File size: 24,181 Bytes
19a01cb |
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 |
2023-10-25 18:37:38,397 ----------------------------------------------------------------------------------------------------
2023-10-25 18:37:38,398 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): BertModel(
(embeddings): BertEmbeddings(
(word_embeddings): Embedding(64001, 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()
)"
2023-10-25 18:37:38,398 ----------------------------------------------------------------------------------------------------
2023-10-25 18:37:38,398 MultiCorpus: 7142 train + 698 dev + 2570 test sentences
- NER_HIPE_2022 Corpus: 7142 train + 698 dev + 2570 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/fr/with_doc_seperator
2023-10-25 18:37:38,399 ----------------------------------------------------------------------------------------------------
2023-10-25 18:37:38,399 Train: 7142 sentences
2023-10-25 18:37:38,399 (train_with_dev=False, train_with_test=False)
2023-10-25 18:37:38,399 ----------------------------------------------------------------------------------------------------
2023-10-25 18:37:38,399 Training Params:
2023-10-25 18:37:38,399 - learning_rate: "5e-05"
2023-10-25 18:37:38,399 - mini_batch_size: "4"
2023-10-25 18:37:38,399 - max_epochs: "10"
2023-10-25 18:37:38,399 - shuffle: "True"
2023-10-25 18:37:38,399 ----------------------------------------------------------------------------------------------------
2023-10-25 18:37:38,399 Plugins:
2023-10-25 18:37:38,399 - TensorboardLogger
2023-10-25 18:37:38,399 - LinearScheduler | warmup_fraction: '0.1'
2023-10-25 18:37:38,399 ----------------------------------------------------------------------------------------------------
2023-10-25 18:37:38,399 Final evaluation on model from best epoch (best-model.pt)
2023-10-25 18:37:38,399 - metric: "('micro avg', 'f1-score')"
2023-10-25 18:37:38,399 ----------------------------------------------------------------------------------------------------
2023-10-25 18:37:38,399 Computation:
2023-10-25 18:37:38,399 - compute on device: cuda:0
2023-10-25 18:37:38,399 - embedding storage: none
2023-10-25 18:37:38,399 ----------------------------------------------------------------------------------------------------
2023-10-25 18:37:38,399 Model training base path: "hmbench-newseye/fr-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-5"
2023-10-25 18:37:38,399 ----------------------------------------------------------------------------------------------------
2023-10-25 18:37:38,399 ----------------------------------------------------------------------------------------------------
2023-10-25 18:37:38,399 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-25 18:37:47,542 epoch 1 - iter 178/1786 - loss 1.40453799 - time (sec): 9.14 - samples/sec: 2496.26 - lr: 0.000005 - momentum: 0.000000
2023-10-25 18:37:56,903 epoch 1 - iter 356/1786 - loss 0.90807035 - time (sec): 18.50 - samples/sec: 2525.74 - lr: 0.000010 - momentum: 0.000000
2023-10-25 18:38:06,238 epoch 1 - iter 534/1786 - loss 0.69820556 - time (sec): 27.84 - samples/sec: 2559.06 - lr: 0.000015 - momentum: 0.000000
2023-10-25 18:38:15,506 epoch 1 - iter 712/1786 - loss 0.56410304 - time (sec): 37.11 - samples/sec: 2650.91 - lr: 0.000020 - momentum: 0.000000
2023-10-25 18:38:24,825 epoch 1 - iter 890/1786 - loss 0.48580068 - time (sec): 46.42 - samples/sec: 2644.52 - lr: 0.000025 - momentum: 0.000000
2023-10-25 18:38:34,115 epoch 1 - iter 1068/1786 - loss 0.43280563 - time (sec): 55.71 - samples/sec: 2626.42 - lr: 0.000030 - momentum: 0.000000
2023-10-25 18:38:43,163 epoch 1 - iter 1246/1786 - loss 0.39270958 - time (sec): 64.76 - samples/sec: 2638.36 - lr: 0.000035 - momentum: 0.000000
2023-10-25 18:38:52,184 epoch 1 - iter 1424/1786 - loss 0.36098717 - time (sec): 73.78 - samples/sec: 2661.58 - lr: 0.000040 - momentum: 0.000000
2023-10-25 18:39:01,242 epoch 1 - iter 1602/1786 - loss 0.33608692 - time (sec): 82.84 - samples/sec: 2688.79 - lr: 0.000045 - momentum: 0.000000
2023-10-25 18:39:10,573 epoch 1 - iter 1780/1786 - loss 0.31930380 - time (sec): 92.17 - samples/sec: 2689.56 - lr: 0.000050 - momentum: 0.000000
2023-10-25 18:39:10,863 ----------------------------------------------------------------------------------------------------
2023-10-25 18:39:10,864 EPOCH 1 done: loss 0.3186 - lr: 0.000050
2023-10-25 18:39:15,004 DEV : loss 0.10816145688295364 - f1-score (micro avg) 0.7071
2023-10-25 18:39:15,026 saving best model
2023-10-25 18:39:15,535 ----------------------------------------------------------------------------------------------------
2023-10-25 18:39:25,230 epoch 2 - iter 178/1786 - loss 0.11968307 - time (sec): 9.69 - samples/sec: 2666.97 - lr: 0.000049 - momentum: 0.000000
2023-10-25 18:39:34,400 epoch 2 - iter 356/1786 - loss 0.12356003 - time (sec): 18.86 - samples/sec: 2568.18 - lr: 0.000049 - momentum: 0.000000
2023-10-25 18:39:43,423 epoch 2 - iter 534/1786 - loss 0.12183286 - time (sec): 27.89 - samples/sec: 2629.03 - lr: 0.000048 - momentum: 0.000000
2023-10-25 18:39:52,372 epoch 2 - iter 712/1786 - loss 0.12132289 - time (sec): 36.83 - samples/sec: 2689.32 - lr: 0.000048 - momentum: 0.000000
2023-10-25 18:40:01,037 epoch 2 - iter 890/1786 - loss 0.12094300 - time (sec): 45.50 - samples/sec: 2683.34 - lr: 0.000047 - momentum: 0.000000
2023-10-25 18:40:09,856 epoch 2 - iter 1068/1786 - loss 0.12136206 - time (sec): 54.32 - samples/sec: 2703.63 - lr: 0.000047 - momentum: 0.000000
2023-10-25 18:40:18,988 epoch 2 - iter 1246/1786 - loss 0.12150787 - time (sec): 63.45 - samples/sec: 2703.34 - lr: 0.000046 - momentum: 0.000000
2023-10-25 18:40:28,126 epoch 2 - iter 1424/1786 - loss 0.12125528 - time (sec): 72.59 - samples/sec: 2734.94 - lr: 0.000046 - momentum: 0.000000
2023-10-25 18:40:37,434 epoch 2 - iter 1602/1786 - loss 0.12087874 - time (sec): 81.90 - samples/sec: 2723.53 - lr: 0.000045 - momentum: 0.000000
2023-10-25 18:40:46,385 epoch 2 - iter 1780/1786 - loss 0.12089949 - time (sec): 90.85 - samples/sec: 2730.30 - lr: 0.000044 - momentum: 0.000000
2023-10-25 18:40:46,664 ----------------------------------------------------------------------------------------------------
2023-10-25 18:40:46,665 EPOCH 2 done: loss 0.1209 - lr: 0.000044
2023-10-25 18:40:50,825 DEV : loss 0.11196932196617126 - f1-score (micro avg) 0.7568
2023-10-25 18:40:50,846 saving best model
2023-10-25 18:40:51,544 ----------------------------------------------------------------------------------------------------
2023-10-25 18:41:00,933 epoch 3 - iter 178/1786 - loss 0.07216058 - time (sec): 9.39 - samples/sec: 2668.52 - lr: 0.000044 - momentum: 0.000000
2023-10-25 18:41:11,563 epoch 3 - iter 356/1786 - loss 0.08332331 - time (sec): 20.02 - samples/sec: 2458.48 - lr: 0.000043 - momentum: 0.000000
2023-10-25 18:41:21,065 epoch 3 - iter 534/1786 - loss 0.08767583 - time (sec): 29.52 - samples/sec: 2493.08 - lr: 0.000043 - momentum: 0.000000
2023-10-25 18:41:30,647 epoch 3 - iter 712/1786 - loss 0.08415718 - time (sec): 39.10 - samples/sec: 2524.84 - lr: 0.000042 - momentum: 0.000000
2023-10-25 18:41:40,016 epoch 3 - iter 890/1786 - loss 0.08739794 - time (sec): 48.47 - samples/sec: 2552.95 - lr: 0.000042 - momentum: 0.000000
2023-10-25 18:41:49,509 epoch 3 - iter 1068/1786 - loss 0.08687276 - time (sec): 57.96 - samples/sec: 2572.12 - lr: 0.000041 - momentum: 0.000000
2023-10-25 18:41:59,111 epoch 3 - iter 1246/1786 - loss 0.08597967 - time (sec): 67.57 - samples/sec: 2587.39 - lr: 0.000041 - momentum: 0.000000
2023-10-25 18:42:08,462 epoch 3 - iter 1424/1786 - loss 0.08636941 - time (sec): 76.92 - samples/sec: 2559.26 - lr: 0.000040 - momentum: 0.000000
2023-10-25 18:42:17,442 epoch 3 - iter 1602/1786 - loss 0.08593164 - time (sec): 85.90 - samples/sec: 2587.64 - lr: 0.000039 - momentum: 0.000000
2023-10-25 18:42:26,319 epoch 3 - iter 1780/1786 - loss 0.08507425 - time (sec): 94.77 - samples/sec: 2616.24 - lr: 0.000039 - momentum: 0.000000
2023-10-25 18:42:26,608 ----------------------------------------------------------------------------------------------------
2023-10-25 18:42:26,608 EPOCH 3 done: loss 0.0850 - lr: 0.000039
2023-10-25 18:42:30,919 DEV : loss 0.13416369259357452 - f1-score (micro avg) 0.7734
2023-10-25 18:42:30,941 saving best model
2023-10-25 18:42:31,602 ----------------------------------------------------------------------------------------------------
2023-10-25 18:42:40,853 epoch 4 - iter 178/1786 - loss 0.05392007 - time (sec): 9.25 - samples/sec: 2660.38 - lr: 0.000038 - momentum: 0.000000
2023-10-25 18:42:50,351 epoch 4 - iter 356/1786 - loss 0.05706668 - time (sec): 18.75 - samples/sec: 2686.77 - lr: 0.000038 - momentum: 0.000000
2023-10-25 18:42:59,263 epoch 4 - iter 534/1786 - loss 0.06080001 - time (sec): 27.66 - samples/sec: 2693.61 - lr: 0.000037 - momentum: 0.000000
2023-10-25 18:43:08,040 epoch 4 - iter 712/1786 - loss 0.06411629 - time (sec): 36.44 - samples/sec: 2714.05 - lr: 0.000037 - momentum: 0.000000
2023-10-25 18:43:17,331 epoch 4 - iter 890/1786 - loss 0.06293852 - time (sec): 45.73 - samples/sec: 2693.11 - lr: 0.000036 - momentum: 0.000000
2023-10-25 18:43:26,753 epoch 4 - iter 1068/1786 - loss 0.06522331 - time (sec): 55.15 - samples/sec: 2696.19 - lr: 0.000036 - momentum: 0.000000
2023-10-25 18:43:35,821 epoch 4 - iter 1246/1786 - loss 0.06635384 - time (sec): 64.22 - samples/sec: 2691.82 - lr: 0.000035 - momentum: 0.000000
2023-10-25 18:43:44,420 epoch 4 - iter 1424/1786 - loss 0.06654842 - time (sec): 72.82 - samples/sec: 2724.98 - lr: 0.000034 - momentum: 0.000000
2023-10-25 18:43:53,437 epoch 4 - iter 1602/1786 - loss 0.06525833 - time (sec): 81.83 - samples/sec: 2730.35 - lr: 0.000034 - momentum: 0.000000
2023-10-25 18:44:02,280 epoch 4 - iter 1780/1786 - loss 0.06417141 - time (sec): 90.68 - samples/sec: 2735.96 - lr: 0.000033 - momentum: 0.000000
2023-10-25 18:44:02,561 ----------------------------------------------------------------------------------------------------
2023-10-25 18:44:02,561 EPOCH 4 done: loss 0.0641 - lr: 0.000033
2023-10-25 18:44:07,999 DEV : loss 0.1622055321931839 - f1-score (micro avg) 0.7949
2023-10-25 18:44:08,021 saving best model
2023-10-25 18:44:08,751 ----------------------------------------------------------------------------------------------------
2023-10-25 18:44:18,039 epoch 5 - iter 178/1786 - loss 0.05453094 - time (sec): 9.29 - samples/sec: 2544.86 - lr: 0.000033 - momentum: 0.000000
2023-10-25 18:44:27,561 epoch 5 - iter 356/1786 - loss 0.04577426 - time (sec): 18.81 - samples/sec: 2676.28 - lr: 0.000032 - momentum: 0.000000
2023-10-25 18:44:37,038 epoch 5 - iter 534/1786 - loss 0.04807378 - time (sec): 28.28 - samples/sec: 2655.55 - lr: 0.000032 - momentum: 0.000000
2023-10-25 18:44:46,571 epoch 5 - iter 712/1786 - loss 0.04717079 - time (sec): 37.82 - samples/sec: 2653.43 - lr: 0.000031 - momentum: 0.000000
2023-10-25 18:44:56,134 epoch 5 - iter 890/1786 - loss 0.04803019 - time (sec): 47.38 - samples/sec: 2655.16 - lr: 0.000031 - momentum: 0.000000
2023-10-25 18:45:05,721 epoch 5 - iter 1068/1786 - loss 0.04817761 - time (sec): 56.97 - samples/sec: 2618.31 - lr: 0.000030 - momentum: 0.000000
2023-10-25 18:45:15,218 epoch 5 - iter 1246/1786 - loss 0.04776400 - time (sec): 66.46 - samples/sec: 2624.83 - lr: 0.000029 - momentum: 0.000000
2023-10-25 18:45:24,767 epoch 5 - iter 1424/1786 - loss 0.04756135 - time (sec): 76.01 - samples/sec: 2611.00 - lr: 0.000029 - momentum: 0.000000
2023-10-25 18:45:34,208 epoch 5 - iter 1602/1786 - loss 0.04690501 - time (sec): 85.46 - samples/sec: 2596.86 - lr: 0.000028 - momentum: 0.000000
2023-10-25 18:45:43,226 epoch 5 - iter 1780/1786 - loss 0.04701536 - time (sec): 94.47 - samples/sec: 2622.82 - lr: 0.000028 - momentum: 0.000000
2023-10-25 18:45:43,541 ----------------------------------------------------------------------------------------------------
2023-10-25 18:45:43,541 EPOCH 5 done: loss 0.0471 - lr: 0.000028
2023-10-25 18:45:47,551 DEV : loss 0.17839400470256805 - f1-score (micro avg) 0.7641
2023-10-25 18:45:47,571 ----------------------------------------------------------------------------------------------------
2023-10-25 18:45:57,008 epoch 6 - iter 178/1786 - loss 0.04739676 - time (sec): 9.44 - samples/sec: 2669.74 - lr: 0.000027 - momentum: 0.000000
2023-10-25 18:46:06,535 epoch 6 - iter 356/1786 - loss 0.03785218 - time (sec): 18.96 - samples/sec: 2692.43 - lr: 0.000027 - momentum: 0.000000
2023-10-25 18:46:15,980 epoch 6 - iter 534/1786 - loss 0.03634755 - time (sec): 28.41 - samples/sec: 2618.52 - lr: 0.000026 - momentum: 0.000000
2023-10-25 18:46:25,654 epoch 6 - iter 712/1786 - loss 0.04044514 - time (sec): 38.08 - samples/sec: 2607.08 - lr: 0.000026 - momentum: 0.000000
2023-10-25 18:46:35,381 epoch 6 - iter 890/1786 - loss 0.03838047 - time (sec): 47.81 - samples/sec: 2615.00 - lr: 0.000025 - momentum: 0.000000
2023-10-25 18:46:44,873 epoch 6 - iter 1068/1786 - loss 0.03827037 - time (sec): 57.30 - samples/sec: 2616.02 - lr: 0.000024 - momentum: 0.000000
2023-10-25 18:46:54,684 epoch 6 - iter 1246/1786 - loss 0.03744695 - time (sec): 67.11 - samples/sec: 2607.36 - lr: 0.000024 - momentum: 0.000000
2023-10-25 18:47:03,908 epoch 6 - iter 1424/1786 - loss 0.03713239 - time (sec): 76.34 - samples/sec: 2598.15 - lr: 0.000023 - momentum: 0.000000
2023-10-25 18:47:13,637 epoch 6 - iter 1602/1786 - loss 0.03788438 - time (sec): 86.06 - samples/sec: 2617.86 - lr: 0.000023 - momentum: 0.000000
2023-10-25 18:47:22,668 epoch 6 - iter 1780/1786 - loss 0.03775238 - time (sec): 95.10 - samples/sec: 2609.10 - lr: 0.000022 - momentum: 0.000000
2023-10-25 18:47:22,974 ----------------------------------------------------------------------------------------------------
2023-10-25 18:47:22,975 EPOCH 6 done: loss 0.0377 - lr: 0.000022
2023-10-25 18:47:27,370 DEV : loss 0.17886780202388763 - f1-score (micro avg) 0.7886
2023-10-25 18:47:27,392 ----------------------------------------------------------------------------------------------------
2023-10-25 18:47:37,024 epoch 7 - iter 178/1786 - loss 0.03077925 - time (sec): 9.63 - samples/sec: 2692.33 - lr: 0.000022 - momentum: 0.000000
2023-10-25 18:47:46,517 epoch 7 - iter 356/1786 - loss 0.02932877 - time (sec): 19.12 - samples/sec: 2574.46 - lr: 0.000021 - momentum: 0.000000
2023-10-25 18:47:55,977 epoch 7 - iter 534/1786 - loss 0.02829002 - time (sec): 28.58 - samples/sec: 2597.22 - lr: 0.000021 - momentum: 0.000000
2023-10-25 18:48:05,423 epoch 7 - iter 712/1786 - loss 0.02973574 - time (sec): 38.03 - samples/sec: 2597.81 - lr: 0.000020 - momentum: 0.000000
2023-10-25 18:48:14,909 epoch 7 - iter 890/1786 - loss 0.02980090 - time (sec): 47.52 - samples/sec: 2639.78 - lr: 0.000019 - momentum: 0.000000
2023-10-25 18:48:24,135 epoch 7 - iter 1068/1786 - loss 0.03007659 - time (sec): 56.74 - samples/sec: 2657.82 - lr: 0.000019 - momentum: 0.000000
2023-10-25 18:48:33,402 epoch 7 - iter 1246/1786 - loss 0.02862806 - time (sec): 66.01 - samples/sec: 2678.20 - lr: 0.000018 - momentum: 0.000000
2023-10-25 18:48:42,731 epoch 7 - iter 1424/1786 - loss 0.02826589 - time (sec): 75.34 - samples/sec: 2639.07 - lr: 0.000018 - momentum: 0.000000
2023-10-25 18:48:51,933 epoch 7 - iter 1602/1786 - loss 0.02797061 - time (sec): 84.54 - samples/sec: 2641.63 - lr: 0.000017 - momentum: 0.000000
2023-10-25 18:49:01,510 epoch 7 - iter 1780/1786 - loss 0.02819106 - time (sec): 94.12 - samples/sec: 2632.82 - lr: 0.000017 - momentum: 0.000000
2023-10-25 18:49:01,831 ----------------------------------------------------------------------------------------------------
2023-10-25 18:49:01,831 EPOCH 7 done: loss 0.0281 - lr: 0.000017
2023-10-25 18:49:06,709 DEV : loss 0.19223909080028534 - f1-score (micro avg) 0.7857
2023-10-25 18:49:06,731 ----------------------------------------------------------------------------------------------------
2023-10-25 18:49:16,260 epoch 8 - iter 178/1786 - loss 0.02174503 - time (sec): 9.53 - samples/sec: 2599.42 - lr: 0.000016 - momentum: 0.000000
2023-10-25 18:49:25,630 epoch 8 - iter 356/1786 - loss 0.01913432 - time (sec): 18.90 - samples/sec: 2564.11 - lr: 0.000016 - momentum: 0.000000
2023-10-25 18:49:35,140 epoch 8 - iter 534/1786 - loss 0.01746195 - time (sec): 28.41 - samples/sec: 2589.15 - lr: 0.000015 - momentum: 0.000000
2023-10-25 18:49:44,521 epoch 8 - iter 712/1786 - loss 0.01693117 - time (sec): 37.79 - samples/sec: 2623.00 - lr: 0.000014 - momentum: 0.000000
2023-10-25 18:49:53,894 epoch 8 - iter 890/1786 - loss 0.01671270 - time (sec): 47.16 - samples/sec: 2621.55 - lr: 0.000014 - momentum: 0.000000
2023-10-25 18:50:03,220 epoch 8 - iter 1068/1786 - loss 0.01731034 - time (sec): 56.49 - samples/sec: 2602.71 - lr: 0.000013 - momentum: 0.000000
2023-10-25 18:50:12,498 epoch 8 - iter 1246/1786 - loss 0.01886582 - time (sec): 65.77 - samples/sec: 2608.65 - lr: 0.000013 - momentum: 0.000000
2023-10-25 18:50:21,455 epoch 8 - iter 1424/1786 - loss 0.01893327 - time (sec): 74.72 - samples/sec: 2631.11 - lr: 0.000012 - momentum: 0.000000
2023-10-25 18:50:30,440 epoch 8 - iter 1602/1786 - loss 0.01901785 - time (sec): 83.71 - samples/sec: 2666.98 - lr: 0.000012 - momentum: 0.000000
2023-10-25 18:50:39,483 epoch 8 - iter 1780/1786 - loss 0.01903471 - time (sec): 92.75 - samples/sec: 2673.79 - lr: 0.000011 - momentum: 0.000000
2023-10-25 18:50:39,785 ----------------------------------------------------------------------------------------------------
2023-10-25 18:50:39,785 EPOCH 8 done: loss 0.0190 - lr: 0.000011
2023-10-25 18:50:43,730 DEV : loss 0.22295665740966797 - f1-score (micro avg) 0.7927
2023-10-25 18:50:43,754 ----------------------------------------------------------------------------------------------------
2023-10-25 18:50:53,379 epoch 9 - iter 178/1786 - loss 0.00894777 - time (sec): 9.62 - samples/sec: 2649.14 - lr: 0.000011 - momentum: 0.000000
2023-10-25 18:51:02,883 epoch 9 - iter 356/1786 - loss 0.01120950 - time (sec): 19.13 - samples/sec: 2527.08 - lr: 0.000010 - momentum: 0.000000
2023-10-25 18:51:12,372 epoch 9 - iter 534/1786 - loss 0.01367811 - time (sec): 28.62 - samples/sec: 2533.14 - lr: 0.000009 - momentum: 0.000000
2023-10-25 18:51:21,536 epoch 9 - iter 712/1786 - loss 0.01359022 - time (sec): 37.78 - samples/sec: 2567.40 - lr: 0.000009 - momentum: 0.000000
2023-10-25 18:51:30,710 epoch 9 - iter 890/1786 - loss 0.01350802 - time (sec): 46.95 - samples/sec: 2560.14 - lr: 0.000008 - momentum: 0.000000
2023-10-25 18:51:39,721 epoch 9 - iter 1068/1786 - loss 0.01419765 - time (sec): 55.97 - samples/sec: 2623.59 - lr: 0.000008 - momentum: 0.000000
2023-10-25 18:51:48,814 epoch 9 - iter 1246/1786 - loss 0.01486049 - time (sec): 65.06 - samples/sec: 2629.79 - lr: 0.000007 - momentum: 0.000000
2023-10-25 18:51:57,499 epoch 9 - iter 1424/1786 - loss 0.01410719 - time (sec): 73.74 - samples/sec: 2666.03 - lr: 0.000007 - momentum: 0.000000
2023-10-25 18:52:06,116 epoch 9 - iter 1602/1786 - loss 0.01367858 - time (sec): 82.36 - samples/sec: 2689.33 - lr: 0.000006 - momentum: 0.000000
2023-10-25 18:52:15,069 epoch 9 - iter 1780/1786 - loss 0.01309891 - time (sec): 91.31 - samples/sec: 2717.19 - lr: 0.000006 - momentum: 0.000000
2023-10-25 18:52:15,368 ----------------------------------------------------------------------------------------------------
2023-10-25 18:52:15,369 EPOCH 9 done: loss 0.0132 - lr: 0.000006
2023-10-25 18:52:20,439 DEV : loss 0.22367699444293976 - f1-score (micro avg) 0.7949
2023-10-25 18:52:20,459 ----------------------------------------------------------------------------------------------------
2023-10-25 18:52:29,298 epoch 10 - iter 178/1786 - loss 0.00880098 - time (sec): 8.84 - samples/sec: 2842.17 - lr: 0.000005 - momentum: 0.000000
2023-10-25 18:52:37,837 epoch 10 - iter 356/1786 - loss 0.00764413 - time (sec): 17.38 - samples/sec: 2803.77 - lr: 0.000004 - momentum: 0.000000
2023-10-25 18:52:46,910 epoch 10 - iter 534/1786 - loss 0.00803735 - time (sec): 26.45 - samples/sec: 2802.88 - lr: 0.000004 - momentum: 0.000000
2023-10-25 18:52:56,296 epoch 10 - iter 712/1786 - loss 0.00807527 - time (sec): 35.84 - samples/sec: 2775.80 - lr: 0.000003 - momentum: 0.000000
2023-10-25 18:53:05,399 epoch 10 - iter 890/1786 - loss 0.00741347 - time (sec): 44.94 - samples/sec: 2772.21 - lr: 0.000003 - momentum: 0.000000
2023-10-25 18:53:14,850 epoch 10 - iter 1068/1786 - loss 0.00783141 - time (sec): 54.39 - samples/sec: 2736.83 - lr: 0.000002 - momentum: 0.000000
2023-10-25 18:53:24,330 epoch 10 - iter 1246/1786 - loss 0.00870824 - time (sec): 63.87 - samples/sec: 2711.41 - lr: 0.000002 - momentum: 0.000000
2023-10-25 18:53:33,732 epoch 10 - iter 1424/1786 - loss 0.00816998 - time (sec): 73.27 - samples/sec: 2716.97 - lr: 0.000001 - momentum: 0.000000
2023-10-25 18:53:43,352 epoch 10 - iter 1602/1786 - loss 0.00820551 - time (sec): 82.89 - samples/sec: 2698.61 - lr: 0.000001 - momentum: 0.000000
2023-10-25 18:53:52,998 epoch 10 - iter 1780/1786 - loss 0.00861252 - time (sec): 92.54 - samples/sec: 2679.83 - lr: 0.000000 - momentum: 0.000000
2023-10-25 18:53:53,317 ----------------------------------------------------------------------------------------------------
2023-10-25 18:53:53,317 EPOCH 10 done: loss 0.0086 - lr: 0.000000
2023-10-25 18:53:58,167 DEV : loss 0.22940541803836823 - f1-score (micro avg) 0.792
2023-10-25 18:53:58,699 ----------------------------------------------------------------------------------------------------
2023-10-25 18:53:58,701 Loading model from best epoch ...
2023-10-25 18:54:00,626 SequenceTagger predicts: Dictionary with 17 tags: O, S-PER, B-PER, E-PER, I-PER, S-LOC, B-LOC, E-LOC, I-LOC, S-ORG, B-ORG, E-ORG, I-ORG, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd
2023-10-25 18:54:13,345
Results:
- F-score (micro) 0.6526
- F-score (macro) 0.5814
- Accuracy 0.5009
By class:
precision recall f1-score support
LOC 0.6415 0.6521 0.6467 1095
PER 0.7798 0.7312 0.7547 1012
ORG 0.3913 0.5294 0.4500 357
HumanProd 0.3594 0.6970 0.4742 33
micro avg 0.6386 0.6672 0.6526 2497
macro avg 0.5430 0.6524 0.5814 2497
weighted avg 0.6580 0.6672 0.6601 2497
2023-10-25 18:54:13,345 ----------------------------------------------------------------------------------------------------
|