File size: 24,232 Bytes
d54efa1 |
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 |
2023-10-19 12:12:24,058 ----------------------------------------------------------------------------------------------------
2023-10-19 12:12:24,058 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): BertModel(
(embeddings): BertEmbeddings(
(word_embeddings): Embedding(32001, 128)
(position_embeddings): Embedding(512, 128)
(token_type_embeddings): Embedding(2, 128)
(LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(encoder): BertEncoder(
(layer): ModuleList(
(0-1): 2 x BertLayer(
(attention): BertAttention(
(self): BertSelfAttention(
(query): Linear(in_features=128, out_features=128, bias=True)
(key): Linear(in_features=128, out_features=128, bias=True)
(value): Linear(in_features=128, out_features=128, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(output): BertSelfOutput(
(dense): Linear(in_features=128, out_features=128, bias=True)
(LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(intermediate): BertIntermediate(
(dense): Linear(in_features=128, out_features=512, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): BertOutput(
(dense): Linear(in_features=512, out_features=128, bias=True)
(LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
(pooler): BertPooler(
(dense): Linear(in_features=128, out_features=128, bias=True)
(activation): Tanh()
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=128, out_features=17, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-19 12:12:24,058 ----------------------------------------------------------------------------------------------------
2023-10-19 12:12:24,058 MultiCorpus: 20847 train + 1123 dev + 3350 test sentences
- NER_HIPE_2022 Corpus: 20847 train + 1123 dev + 3350 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/de/with_doc_seperator
2023-10-19 12:12:24,058 ----------------------------------------------------------------------------------------------------
2023-10-19 12:12:24,058 Train: 20847 sentences
2023-10-19 12:12:24,058 (train_with_dev=False, train_with_test=False)
2023-10-19 12:12:24,058 ----------------------------------------------------------------------------------------------------
2023-10-19 12:12:24,058 Training Params:
2023-10-19 12:12:24,058 - learning_rate: "5e-05"
2023-10-19 12:12:24,058 - mini_batch_size: "8"
2023-10-19 12:12:24,058 - max_epochs: "10"
2023-10-19 12:12:24,058 - shuffle: "True"
2023-10-19 12:12:24,058 ----------------------------------------------------------------------------------------------------
2023-10-19 12:12:24,058 Plugins:
2023-10-19 12:12:24,058 - TensorboardLogger
2023-10-19 12:12:24,059 - LinearScheduler | warmup_fraction: '0.1'
2023-10-19 12:12:24,059 ----------------------------------------------------------------------------------------------------
2023-10-19 12:12:24,059 Final evaluation on model from best epoch (best-model.pt)
2023-10-19 12:12:24,059 - metric: "('micro avg', 'f1-score')"
2023-10-19 12:12:24,059 ----------------------------------------------------------------------------------------------------
2023-10-19 12:12:24,059 Computation:
2023-10-19 12:12:24,059 - compute on device: cuda:0
2023-10-19 12:12:24,059 - embedding storage: none
2023-10-19 12:12:24,059 ----------------------------------------------------------------------------------------------------
2023-10-19 12:12:24,059 Model training base path: "hmbench-newseye/de-dbmdz/bert-tiny-historic-multilingual-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-3"
2023-10-19 12:12:24,059 ----------------------------------------------------------------------------------------------------
2023-10-19 12:12:24,059 ----------------------------------------------------------------------------------------------------
2023-10-19 12:12:24,059 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-19 12:12:30,245 epoch 1 - iter 260/2606 - loss 3.29215868 - time (sec): 6.19 - samples/sec: 6125.45 - lr: 0.000005 - momentum: 0.000000
2023-10-19 12:12:36,395 epoch 1 - iter 520/2606 - loss 2.54663351 - time (sec): 12.34 - samples/sec: 6036.79 - lr: 0.000010 - momentum: 0.000000
2023-10-19 12:12:42,287 epoch 1 - iter 780/2606 - loss 1.98000977 - time (sec): 18.23 - samples/sec: 5915.66 - lr: 0.000015 - momentum: 0.000000
2023-10-19 12:12:48,508 epoch 1 - iter 1040/2606 - loss 1.60194669 - time (sec): 24.45 - samples/sec: 5924.28 - lr: 0.000020 - momentum: 0.000000
2023-10-19 12:12:54,610 epoch 1 - iter 1300/2606 - loss 1.39106299 - time (sec): 30.55 - samples/sec: 5902.93 - lr: 0.000025 - momentum: 0.000000
2023-10-19 12:13:01,049 epoch 1 - iter 1560/2606 - loss 1.23459866 - time (sec): 36.99 - samples/sec: 5904.98 - lr: 0.000030 - momentum: 0.000000
2023-10-19 12:13:07,064 epoch 1 - iter 1820/2606 - loss 1.13126722 - time (sec): 43.00 - samples/sec: 5895.47 - lr: 0.000035 - momentum: 0.000000
2023-10-19 12:13:13,149 epoch 1 - iter 2080/2606 - loss 1.04840712 - time (sec): 49.09 - samples/sec: 5904.15 - lr: 0.000040 - momentum: 0.000000
2023-10-19 12:13:19,453 epoch 1 - iter 2340/2606 - loss 0.97142483 - time (sec): 55.39 - samples/sec: 5935.98 - lr: 0.000045 - momentum: 0.000000
2023-10-19 12:13:25,404 epoch 1 - iter 2600/2606 - loss 0.91090610 - time (sec): 61.34 - samples/sec: 5977.09 - lr: 0.000050 - momentum: 0.000000
2023-10-19 12:13:25,524 ----------------------------------------------------------------------------------------------------
2023-10-19 12:13:25,524 EPOCH 1 done: loss 0.9097 - lr: 0.000050
2023-10-19 12:13:27,771 DEV : loss 0.14083856344223022 - f1-score (micro avg) 0.049
2023-10-19 12:13:27,795 saving best model
2023-10-19 12:13:27,827 ----------------------------------------------------------------------------------------------------
2023-10-19 12:13:33,910 epoch 2 - iter 260/2606 - loss 0.33750708 - time (sec): 6.08 - samples/sec: 6097.84 - lr: 0.000049 - momentum: 0.000000
2023-10-19 12:13:40,145 epoch 2 - iter 520/2606 - loss 0.34632728 - time (sec): 12.32 - samples/sec: 6177.17 - lr: 0.000049 - momentum: 0.000000
2023-10-19 12:13:46,357 epoch 2 - iter 780/2606 - loss 0.33738274 - time (sec): 18.53 - samples/sec: 6059.22 - lr: 0.000048 - momentum: 0.000000
2023-10-19 12:13:52,438 epoch 2 - iter 1040/2606 - loss 0.33995850 - time (sec): 24.61 - samples/sec: 6083.78 - lr: 0.000048 - momentum: 0.000000
2023-10-19 12:13:57,789 epoch 2 - iter 1300/2606 - loss 0.34055467 - time (sec): 29.96 - samples/sec: 6142.43 - lr: 0.000047 - momentum: 0.000000
2023-10-19 12:14:03,922 epoch 2 - iter 1560/2606 - loss 0.33755029 - time (sec): 36.09 - samples/sec: 6101.82 - lr: 0.000047 - momentum: 0.000000
2023-10-19 12:14:10,023 epoch 2 - iter 1820/2606 - loss 0.33322099 - time (sec): 42.19 - samples/sec: 6115.49 - lr: 0.000046 - momentum: 0.000000
2023-10-19 12:14:16,125 epoch 2 - iter 2080/2606 - loss 0.32900406 - time (sec): 48.30 - samples/sec: 6057.42 - lr: 0.000046 - momentum: 0.000000
2023-10-19 12:14:22,224 epoch 2 - iter 2340/2606 - loss 0.32781761 - time (sec): 54.40 - samples/sec: 6080.19 - lr: 0.000045 - momentum: 0.000000
2023-10-19 12:14:28,034 epoch 2 - iter 2600/2606 - loss 0.32378050 - time (sec): 60.21 - samples/sec: 6090.09 - lr: 0.000044 - momentum: 0.000000
2023-10-19 12:14:28,158 ----------------------------------------------------------------------------------------------------
2023-10-19 12:14:28,158 EPOCH 2 done: loss 0.3239 - lr: 0.000044
2023-10-19 12:14:33,294 DEV : loss 0.13072499632835388 - f1-score (micro avg) 0.2529
2023-10-19 12:14:33,317 saving best model
2023-10-19 12:14:33,350 ----------------------------------------------------------------------------------------------------
2023-10-19 12:14:39,474 epoch 3 - iter 260/2606 - loss 0.25574137 - time (sec): 6.12 - samples/sec: 5781.84 - lr: 0.000044 - momentum: 0.000000
2023-10-19 12:14:45,559 epoch 3 - iter 520/2606 - loss 0.26774606 - time (sec): 12.21 - samples/sec: 5992.37 - lr: 0.000043 - momentum: 0.000000
2023-10-19 12:14:51,497 epoch 3 - iter 780/2606 - loss 0.26961024 - time (sec): 18.15 - samples/sec: 5774.06 - lr: 0.000043 - momentum: 0.000000
2023-10-19 12:14:57,569 epoch 3 - iter 1040/2606 - loss 0.26961308 - time (sec): 24.22 - samples/sec: 5862.06 - lr: 0.000042 - momentum: 0.000000
2023-10-19 12:15:03,380 epoch 3 - iter 1300/2606 - loss 0.27095396 - time (sec): 30.03 - samples/sec: 5966.12 - lr: 0.000042 - momentum: 0.000000
2023-10-19 12:15:09,637 epoch 3 - iter 1560/2606 - loss 0.27006005 - time (sec): 36.29 - samples/sec: 6027.28 - lr: 0.000041 - momentum: 0.000000
2023-10-19 12:15:15,650 epoch 3 - iter 1820/2606 - loss 0.26808602 - time (sec): 42.30 - samples/sec: 6043.76 - lr: 0.000041 - momentum: 0.000000
2023-10-19 12:15:22,109 epoch 3 - iter 2080/2606 - loss 0.26544069 - time (sec): 48.76 - samples/sec: 6025.06 - lr: 0.000040 - momentum: 0.000000
2023-10-19 12:15:28,256 epoch 3 - iter 2340/2606 - loss 0.26734771 - time (sec): 54.91 - samples/sec: 6006.07 - lr: 0.000039 - momentum: 0.000000
2023-10-19 12:15:34,479 epoch 3 - iter 2600/2606 - loss 0.26623336 - time (sec): 61.13 - samples/sec: 6001.01 - lr: 0.000039 - momentum: 0.000000
2023-10-19 12:15:34,625 ----------------------------------------------------------------------------------------------------
2023-10-19 12:15:34,625 EPOCH 3 done: loss 0.2661 - lr: 0.000039
2023-10-19 12:15:39,746 DEV : loss 0.14174337685108185 - f1-score (micro avg) 0.2721
2023-10-19 12:15:39,769 saving best model
2023-10-19 12:15:39,801 ----------------------------------------------------------------------------------------------------
2023-10-19 12:15:45,896 epoch 4 - iter 260/2606 - loss 0.25139617 - time (sec): 6.09 - samples/sec: 6246.82 - lr: 0.000038 - momentum: 0.000000
2023-10-19 12:15:52,182 epoch 4 - iter 520/2606 - loss 0.23480749 - time (sec): 12.38 - samples/sec: 6268.11 - lr: 0.000038 - momentum: 0.000000
2023-10-19 12:15:58,276 epoch 4 - iter 780/2606 - loss 0.24352221 - time (sec): 18.47 - samples/sec: 6182.49 - lr: 0.000037 - momentum: 0.000000
2023-10-19 12:16:04,376 epoch 4 - iter 1040/2606 - loss 0.24831904 - time (sec): 24.57 - samples/sec: 6075.18 - lr: 0.000037 - momentum: 0.000000
2023-10-19 12:16:10,652 epoch 4 - iter 1300/2606 - loss 0.24218416 - time (sec): 30.85 - samples/sec: 6009.23 - lr: 0.000036 - momentum: 0.000000
2023-10-19 12:16:16,800 epoch 4 - iter 1560/2606 - loss 0.23924425 - time (sec): 37.00 - samples/sec: 5955.66 - lr: 0.000036 - momentum: 0.000000
2023-10-19 12:16:23,005 epoch 4 - iter 1820/2606 - loss 0.23697081 - time (sec): 43.20 - samples/sec: 5993.23 - lr: 0.000035 - momentum: 0.000000
2023-10-19 12:16:29,000 epoch 4 - iter 2080/2606 - loss 0.23564226 - time (sec): 49.20 - samples/sec: 5965.50 - lr: 0.000034 - momentum: 0.000000
2023-10-19 12:16:35,148 epoch 4 - iter 2340/2606 - loss 0.23533143 - time (sec): 55.35 - samples/sec: 5934.69 - lr: 0.000034 - momentum: 0.000000
2023-10-19 12:16:41,289 epoch 4 - iter 2600/2606 - loss 0.23376594 - time (sec): 61.49 - samples/sec: 5959.34 - lr: 0.000033 - momentum: 0.000000
2023-10-19 12:16:41,416 ----------------------------------------------------------------------------------------------------
2023-10-19 12:16:41,417 EPOCH 4 done: loss 0.2338 - lr: 0.000033
2023-10-19 12:16:46,618 DEV : loss 0.14498400688171387 - f1-score (micro avg) 0.2627
2023-10-19 12:16:46,641 ----------------------------------------------------------------------------------------------------
2023-10-19 12:16:52,681 epoch 5 - iter 260/2606 - loss 0.19023305 - time (sec): 6.04 - samples/sec: 5769.39 - lr: 0.000033 - momentum: 0.000000
2023-10-19 12:16:58,701 epoch 5 - iter 520/2606 - loss 0.21196067 - time (sec): 12.06 - samples/sec: 5855.59 - lr: 0.000032 - momentum: 0.000000
2023-10-19 12:17:04,898 epoch 5 - iter 780/2606 - loss 0.21458042 - time (sec): 18.26 - samples/sec: 5909.45 - lr: 0.000032 - momentum: 0.000000
2023-10-19 12:17:11,106 epoch 5 - iter 1040/2606 - loss 0.21431244 - time (sec): 24.46 - samples/sec: 5961.35 - lr: 0.000031 - momentum: 0.000000
2023-10-19 12:17:17,204 epoch 5 - iter 1300/2606 - loss 0.21745310 - time (sec): 30.56 - samples/sec: 5900.25 - lr: 0.000031 - momentum: 0.000000
2023-10-19 12:17:23,385 epoch 5 - iter 1560/2606 - loss 0.21540934 - time (sec): 36.74 - samples/sec: 5920.43 - lr: 0.000030 - momentum: 0.000000
2023-10-19 12:17:29,536 epoch 5 - iter 1820/2606 - loss 0.21605849 - time (sec): 42.89 - samples/sec: 5966.77 - lr: 0.000029 - momentum: 0.000000
2023-10-19 12:17:35,911 epoch 5 - iter 2080/2606 - loss 0.21388261 - time (sec): 49.27 - samples/sec: 5952.00 - lr: 0.000029 - momentum: 0.000000
2023-10-19 12:17:42,028 epoch 5 - iter 2340/2606 - loss 0.21150192 - time (sec): 55.39 - samples/sec: 5950.63 - lr: 0.000028 - momentum: 0.000000
2023-10-19 12:17:48,297 epoch 5 - iter 2600/2606 - loss 0.21236748 - time (sec): 61.65 - samples/sec: 5944.70 - lr: 0.000028 - momentum: 0.000000
2023-10-19 12:17:48,443 ----------------------------------------------------------------------------------------------------
2023-10-19 12:17:48,443 EPOCH 5 done: loss 0.2122 - lr: 0.000028
2023-10-19 12:17:53,633 DEV : loss 0.16542117297649384 - f1-score (micro avg) 0.2692
2023-10-19 12:17:53,657 ----------------------------------------------------------------------------------------------------
2023-10-19 12:17:59,754 epoch 6 - iter 260/2606 - loss 0.18992843 - time (sec): 6.10 - samples/sec: 5941.18 - lr: 0.000027 - momentum: 0.000000
2023-10-19 12:18:05,959 epoch 6 - iter 520/2606 - loss 0.19973136 - time (sec): 12.30 - samples/sec: 6061.48 - lr: 0.000027 - momentum: 0.000000
2023-10-19 12:18:11,983 epoch 6 - iter 780/2606 - loss 0.20044914 - time (sec): 18.33 - samples/sec: 6043.26 - lr: 0.000026 - momentum: 0.000000
2023-10-19 12:18:18,208 epoch 6 - iter 1040/2606 - loss 0.19582093 - time (sec): 24.55 - samples/sec: 6172.37 - lr: 0.000026 - momentum: 0.000000
2023-10-19 12:18:24,346 epoch 6 - iter 1300/2606 - loss 0.19555968 - time (sec): 30.69 - samples/sec: 6129.34 - lr: 0.000025 - momentum: 0.000000
2023-10-19 12:18:30,281 epoch 6 - iter 1560/2606 - loss 0.19819069 - time (sec): 36.62 - samples/sec: 6027.05 - lr: 0.000024 - momentum: 0.000000
2023-10-19 12:18:36,257 epoch 6 - iter 1820/2606 - loss 0.19310853 - time (sec): 42.60 - samples/sec: 6027.45 - lr: 0.000024 - momentum: 0.000000
2023-10-19 12:18:42,270 epoch 6 - iter 2080/2606 - loss 0.19523538 - time (sec): 48.61 - samples/sec: 6028.21 - lr: 0.000023 - momentum: 0.000000
2023-10-19 12:18:48,569 epoch 6 - iter 2340/2606 - loss 0.19586601 - time (sec): 54.91 - samples/sec: 6030.74 - lr: 0.000023 - momentum: 0.000000
2023-10-19 12:18:54,595 epoch 6 - iter 2600/2606 - loss 0.19577802 - time (sec): 60.94 - samples/sec: 6012.96 - lr: 0.000022 - momentum: 0.000000
2023-10-19 12:18:54,761 ----------------------------------------------------------------------------------------------------
2023-10-19 12:18:54,762 EPOCH 6 done: loss 0.1958 - lr: 0.000022
2023-10-19 12:19:00,041 DEV : loss 0.1602775603532791 - f1-score (micro avg) 0.287
2023-10-19 12:19:00,064 saving best model
2023-10-19 12:19:00,100 ----------------------------------------------------------------------------------------------------
2023-10-19 12:19:06,390 epoch 7 - iter 260/2606 - loss 0.19220321 - time (sec): 6.29 - samples/sec: 5805.07 - lr: 0.000022 - momentum: 0.000000
2023-10-19 12:19:12,713 epoch 7 - iter 520/2606 - loss 0.17774848 - time (sec): 12.61 - samples/sec: 5762.84 - lr: 0.000021 - momentum: 0.000000
2023-10-19 12:19:18,869 epoch 7 - iter 780/2606 - loss 0.18234572 - time (sec): 18.77 - samples/sec: 5790.14 - lr: 0.000021 - momentum: 0.000000
2023-10-19 12:19:25,087 epoch 7 - iter 1040/2606 - loss 0.18561093 - time (sec): 24.99 - samples/sec: 5882.59 - lr: 0.000020 - momentum: 0.000000
2023-10-19 12:19:31,197 epoch 7 - iter 1300/2606 - loss 0.18840634 - time (sec): 31.10 - samples/sec: 5838.80 - lr: 0.000019 - momentum: 0.000000
2023-10-19 12:19:37,367 epoch 7 - iter 1560/2606 - loss 0.18474661 - time (sec): 37.27 - samples/sec: 5861.62 - lr: 0.000019 - momentum: 0.000000
2023-10-19 12:19:43,563 epoch 7 - iter 1820/2606 - loss 0.18465378 - time (sec): 43.46 - samples/sec: 5858.84 - lr: 0.000018 - momentum: 0.000000
2023-10-19 12:19:49,879 epoch 7 - iter 2080/2606 - loss 0.18296343 - time (sec): 49.78 - samples/sec: 5858.93 - lr: 0.000018 - momentum: 0.000000
2023-10-19 12:19:56,085 epoch 7 - iter 2340/2606 - loss 0.18362259 - time (sec): 55.98 - samples/sec: 5851.91 - lr: 0.000017 - momentum: 0.000000
2023-10-19 12:20:02,342 epoch 7 - iter 2600/2606 - loss 0.18183094 - time (sec): 62.24 - samples/sec: 5886.07 - lr: 0.000017 - momentum: 0.000000
2023-10-19 12:20:02,494 ----------------------------------------------------------------------------------------------------
2023-10-19 12:20:02,494 EPOCH 7 done: loss 0.1820 - lr: 0.000017
2023-10-19 12:20:07,043 DEV : loss 0.1663893163204193 - f1-score (micro avg) 0.2606
2023-10-19 12:20:07,067 ----------------------------------------------------------------------------------------------------
2023-10-19 12:20:13,480 epoch 8 - iter 260/2606 - loss 0.17327680 - time (sec): 6.41 - samples/sec: 5899.14 - lr: 0.000016 - momentum: 0.000000
2023-10-19 12:20:20,259 epoch 8 - iter 520/2606 - loss 0.17288993 - time (sec): 13.19 - samples/sec: 5772.22 - lr: 0.000016 - momentum: 0.000000
2023-10-19 12:20:26,391 epoch 8 - iter 780/2606 - loss 0.17832513 - time (sec): 19.32 - samples/sec: 5847.05 - lr: 0.000015 - momentum: 0.000000
2023-10-19 12:20:32,449 epoch 8 - iter 1040/2606 - loss 0.17895989 - time (sec): 25.38 - samples/sec: 5784.84 - lr: 0.000014 - momentum: 0.000000
2023-10-19 12:20:38,654 epoch 8 - iter 1300/2606 - loss 0.17337141 - time (sec): 31.59 - samples/sec: 5857.95 - lr: 0.000014 - momentum: 0.000000
2023-10-19 12:20:44,733 epoch 8 - iter 1560/2606 - loss 0.17298900 - time (sec): 37.67 - samples/sec: 5825.43 - lr: 0.000013 - momentum: 0.000000
2023-10-19 12:20:50,908 epoch 8 - iter 1820/2606 - loss 0.17399534 - time (sec): 43.84 - samples/sec: 5852.70 - lr: 0.000013 - momentum: 0.000000
2023-10-19 12:20:57,214 epoch 8 - iter 2080/2606 - loss 0.17397486 - time (sec): 50.15 - samples/sec: 5834.64 - lr: 0.000012 - momentum: 0.000000
2023-10-19 12:21:03,425 epoch 8 - iter 2340/2606 - loss 0.17437299 - time (sec): 56.36 - samples/sec: 5844.70 - lr: 0.000012 - momentum: 0.000000
2023-10-19 12:21:09,586 epoch 8 - iter 2600/2606 - loss 0.17324575 - time (sec): 62.52 - samples/sec: 5858.13 - lr: 0.000011 - momentum: 0.000000
2023-10-19 12:21:09,745 ----------------------------------------------------------------------------------------------------
2023-10-19 12:21:09,745 EPOCH 8 done: loss 0.1734 - lr: 0.000011
2023-10-19 12:21:14,256 DEV : loss 0.1701161116361618 - f1-score (micro avg) 0.2849
2023-10-19 12:21:14,281 ----------------------------------------------------------------------------------------------------
2023-10-19 12:21:20,114 epoch 9 - iter 260/2606 - loss 0.16285426 - time (sec): 5.83 - samples/sec: 6017.71 - lr: 0.000011 - momentum: 0.000000
2023-10-19 12:21:26,305 epoch 9 - iter 520/2606 - loss 0.15650374 - time (sec): 12.02 - samples/sec: 5899.13 - lr: 0.000010 - momentum: 0.000000
2023-10-19 12:21:32,497 epoch 9 - iter 780/2606 - loss 0.15289815 - time (sec): 18.22 - samples/sec: 5990.64 - lr: 0.000009 - momentum: 0.000000
2023-10-19 12:21:38,744 epoch 9 - iter 1040/2606 - loss 0.15983102 - time (sec): 24.46 - samples/sec: 5990.22 - lr: 0.000009 - momentum: 0.000000
2023-10-19 12:21:44,897 epoch 9 - iter 1300/2606 - loss 0.15901456 - time (sec): 30.62 - samples/sec: 6016.51 - lr: 0.000008 - momentum: 0.000000
2023-10-19 12:21:51,948 epoch 9 - iter 1560/2606 - loss 0.16225996 - time (sec): 37.67 - samples/sec: 5873.77 - lr: 0.000008 - momentum: 0.000000
2023-10-19 12:21:58,092 epoch 9 - iter 1820/2606 - loss 0.16300920 - time (sec): 43.81 - samples/sec: 5858.05 - lr: 0.000007 - momentum: 0.000000
2023-10-19 12:22:04,260 epoch 9 - iter 2080/2606 - loss 0.16260252 - time (sec): 49.98 - samples/sec: 5885.13 - lr: 0.000007 - momentum: 0.000000
2023-10-19 12:22:10,385 epoch 9 - iter 2340/2606 - loss 0.16370810 - time (sec): 56.10 - samples/sec: 5887.77 - lr: 0.000006 - momentum: 0.000000
2023-10-19 12:22:16,530 epoch 9 - iter 2600/2606 - loss 0.16412782 - time (sec): 62.25 - samples/sec: 5886.33 - lr: 0.000006 - momentum: 0.000000
2023-10-19 12:22:16,678 ----------------------------------------------------------------------------------------------------
2023-10-19 12:22:16,678 EPOCH 9 done: loss 0.1641 - lr: 0.000006
2023-10-19 12:22:21,175 DEV : loss 0.17923708260059357 - f1-score (micro avg) 0.2855
2023-10-19 12:22:21,199 ----------------------------------------------------------------------------------------------------
2023-10-19 12:22:27,204 epoch 10 - iter 260/2606 - loss 0.16832198 - time (sec): 6.00 - samples/sec: 5414.49 - lr: 0.000005 - momentum: 0.000000
2023-10-19 12:22:33,408 epoch 10 - iter 520/2606 - loss 0.15961189 - time (sec): 12.21 - samples/sec: 5784.32 - lr: 0.000004 - momentum: 0.000000
2023-10-19 12:22:39,261 epoch 10 - iter 780/2606 - loss 0.16066261 - time (sec): 18.06 - samples/sec: 5815.64 - lr: 0.000004 - momentum: 0.000000
2023-10-19 12:22:45,464 epoch 10 - iter 1040/2606 - loss 0.16755305 - time (sec): 24.26 - samples/sec: 5869.45 - lr: 0.000003 - momentum: 0.000000
2023-10-19 12:22:51,624 epoch 10 - iter 1300/2606 - loss 0.16288242 - time (sec): 30.43 - samples/sec: 5908.64 - lr: 0.000003 - momentum: 0.000000
2023-10-19 12:22:57,929 epoch 10 - iter 1560/2606 - loss 0.16326348 - time (sec): 36.73 - samples/sec: 5945.25 - lr: 0.000002 - momentum: 0.000000
2023-10-19 12:23:03,980 epoch 10 - iter 1820/2606 - loss 0.16190089 - time (sec): 42.78 - samples/sec: 5896.27 - lr: 0.000002 - momentum: 0.000000
2023-10-19 12:23:09,938 epoch 10 - iter 2080/2606 - loss 0.16132966 - time (sec): 48.74 - samples/sec: 5965.88 - lr: 0.000001 - momentum: 0.000000
2023-10-19 12:23:16,219 epoch 10 - iter 2340/2606 - loss 0.16166399 - time (sec): 55.02 - samples/sec: 5992.67 - lr: 0.000001 - momentum: 0.000000
2023-10-19 12:23:23,237 epoch 10 - iter 2600/2606 - loss 0.16180342 - time (sec): 62.04 - samples/sec: 5914.47 - lr: 0.000000 - momentum: 0.000000
2023-10-19 12:23:23,378 ----------------------------------------------------------------------------------------------------
2023-10-19 12:23:23,378 EPOCH 10 done: loss 0.1617 - lr: 0.000000
2023-10-19 12:23:27,891 DEV : loss 0.18292318284511566 - f1-score (micro avg) 0.2881
2023-10-19 12:23:27,916 saving best model
2023-10-19 12:23:27,974 ----------------------------------------------------------------------------------------------------
2023-10-19 12:23:27,974 Loading model from best epoch ...
2023-10-19 12:23:28,044 SequenceTagger predicts: Dictionary with 17 tags: O, S-LOC, B-LOC, E-LOC, I-LOC, S-PER, B-PER, E-PER, I-PER, S-ORG, B-ORG, E-ORG, I-ORG, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd
2023-10-19 12:23:34,357
Results:
- F-score (micro) 0.321
- F-score (macro) 0.182
- Accuracy 0.1932
By class:
precision recall f1-score support
LOC 0.4269 0.5288 0.4724 1214
PER 0.1741 0.2178 0.1935 808
ORG 0.0652 0.0595 0.0622 353
HumanProd 0.0000 0.0000 0.0000 15
micro avg 0.2957 0.3510 0.3210 2390
macro avg 0.1665 0.2015 0.1820 2390
weighted avg 0.2853 0.3510 0.3146 2390
2023-10-19 12:23:34,357 ----------------------------------------------------------------------------------------------------
|