:43,561 - TensorboardLogger
2024-03-26 10:15:43,561 - LinearScheduler | warmup_fraction: '0.1'
2024-03-26 10:15:43,561 ----------------------------------------------------------------------------------------------------
2024-03-26 10:15:43,561 Final evaluation on model from best epoch (best-model.pt)
2024-03-26 10:15:43,561 - metric: "('micro avg', 'f1-score')"
2024-03-26 10:15:43,561 ----------------------------------------------------------------------------------------------------
2024-03-26 10:15:43,561 Computation:
2024-03-26 10:15:43,561 - compute on device: cuda:0
2024-03-26 10:15:43,561 - embedding storage: none
2024-03-26 10:15:43,561 ----------------------------------------------------------------------------------------------------
2024-03-26 10:15:43,561 Model training base path: "flair-co-funer-gbert_base-bs16-e10-lr3e-05-4"
2024-03-26 10:15:43,561 ----------------------------------------------------------------------------------------------------
2024-03-26 10:15:43,561 ----------------------------------------------------------------------------------------------------
2024-03-26 10:15:43,561 Logging anything other than scalars to TensorBoard is currently not supported.
2024-03-26 10:15:45,022 epoch 1 - iter 4/48 - loss 3.33606037 - time (sec): 1.46 - samples/sec: 1786.73 - lr: 0.000002 - momentum: 0.000000
2024-03-26 10:15:46,837 epoch 1 - iter 8/48 - loss 3.28043995 - time (sec): 3.28 - samples/sec: 1563.83 - lr: 0.000004 - momentum: 0.000000
2024-03-26 10:15:48,161 epoch 1 - iter 12/48 - loss 3.18154181 - time (sec): 4.60 - samples/sec: 1587.29 - lr: 0.000007 - momentum: 0.000000
2024-03-26 10:15:50,702 epoch 1 - iter 16/48 - loss 2.99682926 - time (sec): 7.14 - samples/sec: 1498.23 - lr: 0.000009 - momentum: 0.000000
2024-03-26 10:15:52,809 epoch 1 - iter 20/48 - loss 2.86564017 - time (sec): 9.25 - samples/sec: 1481.19 - lr: 0.000012 - momentum: 0.000000
2024-03-26 10:15:55,477 epoch 1 - iter 24/48 - loss 2.71552302 - time (sec): 11.92 - samples/sec: 1420.36 - lr: 0.000014 - momentum: 0.000000
2024-03-26 10:15:57,959 epoch 1 - iter 28/48 - loss 2.59674920 - time (sec): 14.40 - samples/sec: 1408.98 - lr: 0.000017 - momentum: 0.000000
2024-03-26 10:15:59,832 epoch 1 - iter 32/48 - loss 2.50531239 - time (sec): 16.27 - samples/sec: 1405.82 - lr: 0.000019 - momentum: 0.000000
2024-03-26 10:16:00,713 epoch 1 - iter 36/48 - loss 2.43592440 - time (sec): 17.15 - samples/sec: 1456.40 - lr: 0.000022 - momentum: 0.000000
2024-03-26 10:16:02,569 epoch 1 - iter 40/48 - loss 2.34208452 - time (sec): 19.01 - samples/sec: 1465.32 - lr: 0.000024 - momentum: 0.000000
2024-03-26 10:16:04,586 epoch 1 - iter 44/48 - loss 2.23846575 - time (sec): 21.02 - samples/sec: 1485.36 - lr: 0.000027 - momentum: 0.000000
2024-03-26 10:16:06,292 epoch 1 - iter 48/48 - loss 2.15856647 - time (sec): 22.73 - samples/sec: 1516.53 - lr: 0.000029 - momentum: 0.000000
2024-03-26 10:16:06,292 ----------------------------------------------------------------------------------------------------
2024-03-26 10:16:06,292 EPOCH 1 done: loss 2.1586 - lr: 0.000029
2024-03-26 10:16:07,099 DEV : loss 0.8691769242286682 - f1-score (micro avg) 0.3937
2024-03-26 10:16:07,100 saving best model
2024-03-26 10:16:07,381 ----------------------------------------------------------------------------------------------------
2024-03-26 10:16:08,612 epoch 2 - iter 4/48 - loss 1.18353349 - time (sec): 1.23 - samples/sec: 1924.87 - lr: 0.000030 - momentum: 0.000000
2024-03-26 10:16:10,851 epoch 2 - iter 8/48 - loss 0.99048634 - time (sec): 3.47 - samples/sec: 1572.73 - lr: 0.000030 - momentum: 0.000000
2024-03-26 10:16:12,631 epoch 2 - iter 12/48 - loss 0.94829107 - time (sec): 5.25 - samples/sec: 1623.34 - lr: 0.000029 - momentum: 0.000000
2024-03-26 10:16:15,015 epoch 2 - iter 16/48 - loss 0.86335015 - time (sec): 7.63 - samples/sec: 1478.02 - lr: 0.000029 - momentum: 0.000000
2024-03-26 10:16:18,426 epoch 2 - iter 20/48 - loss 0.78901466 - time (sec): 11.04 - samples/sec: 1335.95 - lr: 0.000029 - momentum: 0.000000
2024-03-26 10:16:19,906 epoch 2 - iter 24/48 - loss 0.77282323 - time (sec): 12.52 - samples/sec: 1392.74 - lr: 0.000028 - momentum: 0.000000
2024-03-26 10:16:22,559 epoch 2 - iter 28/48 - loss 0.74398129 - time (sec): 15.18 - samples/sec: 1364.18 - lr: 0.000028 - momentum: 0.000000
2024-03-26 10:16:25,258 epoch 2 - iter 32/48 - loss 0.71068818 - time (sec): 17.88 - samples/sec: 1365.43 - lr: 0.000028 - momentum: 0.000000
2024-03-26 10:16:27,337 epoch 2 - iter 36/48 - loss 0.69857610 - time (sec): 19.96 - samples/sec: 1355.18 - lr: 0.000028 - momentum: 0.000000
2024-03-26 10:16:29,805 epoch 2 - iter 40/48 - loss 0.67325179 - time (sec): 22.42 - samples/sec: 1346.07 - lr: 0.000027 - momentum: 0.000000
2024-03-26 10:16:30,853 epoch 2 - iter 44/48 - loss 0.66012014 - time (sec): 23.47 - samples/sec: 1381.45 - lr: 0.000027 - momentum: 0.000000
2024-03-26 10:16:32,008 epoch 2 - iter 48/48 - loss 0.64582622 - time (sec): 24.63 - samples/sec: 1399.82 - lr: 0.000027 - momentum: 0.000000
2024-03-26 10:16:32,008 ----------------------------------------------------------------------------------------------------
2024-03-26 10:16:32,008 EPOCH 2 done: loss 0.6458 - lr: 0.000027
2024-03-26 10:16:32,902 DEV : loss 0.3441326320171356 - f1-score (micro avg) 0.7893
2024-03-26 10:16:32,904 saving best model
2024-03-26 10:16:33,379 ----------------------------------------------------------------------------------------------------
2024-03-26 10:16:35,456 epoch 3 - iter 4/48 - loss 0.38138387 - time (sec): 2.08 - samples/sec: 1183.00 - lr: 0.000026 - momentum: 0.000000
2024-03-26 10:16:37,001 epoch 3 - iter 8/48 - loss 0.33198739 - time (sec): 3.62 - samples/sec: 1323.41 - lr: 0.000026 - momentum: 0.000000
2024-03-26 10:16:39,563 epoch 3 - iter 12/48 - loss 0.33511209 - time (sec): 6.18 - samples/sec: 1258.61 - lr: 0.000026 - momentum: 0.000000
2024-03-26 10:16:41,567 epoch 3 - iter 16/48 - loss 0.33306197 - time (sec): 8.19 - samples/sec: 1302.69 - lr: 0.000026 - momentum: 0.000000
2024-03-26 10:16:43,445 epoch 3 - iter 20/48 - loss 0.33037874 - time (sec): 10.06 - samples/sec: 1375.67 - lr: 0.000025 - momentum: 0.000000
2024-03-26 10:16:45,656 epoch 3 - iter 24/48 - loss 0.32391011 - time (sec): 12.27 - samples/sec: 1392.80 - lr: 0.000025 - momentum: 0.000000
2024-03-26 10:16:48,105 epoch 3 - iter 28/48 - loss 0.31328468 - time (sec): 14.72 - samples/sec: 1354.65 - lr: 0.000025 - momentum: 0.000000
2024-03-26 10:16:50,640 epoch 3 - iter 32/48 - loss 0.30770160 - time (sec): 17.26 - samples/sec: 1332.24 - lr: 0.000025 - momentum: 0.000000
2024-03-26 10:16:52,737 epoch 3 - iter 36/48 - loss 0.30585269 - time (sec): 19.36 - samples/sec: 1337.44 - lr: 0.000024 - momentum: 0.000000
2024-03-26 10:16:55,028 epoch 3 - iter 40/48 - loss 0.30995459 - time (sec): 21.65 - samples/sec: 1353.84 - lr: 0.000024 - momentum: 0.000000
2024-03-26 10:16:57,547 epoch 3 - iter 44/48 - loss 0.30188412 - time (sec): 24.17 - samples/sec: 1336.95 - lr: 0.000024 - momentum: 0.000000
2024-03-26 10:16:59,044 epoch 3 - iter 48/48 - loss 0.30424410 - time (sec): 25.66 - samples/sec: 1343.29 - lr: 0.000023 - momentum: 0.000000
2024-03-26 10:16:59,044 ----------------------------------------------------------------------------------------------------
2024-03-26 10:16:59,044 EPOCH 3 done: loss 0.3042 - lr: 0.000023
2024-03-26 10:16:59,951 DEV : loss 0.24834905564785004 - f1-score (micro avg) 0.8417
2024-03-26 10:16:59,952 saving best model
2024-03-26 10:17:00,388 ----------------------------------------------------------------------------------------------------
2024-03-26 10:17:03,381 epoch 4 - iter 4/48 - loss 0.16357798 - time (sec): 2.99 - samples/sec: 1218.46 - lr: 0.000023 - momentum: 0.000000
2024-03-26 10:17:04,675 epoch 4 - iter 8/48 - loss 0.19203458 - time (sec): 4.28 - samples/sec: 1372.52 - lr: 0.000023 - momentum: 0.000000
2024-03-26 10:17:06,741 epoch 4 - iter 12/48 - loss 0.20034927 - time (sec): 6.35 - samples/sec: 1452.34 - lr: 0.000023 - momentum: 0.000000
2024-03-26 10:17:09,275 epoch 4 - iter 16/48 - loss 0.19820873 - time (sec): 8.88 - samples/sec: 1371.25 - lr: 0.000022 - momentum: 0.000000
2024-03-26 10:17:10,256 epoch 4 - iter 20/48 - loss 0.19764534 - time (sec): 9.87 - samples/sec: 1455.96 - lr: 0.000022 - momentum: 0.000000
2024-03-26 10:17:11,638 epoch 4 - iter 24/48 - loss 0.20056094 - time (sec): 11.25 - samples/sec: 1502.68 - lr: 0.000022 - momentum: 0.000000
2024-03-26 10:17:14,733 epoch 4 - iter 28/48 - loss 0.19142315 - time (sec): 14.34 - samples/sec: 1407.17 - lr: 0.000022 - momentum: 0.000000
2024-03-26 10:17:17,197 epoch 4 - iter 32/48 - loss 0.20173939 - time (sec): 16.81 - samples/sec: 1399.26 - lr: 0.000021 - momentum: 0.000000
2024-03-26 10:17:18,705 epoch 4 - iter 36/48 - loss 0.20107167 - time (sec): 18.32 - samples/sec: 1434.77 - lr: 0.000021 - momentum: 0.000000
2024-03-26 10:17:20,679 epoch 4 - iter 40/48 - loss 0.19651999 - time (sec): 20.29 - samples/sec: 1449.07 - lr: 0.000021 - momentum: 0.000000
2024-03-26 10:17:22,559 epoch 4 - iter 44/48 - loss 0.19276541 - time (sec): 22.17 - samples/sec: 1463.05 - lr: 0.000020 - momentum: 0.000000
2024-03-26 10:17:23,594 epoch 4 - iter 48/48 - loss 0.19451533 - time (sec): 23.20 - samples/sec: 1485.60 - lr: 0.000020 - momentum: 0.000000
2024-03-26 10:17:23,594 ----------------------------------------------------------------------------------------------------
2024-03-26 10:17:23,594 EPOCH 4 done: loss 0.1945 - lr: 0.000020
2024-03-26 10:17:24,489 DEV : loss 0.21226930618286133 - f1-score (micro avg) 0.8725
2024-03-26 10:17:24,490 saving best model
2024-03-26 10:17:24,914 ----------------------------------------------------------------------------------------------------
2024-03-26 10:17:25,970 epoch 5 - iter 4/48 - loss 0.22900511 - time (sec): 1.05 - samples/sec: 2412.52 - lr: 0.000020 - momentum: 0.000000
2024-03-26 10:17:27,831 epoch 5 - iter 8/48 - loss 0.20745093 - time (sec): 2.92 - samples/sec: 1777.12 - lr: 0.000020 - momentum: 0.000000
2024-03-26 10:17:29,928 epoch 5 - iter 12/48 - loss 0.19394759 - time (sec): 5.01 - samples/sec: 1596.67 - lr: 0.000019 - momentum: 0.000000
2024-03-26 10:17:32,188 epoch 5 - iter 16/48 - loss 0.17919168 - time (sec): 7.27 - samples/sec: 1525.06 - lr: 0.000019 - momentum: 0.000000
2024-03-26 10:17:34,431 epoch 5 - iter 20/48 - loss 0.17309969 - time (sec): 9.52 - samples/sec: 1438.12 - lr: 0.000019 - momentum: 0.000000
2024-03-26 10:17:36,589 epoch 5 - iter 24/48 - loss 0.16467634 - time (sec): 11.67 - samples/sec: 1455.52 - lr: 0.000018 - momentum: 0.000000
2024-03-26 10:17:38,184 epoch 5 - iter 28/48 - loss 0.16079508 - time (sec): 13.27 - samples/sec: 1483.39 - lr: 0.000018 - momentum: 0.000000
2024-03-26 10:17:40,278 epoch 5 - iter 32/48 - loss 0.15041135 - time (sec): 15.36 - samples/sec: 1503.84 - lr: 0.000018 - momentum: 0.000000
2024-03-26 10:17:41,672 epoch 5 - iter 36/48 - loss 0.14767635 - time (sec): 16.76 - samples/sec: 1527.72 - lr: 0.000018 - momentum: 0.000000
2024-03-26 10:17:44,219 epoch 5 - iter 40/48 - loss 0.14147108 - time (sec): 19.30 - samples/sec: 1493.43 - lr: 0.000017 - momentum: 0.000000
2024-03-26 10:17:47,159 epoch 5 - iter 44/48 - loss 0.13978631 - time (sec): 22.24 - samples/sec: 1439.87 - lr: 0.000017 - momentum: 0.000000
2024-03-26 10:17:48,665 epoch 5 - iter 48/48 - loss 0.14191100 - time (sec): 23.75 - samples/sec: 1451.53 - lr: 0.000017 - momentum: 0.000000
2024-03-26 10:17:48,665 ----------------------------------------------------------------------------------------------------
2024-03-26 10:17:48,665 EPOCH 5 done: loss 0.1419 - lr: 0.000017
2024-03-26 10:17:49,581 DEV : loss 0.17432889342308044 - f1-score (micro avg) 0.8924
2024-03-26 10:17:49,583 saving best model
2024-03-26 10:17:50,029 ----------------------------------------------------------------------------------------------------
2024-03-26 10:17:51,893 epoch 6 - iter 4/48 - loss 0.14473226 - time (sec): 1.86 - samples/sec: 1578.61 - lr: 0.000017 - momentum: 0.000000
2024-03-26 10:17:53,612 epoch 6 - iter 8/48 - loss 0.12456982 - time (sec): 3.58 - samples/sec: 1618.91 - lr: 0.000016 - momentum: 0.000000
2024-03-26 10:17:55,912 epoch 6 - iter 12/48 - loss 0.12108378 - time (sec): 5.88 - samples/sec: 1499.33 - lr: 0.000016 - momentum: 0.000000
2024-03-26 10:17:57,476 epoch 6 - iter 16/48 - loss 0.11264219 - time (sec): 7.44 - samples/sec: 1522.01 - lr: 0.000016 - momentum: 0.000000
2024-03-26 10:18:00,033 epoch 6 - iter 20/48 - loss 0.10247223 - time (sec): 10.00 - samples/sec: 1436.35 - lr: 0.000015 - momentum: 0.000000
2024-03-26 10:18:02,087 epoch 6 - iter 24/48 - loss 0.10312786 - time (sec): 12.06 - samples/sec: 1450.74 - lr: 0.000015 - momentum: 0.000000
2024-03-26 10:18:04,714 epoch 6 - iter 28/48 - loss 0.10393896 - time (sec): 14.68 - samples/sec: 1425.71 - lr: 0.000015 - momentum: 0.000000
2024-03-26 10:18:06,764 epoch 6 - iter 32/48 - loss 0.10212557 - time (sec): 16.73 - samples/sec: 1404.50 - lr: 0.000015 - momentum: 0.000000
2024-03-26 10:18:07,864 epoch 6 - iter 36/48 - loss 0.10345036 - time (sec): 17.83 - samples/sec: 1454.68 - lr: 0.000014 - momentum: 0.000000
2024-03-26 10:18:10,059 epoch 6 - iter 40/48 - loss 0.10498644 - time (sec): 20.03 - samples/sec: 1443.81 - lr: 0.000014 - momentum: 0.000000
2024-03-26 10:18:11,667 epoch 6 - iter 44/48 - loss 0.10975459 - time (sec): 21.64 - samples/sec: 1467.53 - lr: 0.000014 - momentum: 0.000000
2024-03-26 10:18:13,439 epoch 6 - iter 48/48 - loss 0.10750133 - time (sec): 23.41 - samples/sec: 1472.63 - lr: 0.000014 - momentum: 0.000000
2024-03-26 10:18:13,440 ----------------------------------------------------------------------------------------------------
2024-03-26 10:18:13,440 EPOCH 6 done: loss 0.1075 - lr: 0.000014
2024-03-26 10:18:14,342 DEV : loss 0.16816526651382446 - f1-score (micro avg) 0.8996
2024-03-26 10:18:14,344 saving best model
2024-03-26 10:18:14,755 ----------------------------------------------------------------------------------------------------
2024-03-26 10:18:16,361 epoch 7 - iter 4/48 - loss 0.09156530 - time (sec): 1.61 - samples/sec: 1743.14 - lr: 0.000013 - momentum: 0.000000
2024-03-26 10:18:18,453 epoch 7 - iter 8/48 - loss 0.07762594 - time (sec): 3.70 - samples/sec: 1655.21 - lr: 0.000013 - momentum: 0.000000
2024-03-26 10:18:20,673 epoch 7 - iter 12/48 - loss 0.07628831 - time (sec): 5.92 - samples/sec: 1488.77 - lr: 0.000013 - momentum: 0.000000
2024-03-26 10:18:21,840 epoch 7 - iter 16/48 - loss 0.08639863 - time (sec): 7.08 - samples/sec: 1588.88 - lr: 0.000012 - momentum: 0.000000
2024-03-26 10:18:23,946 epoch 7 - iter 20/48 - loss 0.08684581 - time (sec): 9.19 - samples/sec: 1560.88 - lr: 0.000012 - momentum: 0.000000
2024-03-26 10:18:25,446 epoch 7 - iter 24/48 - loss 0.08404483 - time (sec): 10.69 - samples/sec: 1610.09 - lr: 0.000012 - momentum: 0.000000
2024-03-26 10:18:27,540 epoch 7 - iter 28/48 - loss 0.08184808 - time (sec): 12.78 - samples/sec: 1570.22 - lr: 0.000012 - momentum: 0.000000
2024-03-26 10:18:30,297 epoch 7 - iter 32/48 - loss 0.08200584 - time (sec): 15.54 - samples/sec: 1498.51 - lr: 0.000011 - momentum: 0.000000
2024-03-26 10:18:32,241 epoch 7 - iter 36/48 - loss 0.08034921 - time (sec): 17.49 - samples/sec: 1500.73 - lr: 0.000011 - momentum: 0.000000
2024-03-26 10:18:33,354 epoch 7 - iter 40/48 - loss 0.08407326 - time (sec): 18.60 - samples/sec: 1532.40 - lr: 0.000011 - momentum: 0.000000
2024-03-26 10:18:35,939 epoch 7 - iter 44/48 - loss 0.08503551 - time (sec): 21.18 - samples/sec: 1513.53 - lr: 0.000010 - momentum: 0.000000
2024-03-26 10:18:37,041 epoch 7 - iter 48/48 - loss 0.08657468 - time (sec): 22.28 - samples/sec: 1546.88 - lr: 0.000010 - momentum: 0.000000
2024-03-26 10:18:37,041 ----------------------------------------------------------------------------------------------------
2024-03-26 10:18:37,041 EPOCH 7 done: loss 0.0866 - lr: 0.000010
2024-03-26 10:18:37,939 DEV : loss 0.18013019859790802 - f1-score (micro avg) 0.9053
2024-03-26 10:18:37,940 saving best model
2024-03-26 10:18:38,380 ----------------------------------------------------------------------------------------------------
2024-03-26 10:18:40,490 epoch 8 - iter 4/48 - loss 0.05940590 - time (sec): 2.11 - samples/sec: 1314.77 - lr: 0.000010 - momentum: 0.000000
2024-03-26 10:18:43,093 epoch 8 - iter 8/48 - loss 0.05066426 - time (sec): 4.71 - samples/sec: 1281.95 - lr: 0.000010 - momentum: 0.000000
2024-03-26 10:18:44,741 epoch 8 - iter 12/48 - loss 0.05202001 - time (sec): 6.36 - samples/sec: 1333.67 - lr: 0.000009 - momentum: 0.000000
2024-03-26 10:18:47,332 epoch 8 - iter 16/48 - loss 0.06461025 - time (sec): 8.95 - samples/sec: 1286.28 - lr: 0.000009 - momentum: 0.000000
2024-03-26 10:18:48,957 epoch 8 - iter 20/48 - loss 0.06519928 - time (sec): 10.57 - samples/sec: 1343.58 - lr: 0.000009 - momentum: 0.000000
2024-03-26 10:18:50,408 epoch 8 - iter 24/48 - loss 0.07047337 - time (sec): 12.03 - samples/sec: 1413.48 - lr: 0.000009 - momentum: 0.000000
2024-03-26 10:18:52,258 epoch 8 - iter 28/48 - loss 0.07587165 - time (sec): 13.88 - samples/sec: 1437.19 - lr: 0.000008 - momentum: 0.000000
2024-03-26 10:18:54,884 epoch 8 - iter 32/48 - loss 0.07553527 - time (sec): 16.50 - samples/sec: 1423.63 - lr: 0.000008 - momentum: 0.000000
2024-03-26 10:18:57,265 epoch 8 - iter 36/48 - loss 0.07533218 - time (sec): 18.88 - samples/sec: 1415.70 - lr: 0.000008 - momentum: 0.000000
2024-03-26 10:18:59,444 epoch 8 - iter 40/48 - loss 0.07539987 - time (sec): 21.06 - samples/sec: 1396.95 - lr: 0.000007 - momentum: 0.000000
2024-03-26 10:19:01,664 epoch 8 - iter 44/48 - loss 0.07328438 - time (sec): 23.28 - samples/sec: 1387.55 - lr: 0.000007 - momentum: 0.000000
2024-03-26 10:19:03,207 epoch 8 - iter 48/48 - loss 0.07364478 - time (sec): 24.83 - samples/sec: 1388.59 - lr: 0.000007 - momentum: 0.000000
2024-03-26 10:19:03,208 ----------------------------------------------------------------------------------------------------
2024-03-26 10:19:03,208 EPOCH 8 done: loss 0.0736 - lr: 0.000007
2024-03-26 10:19:04,106 DEV : loss 0.16676065325737 - f1-score (micro avg) 0.9158
2024-03-26 10:19:04,107 saving best model
2024-03-26 10:19:04,540 ----------------------------------------------------------------------------------------------------
2024-03-26 10:19:06,396 epoch 9 - iter 4/48 - loss 0.07456918 - time (sec): 1.85 - samples/sec: 1557.74 - lr: 0.000007 - momentum: 0.000000
2024-03-26 10:19:09,537 epoch 9 - iter 8/48 - loss 0.07242803 - time (sec): 4.99 - samples/sec: 1260.29 - lr: 0.000006 - momentum: 0.000000
2024-03-26 10:19:11,171 epoch 9 - iter 12/48 - loss 0.06181526 - time (sec): 6.63 - samples/sec: 1304.71 - lr: 0.000006 - momentum: 0.000000
2024-03-26 10:19:13,041 epoch 9 - iter 16/48 - loss 0.06839209 - time (sec): 8.50 - samples/sec: 1345.05 - lr: 0.000006 - momentum: 0.000000
2024-03-26 10:19:15,912 epoch 9 - iter 20/48 - loss 0.06071099 - time (sec): 11.37 - samples/sec: 1306.30 - lr: 0.000006 - momentum: 0.000000
2024-03-26 10:19:17,437 epoch 9 - iter 24/48 - loss 0.06069450 - time (sec): 12.90 - samples/sec: 1353.07 - lr: 0.000005 - momentum: 0.000000
2024-03-26 10:19:19,370 epoch 9 - iter 28/48 - loss 0.06366137 - time (sec): 14.83 - samples/sec: 1377.70 - lr: 0.000005 - momentum: 0.000000
2024-03-26 10:19:21,689 epoch 9 - iter 32/48 - loss 0.06159224 - time (sec): 17.15 - samples/sec: 1354.78 - lr: 0.000005 - momentum: 0.000000
2024-03-26 10:19:22,986 epoch 9 - iter 36/48 - loss 0.06577623 - time (sec): 18.44 - samples/sec: 1385.96 - lr: 0.000004 - momentum: 0.000000
2024-03-26 10:19:26,188 epoch 9 - iter 40/48 - loss 0.06302902 - time (sec): 21.65 - samples/sec: 1336.88 - lr: 0.000004 - momentum: 0.000000
2024-03-26 10:19:28,302 epoch 9 - iter 44/48 - loss 0.06043263 - time (sec): 23.76 - samples/sec: 1359.42 - lr: 0.000004 - momentum: 0.000000
2024-03-26 10:19:29,289 epoch 9 - iter 48/48 - loss 0.06250231 - time (sec): 24.75 - samples/sec: 1393.01 - lr: 0.000004 - momentum: 0.000000
2024-03-26 10:19:29,289 ----------------------------------------------------------------------------------------------------
2024-03-26 10:19:29,289 EPOCH 9 done: loss 0.0625 - lr: 0.000004
2024-03-26 10:19:30,191 DEV : loss 0.15200284123420715 - f1-score (micro avg) 0.925
2024-03-26 10:19:30,192 saving best model
2024-03-26 10:19:30,613 ----------------------------------------------------------------------------------------------------
2024-03-26 10:19:32,484 epoch 10 - iter 4/48 - loss 0.06632142 - time (sec): 1.87 - samples/sec: 1382.85 - lr: 0.000003 - momentum: 0.000000
2024-03-26 10:19:35,257 epoch 10 - iter 8/48 - loss 0.04726173 - time (sec): 4.64 - samples/sec: 1246.33 - lr: 0.000003 - momentum: 0.000000
2024-03-26 10:19:37,284 epoch 10 - iter 12/48 - loss 0.05495615 - time (sec): 6.67 - samples/sec: 1306.57 - lr: 0.000003 - momentum: 0.000000
2024-03-26 10:19:39,304 epoch 10 - iter 16/48 - loss 0.05438606 - time (sec): 8.69 - samples/sec: 1400.10 - lr: 0.000002 - momentum: 0.000000
2024-03-26 10:19:40,170 epoch 10 - iter 20/48 - loss 0.05397075 - time (sec): 9.56 - samples/sec: 1477.42 - lr: 0.000002 - momentum: 0.000000
2024-03-26 10:19:41,853 epoch 10 - iter 24/48 - loss 0.05286287 - time (sec): 11.24 - samples/sec: 1505.12 - lr: 0.000002 - momentum: 0.000000
2024-03-26 10:19:42,787 epoch 10 - iter 28/48 - loss 0.05312673 - time (sec): 12.17 - samples/sec: 1569.84 - lr: 0.000002 - momentum: 0.000000
2024-03-26 10:19:45,100 epoch 10 - iter 32/48 - loss 0.05163266 - time (sec): 14.48 - samples/sec: 1535.96 - lr: 0.000001 - momentum: 0.000000
2024-03-26 10:19:47,587 epoch 10 - iter 36/48 - loss 0.05428024 - time (sec): 16.97 - samples/sec: 1502.19 - lr: 0.000001 - momentum: 0.000000
2024-03-26 10:19:49,469 epoch 10 - iter 40/48 - loss 0.05721031 - time (sec): 18.85 - samples/sec: 1496.49 - lr: 0.000001 - momentum: 0.000000
2024-03-26 10:19:52,039 epoch 10 - iter 44/48 - loss 0.05600849 - time (sec): 21.42 - samples/sec: 1484.28 - lr: 0.000001 - momentum: 0.000000
2024-03-26 10:19:53,639 epoch 10 - iter 48/48 - loss 0.05587015 - time (sec): 23.02 - samples/sec: 1497.23 - lr: 0.000000 - momentum: 0.000000
2024-03-26 10:19:53,639 ----------------------------------------------------------------------------------------------------
2024-03-26 10:19:53,639 EPOCH 10 done: loss 0.0559 - lr: 0.000000
2024-03-26 10:19:54,542 DEV : loss 0.15424026548862457 - f1-score (micro avg) 0.9273
2024-03-26 10:19:54,543 saving best model
2024-03-26 10:19:55,253 ----------------------------------------------------------------------------------------------------
2024-03-26 10:19:55,253 Loading model from best epoch ...
2024-03-26 10:19:56,194 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 10:19:56,949
Results:
- F-score (micro) 0.9048
- F-score (macro) 0.6876
- Accuracy 0.8284
By class:
precision recall f1-score support
Unternehmen 0.9102 0.8759 0.8927 266
Auslagerung 0.8682 0.8996 0.8836 249
Ort 0.9635 0.9851 0.9742 134
Software 0.0000 0.0000 0.0000 0
micro avg 0.9020 0.9076 0.9048 649
macro avg 0.6855 0.6902 0.6876 649
weighted avg 0.9051 0.9076 0.9060 649
2024-03-26 10:19:56,949 ----------------------------------------------------------------------------------------------------
|