File size: 26,621 Bytes
95f036a |
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 |
2024-03-26 15:31:07,433 ----------------------------------------------------------------------------------------------------
2024-03-26 15:31:07,433 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 15:31:07,434 ----------------------------------------------------------------------------------------------------
2024-03-26 15:31:07,434 Corpus: 758 train + 94 dev + 96 test sentences
2024-03-26 15:31:07,434 ----------------------------------------------------------------------------------------------------
2024-03-26 15:31:07,434 Train: 758 sentences
2024-03-26 15:31:07,434 (train_with_dev=False, train_with_test=False)
2024-03-26 15:31:07,434 ----------------------------------------------------------------------------------------------------
2024-03-26 15:31:07,434 Training Params:
2024-03-26 15:31:07,434 - learning_rate: "3e-05"
2024-03-26 15:31:07,434 - mini_batch_size: "16"
2024-03-26 15:31:07,434 - max_epochs: "10"
2024-03-26 15:31:07,434 - shuffle: "True"
2024-03-26 15:31:07,434 ----------------------------------------------------------------------------------------------------
2024-03-26 15:31:07,434 Plugins:
2024-03-26 15:31:07,434 - TensorboardLogger
2024-03-26 15:31:07,434 - LinearScheduler | warmup_fraction: '0.1'
2024-03-26 15:31:07,434 ----------------------------------------------------------------------------------------------------
2024-03-26 15:31:07,434 Final evaluation on model from best epoch (best-model.pt)
2024-03-26 15:31:07,434 - metric: "('micro avg', 'f1-score')"
2024-03-26 15:31:07,434 ----------------------------------------------------------------------------------------------------
2024-03-26 15:31:07,434 Computation:
2024-03-26 15:31:07,434 - compute on device: cuda:0
2024-03-26 15:31:07,434 - embedding storage: none
2024-03-26 15:31:07,434 ----------------------------------------------------------------------------------------------------
2024-03-26 15:31:07,434 Model training base path: "flair-co-funer-german_dbmdz_bert_base-bs16-e10-lr3e-05-2"
2024-03-26 15:31:07,434 ----------------------------------------------------------------------------------------------------
2024-03-26 15:31:07,434 ----------------------------------------------------------------------------------------------------
2024-03-26 15:31:07,434 Logging anything other than scalars to TensorBoard is currently not supported.
2024-03-26 15:31:09,153 epoch 1 - iter 4/48 - loss 3.07690001 - time (sec): 1.72 - samples/sec: 1757.53 - lr: 0.000002 - momentum: 0.000000
2024-03-26 15:31:11,233 epoch 1 - iter 8/48 - loss 3.05496224 - time (sec): 3.80 - samples/sec: 1634.29 - lr: 0.000004 - momentum: 0.000000
2024-03-26 15:31:13,072 epoch 1 - iter 12/48 - loss 2.97386894 - time (sec): 5.64 - samples/sec: 1581.12 - lr: 0.000007 - momentum: 0.000000
2024-03-26 15:31:15,073 epoch 1 - iter 16/48 - loss 2.85293571 - time (sec): 7.64 - samples/sec: 1588.44 - lr: 0.000009 - momentum: 0.000000
2024-03-26 15:31:17,248 epoch 1 - iter 20/48 - loss 2.74493363 - time (sec): 9.81 - samples/sec: 1557.15 - lr: 0.000012 - momentum: 0.000000
2024-03-26 15:31:20,254 epoch 1 - iter 24/48 - loss 2.64581724 - time (sec): 12.82 - samples/sec: 1418.04 - lr: 0.000014 - momentum: 0.000000
2024-03-26 15:31:22,640 epoch 1 - iter 28/48 - loss 2.52600763 - time (sec): 15.21 - samples/sec: 1401.69 - lr: 0.000017 - momentum: 0.000000
2024-03-26 15:31:23,456 epoch 1 - iter 32/48 - loss 2.44927452 - time (sec): 16.02 - samples/sec: 1457.37 - lr: 0.000019 - momentum: 0.000000
2024-03-26 15:31:24,710 epoch 1 - iter 36/48 - loss 2.36202901 - time (sec): 17.28 - samples/sec: 1513.78 - lr: 0.000022 - momentum: 0.000000
2024-03-26 15:31:26,567 epoch 1 - iter 40/48 - loss 2.28016152 - time (sec): 19.13 - samples/sec: 1520.48 - lr: 0.000024 - momentum: 0.000000
2024-03-26 15:31:28,440 epoch 1 - iter 44/48 - loss 2.19008437 - time (sec): 21.01 - samples/sec: 1521.10 - lr: 0.000027 - momentum: 0.000000
2024-03-26 15:31:29,788 epoch 1 - iter 48/48 - loss 2.11466347 - time (sec): 22.35 - samples/sec: 1542.10 - lr: 0.000029 - momentum: 0.000000
2024-03-26 15:31:29,788 ----------------------------------------------------------------------------------------------------
2024-03-26 15:31:29,788 EPOCH 1 done: loss 2.1147 - lr: 0.000029
2024-03-26 15:31:30,611 DEV : loss 0.8351971507072449 - f1-score (micro avg) 0.4472
2024-03-26 15:31:30,612 saving best model
2024-03-26 15:31:30,882 ----------------------------------------------------------------------------------------------------
2024-03-26 15:31:32,192 epoch 2 - iter 4/48 - loss 1.15516706 - time (sec): 1.31 - samples/sec: 2214.60 - lr: 0.000030 - momentum: 0.000000
2024-03-26 15:31:34,020 epoch 2 - iter 8/48 - loss 0.97683226 - time (sec): 3.14 - samples/sec: 1943.54 - lr: 0.000030 - momentum: 0.000000
2024-03-26 15:31:37,458 epoch 2 - iter 12/48 - loss 0.86710735 - time (sec): 6.58 - samples/sec: 1547.82 - lr: 0.000029 - momentum: 0.000000
2024-03-26 15:31:39,937 epoch 2 - iter 16/48 - loss 0.80348775 - time (sec): 9.05 - samples/sec: 1471.03 - lr: 0.000029 - momentum: 0.000000
2024-03-26 15:31:42,595 epoch 2 - iter 20/48 - loss 0.75215888 - time (sec): 11.71 - samples/sec: 1418.35 - lr: 0.000029 - momentum: 0.000000
2024-03-26 15:31:44,488 epoch 2 - iter 24/48 - loss 0.70436302 - time (sec): 13.61 - samples/sec: 1416.97 - lr: 0.000028 - momentum: 0.000000
2024-03-26 15:31:46,268 epoch 2 - iter 28/48 - loss 0.69084455 - time (sec): 15.39 - samples/sec: 1425.51 - lr: 0.000028 - momentum: 0.000000
2024-03-26 15:31:47,991 epoch 2 - iter 32/48 - loss 0.67192253 - time (sec): 17.11 - samples/sec: 1438.14 - lr: 0.000028 - momentum: 0.000000
2024-03-26 15:31:49,842 epoch 2 - iter 36/48 - loss 0.65489588 - time (sec): 18.96 - samples/sec: 1446.86 - lr: 0.000028 - momentum: 0.000000
2024-03-26 15:31:50,863 epoch 2 - iter 40/48 - loss 0.63872466 - time (sec): 19.98 - samples/sec: 1494.02 - lr: 0.000027 - momentum: 0.000000
2024-03-26 15:31:52,297 epoch 2 - iter 44/48 - loss 0.62970353 - time (sec): 21.41 - samples/sec: 1513.73 - lr: 0.000027 - momentum: 0.000000
2024-03-26 15:31:53,824 epoch 2 - iter 48/48 - loss 0.61156947 - time (sec): 22.94 - samples/sec: 1502.60 - lr: 0.000027 - momentum: 0.000000
2024-03-26 15:31:53,824 ----------------------------------------------------------------------------------------------------
2024-03-26 15:31:53,824 EPOCH 2 done: loss 0.6116 - lr: 0.000027
2024-03-26 15:31:54,747 DEV : loss 0.33092811703681946 - f1-score (micro avg) 0.8046
2024-03-26 15:31:54,748 saving best model
2024-03-26 15:31:55,217 ----------------------------------------------------------------------------------------------------
2024-03-26 15:31:57,865 epoch 3 - iter 4/48 - loss 0.33780163 - time (sec): 2.65 - samples/sec: 1136.57 - lr: 0.000026 - momentum: 0.000000
2024-03-26 15:32:00,011 epoch 3 - iter 8/48 - loss 0.33754782 - time (sec): 4.79 - samples/sec: 1324.89 - lr: 0.000026 - momentum: 0.000000
2024-03-26 15:32:01,586 epoch 3 - iter 12/48 - loss 0.35771746 - time (sec): 6.37 - samples/sec: 1393.06 - lr: 0.000026 - momentum: 0.000000
2024-03-26 15:32:03,330 epoch 3 - iter 16/48 - loss 0.33541111 - time (sec): 8.11 - samples/sec: 1401.05 - lr: 0.000026 - momentum: 0.000000
2024-03-26 15:32:04,480 epoch 3 - iter 20/48 - loss 0.33566464 - time (sec): 9.26 - samples/sec: 1477.25 - lr: 0.000025 - momentum: 0.000000
2024-03-26 15:32:06,324 epoch 3 - iter 24/48 - loss 0.34059768 - time (sec): 11.11 - samples/sec: 1481.53 - lr: 0.000025 - momentum: 0.000000
2024-03-26 15:32:08,770 epoch 3 - iter 28/48 - loss 0.33468202 - time (sec): 13.55 - samples/sec: 1427.62 - lr: 0.000025 - momentum: 0.000000
2024-03-26 15:32:10,614 epoch 3 - iter 32/48 - loss 0.33237610 - time (sec): 15.40 - samples/sec: 1437.92 - lr: 0.000025 - momentum: 0.000000
2024-03-26 15:32:12,049 epoch 3 - iter 36/48 - loss 0.32281669 - time (sec): 16.83 - samples/sec: 1472.21 - lr: 0.000024 - momentum: 0.000000
2024-03-26 15:32:14,321 epoch 3 - iter 40/48 - loss 0.31113411 - time (sec): 19.10 - samples/sec: 1445.24 - lr: 0.000024 - momentum: 0.000000
2024-03-26 15:32:17,590 epoch 3 - iter 44/48 - loss 0.28678284 - time (sec): 22.37 - samples/sec: 1440.30 - lr: 0.000024 - momentum: 0.000000
2024-03-26 15:32:18,841 epoch 3 - iter 48/48 - loss 0.28114112 - time (sec): 23.62 - samples/sec: 1459.23 - lr: 0.000023 - momentum: 0.000000
2024-03-26 15:32:18,841 ----------------------------------------------------------------------------------------------------
2024-03-26 15:32:18,842 EPOCH 3 done: loss 0.2811 - lr: 0.000023
2024-03-26 15:32:19,759 DEV : loss 0.2615453898906708 - f1-score (micro avg) 0.8483
2024-03-26 15:32:19,761 saving best model
2024-03-26 15:32:20,220 ----------------------------------------------------------------------------------------------------
2024-03-26 15:32:21,779 epoch 4 - iter 4/48 - loss 0.27731467 - time (sec): 1.56 - samples/sec: 1636.53 - lr: 0.000023 - momentum: 0.000000
2024-03-26 15:32:23,988 epoch 4 - iter 8/48 - loss 0.23357535 - time (sec): 3.77 - samples/sec: 1590.61 - lr: 0.000023 - momentum: 0.000000
2024-03-26 15:32:25,233 epoch 4 - iter 12/48 - loss 0.21777199 - time (sec): 5.01 - samples/sec: 1667.59 - lr: 0.000023 - momentum: 0.000000
2024-03-26 15:32:27,454 epoch 4 - iter 16/48 - loss 0.21671924 - time (sec): 7.23 - samples/sec: 1558.49 - lr: 0.000022 - momentum: 0.000000
2024-03-26 15:32:29,973 epoch 4 - iter 20/48 - loss 0.20437028 - time (sec): 9.75 - samples/sec: 1433.65 - lr: 0.000022 - momentum: 0.000000
2024-03-26 15:32:31,983 epoch 4 - iter 24/48 - loss 0.20982545 - time (sec): 11.76 - samples/sec: 1431.14 - lr: 0.000022 - momentum: 0.000000
2024-03-26 15:32:34,093 epoch 4 - iter 28/48 - loss 0.20703194 - time (sec): 13.87 - samples/sec: 1434.09 - lr: 0.000022 - momentum: 0.000000
2024-03-26 15:32:36,637 epoch 4 - iter 32/48 - loss 0.20143413 - time (sec): 16.42 - samples/sec: 1404.71 - lr: 0.000021 - momentum: 0.000000
2024-03-26 15:32:39,428 epoch 4 - iter 36/48 - loss 0.19267077 - time (sec): 19.21 - samples/sec: 1392.72 - lr: 0.000021 - momentum: 0.000000
2024-03-26 15:32:41,108 epoch 4 - iter 40/48 - loss 0.18832931 - time (sec): 20.89 - samples/sec: 1392.91 - lr: 0.000021 - momentum: 0.000000
2024-03-26 15:32:43,083 epoch 4 - iter 44/48 - loss 0.18668573 - time (sec): 22.86 - samples/sec: 1396.30 - lr: 0.000020 - momentum: 0.000000
2024-03-26 15:32:44,737 epoch 4 - iter 48/48 - loss 0.18524935 - time (sec): 24.52 - samples/sec: 1406.10 - lr: 0.000020 - momentum: 0.000000
2024-03-26 15:32:44,737 ----------------------------------------------------------------------------------------------------
2024-03-26 15:32:44,737 EPOCH 4 done: loss 0.1852 - lr: 0.000020
2024-03-26 15:32:45,657 DEV : loss 0.22585716843605042 - f1-score (micro avg) 0.8805
2024-03-26 15:32:45,658 saving best model
2024-03-26 15:32:46,110 ----------------------------------------------------------------------------------------------------
2024-03-26 15:32:46,935 epoch 5 - iter 4/48 - loss 0.12116649 - time (sec): 0.82 - samples/sec: 2223.43 - lr: 0.000020 - momentum: 0.000000
2024-03-26 15:32:48,298 epoch 5 - iter 8/48 - loss 0.15184731 - time (sec): 2.19 - samples/sec: 2033.71 - lr: 0.000020 - momentum: 0.000000
2024-03-26 15:32:51,031 epoch 5 - iter 12/48 - loss 0.15337292 - time (sec): 4.92 - samples/sec: 1621.70 - lr: 0.000019 - momentum: 0.000000
2024-03-26 15:32:53,984 epoch 5 - iter 16/48 - loss 0.14647387 - time (sec): 7.87 - samples/sec: 1433.22 - lr: 0.000019 - momentum: 0.000000
2024-03-26 15:32:55,373 epoch 5 - iter 20/48 - loss 0.14839734 - time (sec): 9.26 - samples/sec: 1482.03 - lr: 0.000019 - momentum: 0.000000
2024-03-26 15:32:57,830 epoch 5 - iter 24/48 - loss 0.14399635 - time (sec): 11.72 - samples/sec: 1429.63 - lr: 0.000018 - momentum: 0.000000
2024-03-26 15:32:59,905 epoch 5 - iter 28/48 - loss 0.13921168 - time (sec): 13.79 - samples/sec: 1416.46 - lr: 0.000018 - momentum: 0.000000
2024-03-26 15:33:02,186 epoch 5 - iter 32/48 - loss 0.14270024 - time (sec): 16.08 - samples/sec: 1440.86 - lr: 0.000018 - momentum: 0.000000
2024-03-26 15:33:03,648 epoch 5 - iter 36/48 - loss 0.14605617 - time (sec): 17.54 - samples/sec: 1464.65 - lr: 0.000018 - momentum: 0.000000
2024-03-26 15:33:06,164 epoch 5 - iter 40/48 - loss 0.13970169 - time (sec): 20.05 - samples/sec: 1416.75 - lr: 0.000017 - momentum: 0.000000
2024-03-26 15:33:08,231 epoch 5 - iter 44/48 - loss 0.13750957 - time (sec): 22.12 - samples/sec: 1430.14 - lr: 0.000017 - momentum: 0.000000
2024-03-26 15:33:10,191 epoch 5 - iter 48/48 - loss 0.13714226 - time (sec): 24.08 - samples/sec: 1431.52 - lr: 0.000017 - momentum: 0.000000
2024-03-26 15:33:10,192 ----------------------------------------------------------------------------------------------------
2024-03-26 15:33:10,192 EPOCH 5 done: loss 0.1371 - lr: 0.000017
2024-03-26 15:33:11,115 DEV : loss 0.20046745240688324 - f1-score (micro avg) 0.8771
2024-03-26 15:33:11,116 ----------------------------------------------------------------------------------------------------
2024-03-26 15:33:12,676 epoch 6 - iter 4/48 - loss 0.10272289 - time (sec): 1.56 - samples/sec: 1596.25 - lr: 0.000017 - momentum: 0.000000
2024-03-26 15:33:15,066 epoch 6 - iter 8/48 - loss 0.09870474 - time (sec): 3.95 - samples/sec: 1620.23 - lr: 0.000016 - momentum: 0.000000
2024-03-26 15:33:16,991 epoch 6 - iter 12/48 - loss 0.10078181 - time (sec): 5.87 - samples/sec: 1541.84 - lr: 0.000016 - momentum: 0.000000
2024-03-26 15:33:19,000 epoch 6 - iter 16/48 - loss 0.09675325 - time (sec): 7.88 - samples/sec: 1538.21 - lr: 0.000016 - momentum: 0.000000
2024-03-26 15:33:21,737 epoch 6 - iter 20/48 - loss 0.09973824 - time (sec): 10.62 - samples/sec: 1504.31 - lr: 0.000015 - momentum: 0.000000
2024-03-26 15:33:23,239 epoch 6 - iter 24/48 - loss 0.11510382 - time (sec): 12.12 - samples/sec: 1526.93 - lr: 0.000015 - momentum: 0.000000
2024-03-26 15:33:24,605 epoch 6 - iter 28/48 - loss 0.11454603 - time (sec): 13.49 - samples/sec: 1532.30 - lr: 0.000015 - momentum: 0.000000
2024-03-26 15:33:25,766 epoch 6 - iter 32/48 - loss 0.11090914 - time (sec): 14.65 - samples/sec: 1552.90 - lr: 0.000015 - momentum: 0.000000
2024-03-26 15:33:27,228 epoch 6 - iter 36/48 - loss 0.10548167 - time (sec): 16.11 - samples/sec: 1584.76 - lr: 0.000014 - momentum: 0.000000
2024-03-26 15:33:29,120 epoch 6 - iter 40/48 - loss 0.10799176 - time (sec): 18.00 - samples/sec: 1574.06 - lr: 0.000014 - momentum: 0.000000
2024-03-26 15:33:31,286 epoch 6 - iter 44/48 - loss 0.10341451 - time (sec): 20.17 - samples/sec: 1594.19 - lr: 0.000014 - momentum: 0.000000
2024-03-26 15:33:32,955 epoch 6 - iter 48/48 - loss 0.10237983 - time (sec): 21.84 - samples/sec: 1578.46 - lr: 0.000014 - momentum: 0.000000
2024-03-26 15:33:32,956 ----------------------------------------------------------------------------------------------------
2024-03-26 15:33:32,956 EPOCH 6 done: loss 0.1024 - lr: 0.000014
2024-03-26 15:33:33,865 DEV : loss 0.1799185872077942 - f1-score (micro avg) 0.903
2024-03-26 15:33:33,867 saving best model
2024-03-26 15:33:34,313 ----------------------------------------------------------------------------------------------------
2024-03-26 15:33:35,930 epoch 7 - iter 4/48 - loss 0.07362542 - time (sec): 1.62 - samples/sec: 1506.35 - lr: 0.000013 - momentum: 0.000000
2024-03-26 15:33:37,620 epoch 7 - iter 8/48 - loss 0.07695036 - time (sec): 3.31 - samples/sec: 1497.95 - lr: 0.000013 - momentum: 0.000000
2024-03-26 15:33:39,724 epoch 7 - iter 12/48 - loss 0.08244330 - time (sec): 5.41 - samples/sec: 1454.46 - lr: 0.000013 - momentum: 0.000000
2024-03-26 15:33:41,740 epoch 7 - iter 16/48 - loss 0.08062151 - time (sec): 7.43 - samples/sec: 1500.21 - lr: 0.000012 - momentum: 0.000000
2024-03-26 15:33:42,377 epoch 7 - iter 20/48 - loss 0.07768609 - time (sec): 8.06 - samples/sec: 1607.06 - lr: 0.000012 - momentum: 0.000000
2024-03-26 15:33:43,958 epoch 7 - iter 24/48 - loss 0.07804374 - time (sec): 9.64 - samples/sec: 1588.68 - lr: 0.000012 - momentum: 0.000000
2024-03-26 15:33:46,793 epoch 7 - iter 28/48 - loss 0.07671140 - time (sec): 12.48 - samples/sec: 1492.31 - lr: 0.000012 - momentum: 0.000000
2024-03-26 15:33:49,542 epoch 7 - iter 32/48 - loss 0.07602401 - time (sec): 15.23 - samples/sec: 1422.69 - lr: 0.000011 - momentum: 0.000000
2024-03-26 15:33:52,262 epoch 7 - iter 36/48 - loss 0.07873676 - time (sec): 17.95 - samples/sec: 1436.35 - lr: 0.000011 - momentum: 0.000000
2024-03-26 15:33:54,220 epoch 7 - iter 40/48 - loss 0.08275888 - time (sec): 19.91 - samples/sec: 1444.10 - lr: 0.000011 - momentum: 0.000000
2024-03-26 15:33:56,756 epoch 7 - iter 44/48 - loss 0.08223865 - time (sec): 22.44 - samples/sec: 1419.37 - lr: 0.000010 - momentum: 0.000000
2024-03-26 15:33:58,492 epoch 7 - iter 48/48 - loss 0.08132377 - time (sec): 24.18 - samples/sec: 1425.72 - lr: 0.000010 - momentum: 0.000000
2024-03-26 15:33:58,492 ----------------------------------------------------------------------------------------------------
2024-03-26 15:33:58,492 EPOCH 7 done: loss 0.0813 - lr: 0.000010
2024-03-26 15:33:59,405 DEV : loss 0.17715860903263092 - f1-score (micro avg) 0.9062
2024-03-26 15:33:59,408 saving best model
2024-03-26 15:33:59,861 ----------------------------------------------------------------------------------------------------
2024-03-26 15:34:02,494 epoch 8 - iter 4/48 - loss 0.07748369 - time (sec): 2.63 - samples/sec: 1255.51 - lr: 0.000010 - momentum: 0.000000
2024-03-26 15:34:04,546 epoch 8 - iter 8/48 - loss 0.06084314 - time (sec): 4.68 - samples/sec: 1252.87 - lr: 0.000010 - momentum: 0.000000
2024-03-26 15:34:07,704 epoch 8 - iter 12/48 - loss 0.06309502 - time (sec): 7.84 - samples/sec: 1235.93 - lr: 0.000009 - momentum: 0.000000
2024-03-26 15:34:09,620 epoch 8 - iter 16/48 - loss 0.07258115 - time (sec): 9.76 - samples/sec: 1265.04 - lr: 0.000009 - momentum: 0.000000
2024-03-26 15:34:11,081 epoch 8 - iter 20/48 - loss 0.07019402 - time (sec): 11.22 - samples/sec: 1309.06 - lr: 0.000009 - momentum: 0.000000
2024-03-26 15:34:13,491 epoch 8 - iter 24/48 - loss 0.06964750 - time (sec): 13.63 - samples/sec: 1309.36 - lr: 0.000009 - momentum: 0.000000
2024-03-26 15:34:15,234 epoch 8 - iter 28/48 - loss 0.07274345 - time (sec): 15.37 - samples/sec: 1345.28 - lr: 0.000008 - momentum: 0.000000
2024-03-26 15:34:16,883 epoch 8 - iter 32/48 - loss 0.07155021 - time (sec): 17.02 - samples/sec: 1366.86 - lr: 0.000008 - momentum: 0.000000
2024-03-26 15:34:18,168 epoch 8 - iter 36/48 - loss 0.06972937 - time (sec): 18.30 - samples/sec: 1397.60 - lr: 0.000008 - momentum: 0.000000
2024-03-26 15:34:20,471 epoch 8 - iter 40/48 - loss 0.07030178 - time (sec): 20.61 - samples/sec: 1406.83 - lr: 0.000007 - momentum: 0.000000
2024-03-26 15:34:23,313 epoch 8 - iter 44/48 - loss 0.06695918 - time (sec): 23.45 - samples/sec: 1373.85 - lr: 0.000007 - momentum: 0.000000
2024-03-26 15:34:25,225 epoch 8 - iter 48/48 - loss 0.06596868 - time (sec): 25.36 - samples/sec: 1359.21 - lr: 0.000007 - momentum: 0.000000
2024-03-26 15:34:25,225 ----------------------------------------------------------------------------------------------------
2024-03-26 15:34:25,225 EPOCH 8 done: loss 0.0660 - lr: 0.000007
2024-03-26 15:34:26,138 DEV : loss 0.18558232486248016 - f1-score (micro avg) 0.9211
2024-03-26 15:34:26,141 saving best model
2024-03-26 15:34:26,605 ----------------------------------------------------------------------------------------------------
2024-03-26 15:34:28,417 epoch 9 - iter 4/48 - loss 0.06772714 - time (sec): 1.81 - samples/sec: 1570.73 - lr: 0.000007 - momentum: 0.000000
2024-03-26 15:34:30,821 epoch 9 - iter 8/48 - loss 0.05495595 - time (sec): 4.21 - samples/sec: 1454.92 - lr: 0.000006 - momentum: 0.000000
2024-03-26 15:34:33,163 epoch 9 - iter 12/48 - loss 0.06627844 - time (sec): 6.56 - samples/sec: 1407.75 - lr: 0.000006 - momentum: 0.000000
2024-03-26 15:34:35,189 epoch 9 - iter 16/48 - loss 0.06614638 - time (sec): 8.58 - samples/sec: 1409.16 - lr: 0.000006 - momentum: 0.000000
2024-03-26 15:34:36,638 epoch 9 - iter 20/48 - loss 0.05859714 - time (sec): 10.03 - samples/sec: 1469.04 - lr: 0.000006 - momentum: 0.000000
2024-03-26 15:34:37,839 epoch 9 - iter 24/48 - loss 0.05484876 - time (sec): 11.23 - samples/sec: 1516.66 - lr: 0.000005 - momentum: 0.000000
2024-03-26 15:34:39,529 epoch 9 - iter 28/48 - loss 0.05309367 - time (sec): 12.92 - samples/sec: 1530.38 - lr: 0.000005 - momentum: 0.000000
2024-03-26 15:34:41,777 epoch 9 - iter 32/48 - loss 0.05768250 - time (sec): 15.17 - samples/sec: 1515.76 - lr: 0.000005 - momentum: 0.000000
2024-03-26 15:34:44,442 epoch 9 - iter 36/48 - loss 0.05677322 - time (sec): 17.84 - samples/sec: 1464.59 - lr: 0.000004 - momentum: 0.000000
2024-03-26 15:34:47,361 epoch 9 - iter 40/48 - loss 0.05707459 - time (sec): 20.75 - samples/sec: 1420.11 - lr: 0.000004 - momentum: 0.000000
2024-03-26 15:34:49,162 epoch 9 - iter 44/48 - loss 0.05624355 - time (sec): 22.56 - samples/sec: 1435.56 - lr: 0.000004 - momentum: 0.000000
2024-03-26 15:34:50,192 epoch 9 - iter 48/48 - loss 0.05651383 - time (sec): 23.59 - samples/sec: 1461.56 - lr: 0.000004 - momentum: 0.000000
2024-03-26 15:34:50,192 ----------------------------------------------------------------------------------------------------
2024-03-26 15:34:50,192 EPOCH 9 done: loss 0.0565 - lr: 0.000004
2024-03-26 15:34:51,127 DEV : loss 0.18057239055633545 - f1-score (micro avg) 0.9321
2024-03-26 15:34:51,128 saving best model
2024-03-26 15:34:51,585 ----------------------------------------------------------------------------------------------------
2024-03-26 15:34:53,875 epoch 10 - iter 4/48 - loss 0.02487919 - time (sec): 2.29 - samples/sec: 1442.31 - lr: 0.000003 - momentum: 0.000000
2024-03-26 15:34:55,930 epoch 10 - iter 8/48 - loss 0.03646152 - time (sec): 4.34 - samples/sec: 1422.14 - lr: 0.000003 - momentum: 0.000000
2024-03-26 15:34:57,846 epoch 10 - iter 12/48 - loss 0.03623072 - time (sec): 6.26 - samples/sec: 1409.39 - lr: 0.000003 - momentum: 0.000000
2024-03-26 15:34:59,082 epoch 10 - iter 16/48 - loss 0.03960043 - time (sec): 7.50 - samples/sec: 1470.03 - lr: 0.000002 - momentum: 0.000000
2024-03-26 15:35:00,988 epoch 10 - iter 20/48 - loss 0.04632415 - time (sec): 9.40 - samples/sec: 1457.96 - lr: 0.000002 - momentum: 0.000000
2024-03-26 15:35:03,197 epoch 10 - iter 24/48 - loss 0.05286228 - time (sec): 11.61 - samples/sec: 1430.23 - lr: 0.000002 - momentum: 0.000000
2024-03-26 15:35:04,086 epoch 10 - iter 28/48 - loss 0.05410697 - time (sec): 12.50 - samples/sec: 1503.04 - lr: 0.000002 - momentum: 0.000000
2024-03-26 15:35:05,349 epoch 10 - iter 32/48 - loss 0.05294441 - time (sec): 13.76 - samples/sec: 1543.10 - lr: 0.000001 - momentum: 0.000000
2024-03-26 15:35:08,108 epoch 10 - iter 36/48 - loss 0.05023736 - time (sec): 16.52 - samples/sec: 1494.52 - lr: 0.000001 - momentum: 0.000000
2024-03-26 15:35:10,514 epoch 10 - iter 40/48 - loss 0.05048026 - time (sec): 18.93 - samples/sec: 1519.04 - lr: 0.000001 - momentum: 0.000000
2024-03-26 15:35:13,059 epoch 10 - iter 44/48 - loss 0.04935967 - time (sec): 21.47 - samples/sec: 1493.67 - lr: 0.000001 - momentum: 0.000000
2024-03-26 15:35:14,979 epoch 10 - iter 48/48 - loss 0.04860975 - time (sec): 23.39 - samples/sec: 1473.54 - lr: 0.000000 - momentum: 0.000000
2024-03-26 15:35:14,980 ----------------------------------------------------------------------------------------------------
2024-03-26 15:35:14,980 EPOCH 10 done: loss 0.0486 - lr: 0.000000
2024-03-26 15:35:15,900 DEV : loss 0.1853199601173401 - f1-score (micro avg) 0.9257
2024-03-26 15:35:16,184 ----------------------------------------------------------------------------------------------------
2024-03-26 15:35:16,185 Loading model from best epoch ...
2024-03-26 15:35:17,059 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 15:35:17,909
Results:
- F-score (micro) 0.8995
- F-score (macro) 0.6839
- Accuracy 0.8208
By class:
precision recall f1-score support
Unternehmen 0.9008 0.8872 0.8939 266
Auslagerung 0.8479 0.8956 0.8711 249
Ort 0.9565 0.9851 0.9706 134
Software 0.0000 0.0000 0.0000 0
micro avg 0.8887 0.9106 0.8995 649
macro avg 0.6763 0.6920 0.6839 649
weighted avg 0.8920 0.9106 0.9010 649
2024-03-26 15:35:17,909 ----------------------------------------------------------------------------------------------------
|