File size: 26,652 Bytes
eb40798 |
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 10:03:59,849 ----------------------------------------------------------------------------------------------------
2024-03-26 10:03:59,850 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 10:03:59,850 ----------------------------------------------------------------------------------------------------
2024-03-26 10:03:59,850 Corpus: 758 train + 94 dev + 96 test sentences
2024-03-26 10:03:59,850 ----------------------------------------------------------------------------------------------------
2024-03-26 10:03:59,850 Train: 758 sentences
2024-03-26 10:03:59,850 (train_with_dev=False, train_with_test=False)
2024-03-26 10:03:59,850 ----------------------------------------------------------------------------------------------------
2024-03-26 10:03:59,850 Training Params:
2024-03-26 10:03:59,850 - learning_rate: "5e-05"
2024-03-26 10:03:59,850 - mini_batch_size: "16"
2024-03-26 10:03:59,850 - max_epochs: "10"
2024-03-26 10:03:59,850 - shuffle: "True"
2024-03-26 10:03:59,850 ----------------------------------------------------------------------------------------------------
2024-03-26 10:03:59,850 Plugins:
2024-03-26 10:03:59,850 - TensorboardLogger
2024-03-26 10:03:59,850 - LinearScheduler | warmup_fraction: '0.1'
2024-03-26 10:03:59,850 ----------------------------------------------------------------------------------------------------
2024-03-26 10:03:59,850 Final evaluation on model from best epoch (best-model.pt)
2024-03-26 10:03:59,850 - metric: "('micro avg', 'f1-score')"
2024-03-26 10:03:59,850 ----------------------------------------------------------------------------------------------------
2024-03-26 10:03:59,850 Computation:
2024-03-26 10:03:59,850 - compute on device: cuda:0
2024-03-26 10:03:59,850 - embedding storage: none
2024-03-26 10:03:59,850 ----------------------------------------------------------------------------------------------------
2024-03-26 10:03:59,850 Model training base path: "flair-co-funer-gbert_base-bs16-e10-lr5e-05-3"
2024-03-26 10:03:59,850 ----------------------------------------------------------------------------------------------------
2024-03-26 10:03:59,850 ----------------------------------------------------------------------------------------------------
2024-03-26 10:03:59,850 Logging anything other than scalars to TensorBoard is currently not supported.
2024-03-26 10:04:01,094 epoch 1 - iter 4/48 - loss 3.42847207 - time (sec): 1.24 - samples/sec: 2213.10 - lr: 0.000003 - momentum: 0.000000
2024-03-26 10:04:03,051 epoch 1 - iter 8/48 - loss 3.32248082 - time (sec): 3.20 - samples/sec: 1819.69 - lr: 0.000007 - momentum: 0.000000
2024-03-26 10:04:04,566 epoch 1 - iter 12/48 - loss 3.17815477 - time (sec): 4.72 - samples/sec: 1775.93 - lr: 0.000011 - momentum: 0.000000
2024-03-26 10:04:07,434 epoch 1 - iter 16/48 - loss 2.96264845 - time (sec): 7.58 - samples/sec: 1529.34 - lr: 0.000016 - momentum: 0.000000
2024-03-26 10:04:09,078 epoch 1 - iter 20/48 - loss 2.79126105 - time (sec): 9.23 - samples/sec: 1560.92 - lr: 0.000020 - momentum: 0.000000
2024-03-26 10:04:10,501 epoch 1 - iter 24/48 - loss 2.66809628 - time (sec): 10.65 - samples/sec: 1611.84 - lr: 0.000024 - momentum: 0.000000
2024-03-26 10:04:11,808 epoch 1 - iter 28/48 - loss 2.54471273 - time (sec): 11.96 - samples/sec: 1633.05 - lr: 0.000028 - momentum: 0.000000
2024-03-26 10:04:13,881 epoch 1 - iter 32/48 - loss 2.40956988 - time (sec): 14.03 - samples/sec: 1619.87 - lr: 0.000032 - momentum: 0.000000
2024-03-26 10:04:15,418 epoch 1 - iter 36/48 - loss 2.28409769 - time (sec): 15.57 - samples/sec: 1639.18 - lr: 0.000036 - momentum: 0.000000
2024-03-26 10:04:17,635 epoch 1 - iter 40/48 - loss 2.14660908 - time (sec): 17.78 - samples/sec: 1629.26 - lr: 0.000041 - momentum: 0.000000
2024-03-26 10:04:19,542 epoch 1 - iter 44/48 - loss 2.02954310 - time (sec): 19.69 - samples/sec: 1628.38 - lr: 0.000045 - momentum: 0.000000
2024-03-26 10:04:21,152 epoch 1 - iter 48/48 - loss 1.93848867 - time (sec): 21.30 - samples/sec: 1618.28 - lr: 0.000049 - momentum: 0.000000
2024-03-26 10:04:21,152 ----------------------------------------------------------------------------------------------------
2024-03-26 10:04:21,152 EPOCH 1 done: loss 1.9385 - lr: 0.000049
2024-03-26 10:04:21,961 DEV : loss 0.5561796426773071 - f1-score (micro avg) 0.6403
2024-03-26 10:04:21,962 saving best model
2024-03-26 10:04:22,243 ----------------------------------------------------------------------------------------------------
2024-03-26 10:04:23,645 epoch 2 - iter 4/48 - loss 0.68169693 - time (sec): 1.40 - samples/sec: 1780.37 - lr: 0.000050 - momentum: 0.000000
2024-03-26 10:04:25,098 epoch 2 - iter 8/48 - loss 0.58987382 - time (sec): 2.85 - samples/sec: 1709.75 - lr: 0.000049 - momentum: 0.000000
2024-03-26 10:04:26,501 epoch 2 - iter 12/48 - loss 0.57924523 - time (sec): 4.26 - samples/sec: 1805.47 - lr: 0.000049 - momentum: 0.000000
2024-03-26 10:04:28,350 epoch 2 - iter 16/48 - loss 0.53887829 - time (sec): 6.11 - samples/sec: 1761.61 - lr: 0.000048 - momentum: 0.000000
2024-03-26 10:04:30,676 epoch 2 - iter 20/48 - loss 0.52146094 - time (sec): 8.43 - samples/sec: 1681.60 - lr: 0.000048 - momentum: 0.000000
2024-03-26 10:04:32,671 epoch 2 - iter 24/48 - loss 0.48389242 - time (sec): 10.43 - samples/sec: 1663.33 - lr: 0.000047 - momentum: 0.000000
2024-03-26 10:04:35,363 epoch 2 - iter 28/48 - loss 0.46805210 - time (sec): 13.12 - samples/sec: 1594.97 - lr: 0.000047 - momentum: 0.000000
2024-03-26 10:04:37,528 epoch 2 - iter 32/48 - loss 0.45046616 - time (sec): 15.28 - samples/sec: 1561.55 - lr: 0.000046 - momentum: 0.000000
2024-03-26 10:04:39,221 epoch 2 - iter 36/48 - loss 0.44202409 - time (sec): 16.98 - samples/sec: 1555.99 - lr: 0.000046 - momentum: 0.000000
2024-03-26 10:04:40,911 epoch 2 - iter 40/48 - loss 0.44187690 - time (sec): 18.67 - samples/sec: 1562.70 - lr: 0.000046 - momentum: 0.000000
2024-03-26 10:04:43,039 epoch 2 - iter 44/48 - loss 0.43102293 - time (sec): 20.80 - samples/sec: 1558.14 - lr: 0.000045 - momentum: 0.000000
2024-03-26 10:04:44,533 epoch 2 - iter 48/48 - loss 0.41948321 - time (sec): 22.29 - samples/sec: 1546.53 - lr: 0.000045 - momentum: 0.000000
2024-03-26 10:04:44,533 ----------------------------------------------------------------------------------------------------
2024-03-26 10:04:44,533 EPOCH 2 done: loss 0.4195 - lr: 0.000045
2024-03-26 10:04:45,419 DEV : loss 0.2610315978527069 - f1-score (micro avg) 0.8488
2024-03-26 10:04:45,420 saving best model
2024-03-26 10:04:45,900 ----------------------------------------------------------------------------------------------------
2024-03-26 10:04:47,380 epoch 3 - iter 4/48 - loss 0.27115186 - time (sec): 1.48 - samples/sec: 1657.88 - lr: 0.000044 - momentum: 0.000000
2024-03-26 10:04:50,063 epoch 3 - iter 8/48 - loss 0.21895063 - time (sec): 4.16 - samples/sec: 1376.71 - lr: 0.000044 - momentum: 0.000000
2024-03-26 10:04:51,291 epoch 3 - iter 12/48 - loss 0.23419962 - time (sec): 5.39 - samples/sec: 1511.42 - lr: 0.000043 - momentum: 0.000000
2024-03-26 10:04:52,627 epoch 3 - iter 16/48 - loss 0.21189624 - time (sec): 6.73 - samples/sec: 1641.16 - lr: 0.000043 - momentum: 0.000000
2024-03-26 10:04:54,054 epoch 3 - iter 20/48 - loss 0.21487538 - time (sec): 8.15 - samples/sec: 1657.27 - lr: 0.000042 - momentum: 0.000000
2024-03-26 10:04:56,719 epoch 3 - iter 24/48 - loss 0.20856793 - time (sec): 10.82 - samples/sec: 1544.04 - lr: 0.000042 - momentum: 0.000000
2024-03-26 10:04:58,592 epoch 3 - iter 28/48 - loss 0.21250090 - time (sec): 12.69 - samples/sec: 1561.83 - lr: 0.000041 - momentum: 0.000000
2024-03-26 10:05:01,073 epoch 3 - iter 32/48 - loss 0.20205428 - time (sec): 15.17 - samples/sec: 1502.91 - lr: 0.000041 - momentum: 0.000000
2024-03-26 10:05:02,973 epoch 3 - iter 36/48 - loss 0.20232438 - time (sec): 17.07 - samples/sec: 1498.34 - lr: 0.000040 - momentum: 0.000000
2024-03-26 10:05:05,288 epoch 3 - iter 40/48 - loss 0.19607315 - time (sec): 19.39 - samples/sec: 1473.45 - lr: 0.000040 - momentum: 0.000000
2024-03-26 10:05:07,686 epoch 3 - iter 44/48 - loss 0.20415740 - time (sec): 21.78 - samples/sec: 1460.81 - lr: 0.000040 - momentum: 0.000000
2024-03-26 10:05:09,974 epoch 3 - iter 48/48 - loss 0.19702094 - time (sec): 24.07 - samples/sec: 1432.03 - lr: 0.000039 - momentum: 0.000000
2024-03-26 10:05:09,974 ----------------------------------------------------------------------------------------------------
2024-03-26 10:05:09,974 EPOCH 3 done: loss 0.1970 - lr: 0.000039
2024-03-26 10:05:10,863 DEV : loss 0.20633003115653992 - f1-score (micro avg) 0.8791
2024-03-26 10:05:10,864 saving best model
2024-03-26 10:05:11,310 ----------------------------------------------------------------------------------------------------
2024-03-26 10:05:12,768 epoch 4 - iter 4/48 - loss 0.15569544 - time (sec): 1.46 - samples/sec: 1719.96 - lr: 0.000039 - momentum: 0.000000
2024-03-26 10:05:14,663 epoch 4 - iter 8/48 - loss 0.13810570 - time (sec): 3.35 - samples/sec: 1599.03 - lr: 0.000038 - momentum: 0.000000
2024-03-26 10:05:17,174 epoch 4 - iter 12/48 - loss 0.12486450 - time (sec): 5.86 - samples/sec: 1439.69 - lr: 0.000038 - momentum: 0.000000
2024-03-26 10:05:19,053 epoch 4 - iter 16/48 - loss 0.12364042 - time (sec): 7.74 - samples/sec: 1462.77 - lr: 0.000037 - momentum: 0.000000
2024-03-26 10:05:21,409 epoch 4 - iter 20/48 - loss 0.11913852 - time (sec): 10.10 - samples/sec: 1455.15 - lr: 0.000037 - momentum: 0.000000
2024-03-26 10:05:24,284 epoch 4 - iter 24/48 - loss 0.11463006 - time (sec): 12.97 - samples/sec: 1405.63 - lr: 0.000036 - momentum: 0.000000
2024-03-26 10:05:25,402 epoch 4 - iter 28/48 - loss 0.11545901 - time (sec): 14.09 - samples/sec: 1441.74 - lr: 0.000036 - momentum: 0.000000
2024-03-26 10:05:28,399 epoch 4 - iter 32/48 - loss 0.11436647 - time (sec): 17.09 - samples/sec: 1382.25 - lr: 0.000035 - momentum: 0.000000
2024-03-26 10:05:30,117 epoch 4 - iter 36/48 - loss 0.12044215 - time (sec): 18.81 - samples/sec: 1414.80 - lr: 0.000035 - momentum: 0.000000
2024-03-26 10:05:32,925 epoch 4 - iter 40/48 - loss 0.12453469 - time (sec): 21.61 - samples/sec: 1382.74 - lr: 0.000034 - momentum: 0.000000
2024-03-26 10:05:33,832 epoch 4 - iter 44/48 - loss 0.12667938 - time (sec): 22.52 - samples/sec: 1428.87 - lr: 0.000034 - momentum: 0.000000
2024-03-26 10:05:35,307 epoch 4 - iter 48/48 - loss 0.12821766 - time (sec): 24.00 - samples/sec: 1436.61 - lr: 0.000034 - momentum: 0.000000
2024-03-26 10:05:35,308 ----------------------------------------------------------------------------------------------------
2024-03-26 10:05:35,308 EPOCH 4 done: loss 0.1282 - lr: 0.000034
2024-03-26 10:05:36,195 DEV : loss 0.1677311509847641 - f1-score (micro avg) 0.9039
2024-03-26 10:05:36,196 saving best model
2024-03-26 10:05:36,667 ----------------------------------------------------------------------------------------------------
2024-03-26 10:05:39,099 epoch 5 - iter 4/48 - loss 0.07272465 - time (sec): 2.43 - samples/sec: 1307.92 - lr: 0.000033 - momentum: 0.000000
2024-03-26 10:05:40,508 epoch 5 - iter 8/48 - loss 0.08934142 - time (sec): 3.84 - samples/sec: 1482.60 - lr: 0.000033 - momentum: 0.000000
2024-03-26 10:05:41,952 epoch 5 - iter 12/48 - loss 0.08879200 - time (sec): 5.28 - samples/sec: 1559.78 - lr: 0.000032 - momentum: 0.000000
2024-03-26 10:05:44,122 epoch 5 - iter 16/48 - loss 0.08576217 - time (sec): 7.45 - samples/sec: 1476.19 - lr: 0.000032 - momentum: 0.000000
2024-03-26 10:05:46,174 epoch 5 - iter 20/48 - loss 0.09482743 - time (sec): 9.51 - samples/sec: 1481.30 - lr: 0.000031 - momentum: 0.000000
2024-03-26 10:05:48,645 epoch 5 - iter 24/48 - loss 0.09094812 - time (sec): 11.98 - samples/sec: 1469.90 - lr: 0.000031 - momentum: 0.000000
2024-03-26 10:05:51,186 epoch 5 - iter 28/48 - loss 0.08571930 - time (sec): 14.52 - samples/sec: 1449.92 - lr: 0.000030 - momentum: 0.000000
2024-03-26 10:05:53,044 epoch 5 - iter 32/48 - loss 0.08483497 - time (sec): 16.38 - samples/sec: 1454.13 - lr: 0.000030 - momentum: 0.000000
2024-03-26 10:05:54,846 epoch 5 - iter 36/48 - loss 0.08308133 - time (sec): 18.18 - samples/sec: 1454.89 - lr: 0.000029 - momentum: 0.000000
2024-03-26 10:05:57,131 epoch 5 - iter 40/48 - loss 0.08378739 - time (sec): 20.46 - samples/sec: 1443.31 - lr: 0.000029 - momentum: 0.000000
2024-03-26 10:05:59,072 epoch 5 - iter 44/48 - loss 0.08769191 - time (sec): 22.40 - samples/sec: 1441.80 - lr: 0.000029 - momentum: 0.000000
2024-03-26 10:06:00,117 epoch 5 - iter 48/48 - loss 0.08679457 - time (sec): 23.45 - samples/sec: 1470.09 - lr: 0.000028 - momentum: 0.000000
2024-03-26 10:06:00,117 ----------------------------------------------------------------------------------------------------
2024-03-26 10:06:00,117 EPOCH 5 done: loss 0.0868 - lr: 0.000028
2024-03-26 10:06:01,011 DEV : loss 0.1546858549118042 - f1-score (micro avg) 0.9175
2024-03-26 10:06:01,012 saving best model
2024-03-26 10:06:01,464 ----------------------------------------------------------------------------------------------------
2024-03-26 10:06:04,054 epoch 6 - iter 4/48 - loss 0.06321178 - time (sec): 2.59 - samples/sec: 1228.55 - lr: 0.000028 - momentum: 0.000000
2024-03-26 10:06:06,037 epoch 6 - iter 8/48 - loss 0.06414214 - time (sec): 4.57 - samples/sec: 1284.40 - lr: 0.000027 - momentum: 0.000000
2024-03-26 10:06:07,599 epoch 6 - iter 12/48 - loss 0.06581955 - time (sec): 6.13 - samples/sec: 1438.56 - lr: 0.000027 - momentum: 0.000000
2024-03-26 10:06:09,552 epoch 6 - iter 16/48 - loss 0.06180199 - time (sec): 8.09 - samples/sec: 1438.19 - lr: 0.000026 - momentum: 0.000000
2024-03-26 10:06:10,624 epoch 6 - iter 20/48 - loss 0.06294844 - time (sec): 9.16 - samples/sec: 1525.74 - lr: 0.000026 - momentum: 0.000000
2024-03-26 10:06:12,543 epoch 6 - iter 24/48 - loss 0.06365673 - time (sec): 11.08 - samples/sec: 1508.95 - lr: 0.000025 - momentum: 0.000000
2024-03-26 10:06:13,683 epoch 6 - iter 28/48 - loss 0.06347676 - time (sec): 12.22 - samples/sec: 1556.99 - lr: 0.000025 - momentum: 0.000000
2024-03-26 10:06:15,443 epoch 6 - iter 32/48 - loss 0.06064851 - time (sec): 13.98 - samples/sec: 1576.17 - lr: 0.000024 - momentum: 0.000000
2024-03-26 10:06:17,849 epoch 6 - iter 36/48 - loss 0.07068180 - time (sec): 16.38 - samples/sec: 1549.98 - lr: 0.000024 - momentum: 0.000000
2024-03-26 10:06:19,898 epoch 6 - iter 40/48 - loss 0.06791519 - time (sec): 18.43 - samples/sec: 1541.53 - lr: 0.000023 - momentum: 0.000000
2024-03-26 10:06:21,769 epoch 6 - iter 44/48 - loss 0.06892305 - time (sec): 20.30 - samples/sec: 1549.34 - lr: 0.000023 - momentum: 0.000000
2024-03-26 10:06:23,311 epoch 6 - iter 48/48 - loss 0.06987396 - time (sec): 21.85 - samples/sec: 1577.99 - lr: 0.000023 - momentum: 0.000000
2024-03-26 10:06:23,312 ----------------------------------------------------------------------------------------------------
2024-03-26 10:06:23,312 EPOCH 6 done: loss 0.0699 - lr: 0.000023
2024-03-26 10:06:24,215 DEV : loss 0.15499736368656158 - f1-score (micro avg) 0.9189
2024-03-26 10:06:24,216 saving best model
2024-03-26 10:06:24,665 ----------------------------------------------------------------------------------------------------
2024-03-26 10:06:26,755 epoch 7 - iter 4/48 - loss 0.05729350 - time (sec): 2.09 - samples/sec: 1324.11 - lr: 0.000022 - momentum: 0.000000
2024-03-26 10:06:28,456 epoch 7 - iter 8/48 - loss 0.04997245 - time (sec): 3.79 - samples/sec: 1516.79 - lr: 0.000022 - momentum: 0.000000
2024-03-26 10:06:30,515 epoch 7 - iter 12/48 - loss 0.04178798 - time (sec): 5.85 - samples/sec: 1465.31 - lr: 0.000021 - momentum: 0.000000
2024-03-26 10:06:33,077 epoch 7 - iter 16/48 - loss 0.04138225 - time (sec): 8.41 - samples/sec: 1405.52 - lr: 0.000021 - momentum: 0.000000
2024-03-26 10:06:35,748 epoch 7 - iter 20/48 - loss 0.04467645 - time (sec): 11.08 - samples/sec: 1410.66 - lr: 0.000020 - momentum: 0.000000
2024-03-26 10:06:37,266 epoch 7 - iter 24/48 - loss 0.04533926 - time (sec): 12.60 - samples/sec: 1430.71 - lr: 0.000020 - momentum: 0.000000
2024-03-26 10:06:39,346 epoch 7 - iter 28/48 - loss 0.04288749 - time (sec): 14.68 - samples/sec: 1449.37 - lr: 0.000019 - momentum: 0.000000
2024-03-26 10:06:41,482 epoch 7 - iter 32/48 - loss 0.04691273 - time (sec): 16.81 - samples/sec: 1454.03 - lr: 0.000019 - momentum: 0.000000
2024-03-26 10:06:43,694 epoch 7 - iter 36/48 - loss 0.05060029 - time (sec): 19.03 - samples/sec: 1438.82 - lr: 0.000018 - momentum: 0.000000
2024-03-26 10:06:45,277 epoch 7 - iter 40/48 - loss 0.04841109 - time (sec): 20.61 - samples/sec: 1446.69 - lr: 0.000018 - momentum: 0.000000
2024-03-26 10:06:46,932 epoch 7 - iter 44/48 - loss 0.05180454 - time (sec): 22.26 - samples/sec: 1465.06 - lr: 0.000017 - momentum: 0.000000
2024-03-26 10:06:48,258 epoch 7 - iter 48/48 - loss 0.05225955 - time (sec): 23.59 - samples/sec: 1461.28 - lr: 0.000017 - momentum: 0.000000
2024-03-26 10:06:48,258 ----------------------------------------------------------------------------------------------------
2024-03-26 10:06:48,258 EPOCH 7 done: loss 0.0523 - lr: 0.000017
2024-03-26 10:06:49,158 DEV : loss 0.14614951610565186 - f1-score (micro avg) 0.9302
2024-03-26 10:06:49,159 saving best model
2024-03-26 10:06:49,691 ----------------------------------------------------------------------------------------------------
2024-03-26 10:06:52,031 epoch 8 - iter 4/48 - loss 0.03268631 - time (sec): 2.34 - samples/sec: 1257.01 - lr: 0.000017 - momentum: 0.000000
2024-03-26 10:06:54,563 epoch 8 - iter 8/48 - loss 0.03038363 - time (sec): 4.87 - samples/sec: 1358.16 - lr: 0.000016 - momentum: 0.000000
2024-03-26 10:06:56,543 epoch 8 - iter 12/48 - loss 0.03050544 - time (sec): 6.85 - samples/sec: 1343.37 - lr: 0.000016 - momentum: 0.000000
2024-03-26 10:06:58,522 epoch 8 - iter 16/48 - loss 0.03045574 - time (sec): 8.83 - samples/sec: 1358.09 - lr: 0.000015 - momentum: 0.000000
2024-03-26 10:07:00,040 epoch 8 - iter 20/48 - loss 0.03185719 - time (sec): 10.35 - samples/sec: 1381.75 - lr: 0.000015 - momentum: 0.000000
2024-03-26 10:07:02,371 epoch 8 - iter 24/48 - loss 0.03213468 - time (sec): 12.68 - samples/sec: 1367.90 - lr: 0.000014 - momentum: 0.000000
2024-03-26 10:07:04,500 epoch 8 - iter 28/48 - loss 0.03166780 - time (sec): 14.81 - samples/sec: 1361.22 - lr: 0.000014 - momentum: 0.000000
2024-03-26 10:07:06,790 epoch 8 - iter 32/48 - loss 0.03896785 - time (sec): 17.10 - samples/sec: 1373.13 - lr: 0.000013 - momentum: 0.000000
2024-03-26 10:07:09,945 epoch 8 - iter 36/48 - loss 0.04158676 - time (sec): 20.25 - samples/sec: 1325.12 - lr: 0.000013 - momentum: 0.000000
2024-03-26 10:07:11,914 epoch 8 - iter 40/48 - loss 0.04704426 - time (sec): 22.22 - samples/sec: 1331.89 - lr: 0.000012 - momentum: 0.000000
2024-03-26 10:07:12,710 epoch 8 - iter 44/48 - loss 0.04633177 - time (sec): 23.02 - samples/sec: 1380.29 - lr: 0.000012 - momentum: 0.000000
2024-03-26 10:07:14,516 epoch 8 - iter 48/48 - loss 0.04472291 - time (sec): 24.82 - samples/sec: 1388.72 - lr: 0.000011 - momentum: 0.000000
2024-03-26 10:07:14,516 ----------------------------------------------------------------------------------------------------
2024-03-26 10:07:14,516 EPOCH 8 done: loss 0.0447 - lr: 0.000011
2024-03-26 10:07:15,418 DEV : loss 0.1479821503162384 - f1-score (micro avg) 0.9288
2024-03-26 10:07:15,419 ----------------------------------------------------------------------------------------------------
2024-03-26 10:07:18,093 epoch 9 - iter 4/48 - loss 0.01932148 - time (sec): 2.67 - samples/sec: 1233.74 - lr: 0.000011 - momentum: 0.000000
2024-03-26 10:07:19,745 epoch 9 - iter 8/48 - loss 0.02825076 - time (sec): 4.32 - samples/sec: 1326.82 - lr: 0.000011 - momentum: 0.000000
2024-03-26 10:07:21,833 epoch 9 - iter 12/48 - loss 0.03408380 - time (sec): 6.41 - samples/sec: 1399.55 - lr: 0.000010 - momentum: 0.000000
2024-03-26 10:07:23,896 epoch 9 - iter 16/48 - loss 0.03627351 - time (sec): 8.48 - samples/sec: 1428.64 - lr: 0.000010 - momentum: 0.000000
2024-03-26 10:07:26,193 epoch 9 - iter 20/48 - loss 0.03172262 - time (sec): 10.77 - samples/sec: 1404.36 - lr: 0.000009 - momentum: 0.000000
2024-03-26 10:07:28,092 epoch 9 - iter 24/48 - loss 0.03134365 - time (sec): 12.67 - samples/sec: 1397.35 - lr: 0.000009 - momentum: 0.000000
2024-03-26 10:07:31,265 epoch 9 - iter 28/48 - loss 0.03364734 - time (sec): 15.84 - samples/sec: 1348.27 - lr: 0.000008 - momentum: 0.000000
2024-03-26 10:07:32,622 epoch 9 - iter 32/48 - loss 0.03433911 - time (sec): 17.20 - samples/sec: 1388.75 - lr: 0.000008 - momentum: 0.000000
2024-03-26 10:07:34,972 epoch 9 - iter 36/48 - loss 0.03454591 - time (sec): 19.55 - samples/sec: 1378.29 - lr: 0.000007 - momentum: 0.000000
2024-03-26 10:07:36,421 epoch 9 - iter 40/48 - loss 0.03528763 - time (sec): 21.00 - samples/sec: 1396.11 - lr: 0.000007 - momentum: 0.000000
2024-03-26 10:07:37,918 epoch 9 - iter 44/48 - loss 0.03734495 - time (sec): 22.50 - samples/sec: 1413.93 - lr: 0.000006 - momentum: 0.000000
2024-03-26 10:07:39,311 epoch 9 - iter 48/48 - loss 0.03725859 - time (sec): 23.89 - samples/sec: 1442.89 - lr: 0.000006 - momentum: 0.000000
2024-03-26 10:07:39,311 ----------------------------------------------------------------------------------------------------
2024-03-26 10:07:39,311 EPOCH 9 done: loss 0.0373 - lr: 0.000006
2024-03-26 10:07:40,211 DEV : loss 0.14975792169570923 - f1-score (micro avg) 0.9318
2024-03-26 10:07:40,212 saving best model
2024-03-26 10:07:40,684 ----------------------------------------------------------------------------------------------------
2024-03-26 10:07:43,043 epoch 10 - iter 4/48 - loss 0.01926073 - time (sec): 2.36 - samples/sec: 1396.63 - lr: 0.000006 - momentum: 0.000000
2024-03-26 10:07:44,946 epoch 10 - iter 8/48 - loss 0.01598110 - time (sec): 4.26 - samples/sec: 1372.00 - lr: 0.000005 - momentum: 0.000000
2024-03-26 10:07:46,127 epoch 10 - iter 12/48 - loss 0.02607911 - time (sec): 5.44 - samples/sec: 1533.24 - lr: 0.000005 - momentum: 0.000000
2024-03-26 10:07:47,639 epoch 10 - iter 16/48 - loss 0.03016914 - time (sec): 6.95 - samples/sec: 1613.71 - lr: 0.000004 - momentum: 0.000000
2024-03-26 10:07:49,327 epoch 10 - iter 20/48 - loss 0.03002684 - time (sec): 8.64 - samples/sec: 1652.04 - lr: 0.000004 - momentum: 0.000000
2024-03-26 10:07:51,329 epoch 10 - iter 24/48 - loss 0.02958951 - time (sec): 10.64 - samples/sec: 1601.27 - lr: 0.000003 - momentum: 0.000000
2024-03-26 10:07:53,434 epoch 10 - iter 28/48 - loss 0.02712996 - time (sec): 12.75 - samples/sec: 1564.85 - lr: 0.000003 - momentum: 0.000000
2024-03-26 10:07:55,523 epoch 10 - iter 32/48 - loss 0.02875308 - time (sec): 14.84 - samples/sec: 1572.48 - lr: 0.000002 - momentum: 0.000000
2024-03-26 10:07:56,902 epoch 10 - iter 36/48 - loss 0.02840770 - time (sec): 16.22 - samples/sec: 1572.50 - lr: 0.000002 - momentum: 0.000000
2024-03-26 10:07:59,501 epoch 10 - iter 40/48 - loss 0.02707974 - time (sec): 18.82 - samples/sec: 1533.75 - lr: 0.000001 - momentum: 0.000000
2024-03-26 10:08:01,999 epoch 10 - iter 44/48 - loss 0.03012453 - time (sec): 21.31 - samples/sec: 1508.73 - lr: 0.000001 - momentum: 0.000000
2024-03-26 10:08:03,574 epoch 10 - iter 48/48 - loss 0.03064875 - time (sec): 22.89 - samples/sec: 1506.14 - lr: 0.000000 - momentum: 0.000000
2024-03-26 10:08:03,574 ----------------------------------------------------------------------------------------------------
2024-03-26 10:08:03,574 EPOCH 10 done: loss 0.0306 - lr: 0.000000
2024-03-26 10:08:04,469 DEV : loss 0.14895965158939362 - f1-score (micro avg) 0.9334
2024-03-26 10:08:04,470 saving best model
2024-03-26 10:08:05,226 ----------------------------------------------------------------------------------------------------
2024-03-26 10:08:05,226 Loading model from best epoch ...
2024-03-26 10:08:06,139 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 10:08:06,969
Results:
- F-score (micro) 0.9119
- F-score (macro) 0.6922
- Accuracy 0.8428
By class:
precision recall f1-score support
Unternehmen 0.9183 0.8872 0.9025 266
Auslagerung 0.8731 0.9116 0.8919 249
Ort 0.9635 0.9851 0.9742 134
Software 0.0000 0.0000 0.0000 0
micro avg 0.9070 0.9168 0.9119 649
macro avg 0.6887 0.6960 0.6922 649
weighted avg 0.9103 0.9168 0.9132 649
2024-03-26 10:08:06,969 ----------------------------------------------------------------------------------------------------
|