File size: 25,123 Bytes
3bcde47 |
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 |
2023-10-06 11:40:49,468 ----------------------------------------------------------------------------------------------------
2023-10-06 11:40:49,469 Model: "SequenceTagger(
(embeddings): ByT5Embeddings(
(model): T5EncoderModel(
(shared): Embedding(384, 1472)
(encoder): T5Stack(
(embed_tokens): Embedding(384, 1472)
(block): ModuleList(
(0): T5Block(
(layer): ModuleList(
(0): T5LayerSelfAttention(
(SelfAttention): T5Attention(
(q): Linear(in_features=1472, out_features=384, bias=False)
(k): Linear(in_features=1472, out_features=384, bias=False)
(v): Linear(in_features=1472, out_features=384, bias=False)
(o): Linear(in_features=384, out_features=1472, bias=False)
(relative_attention_bias): Embedding(32, 6)
)
(layer_norm): T5LayerNorm()
(dropout): Dropout(p=0.1, inplace=False)
)
(1): T5LayerFF(
(DenseReluDense): T5DenseGatedActDense(
(wi_0): Linear(in_features=1472, out_features=3584, bias=False)
(wi_1): Linear(in_features=1472, out_features=3584, bias=False)
(wo): Linear(in_features=3584, out_features=1472, bias=False)
(dropout): Dropout(p=0.1, inplace=False)
(act): NewGELUActivation()
)
(layer_norm): T5LayerNorm()
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
(1-11): 11 x T5Block(
(layer): ModuleList(
(0): T5LayerSelfAttention(
(SelfAttention): T5Attention(
(q): Linear(in_features=1472, out_features=384, bias=False)
(k): Linear(in_features=1472, out_features=384, bias=False)
(v): Linear(in_features=1472, out_features=384, bias=False)
(o): Linear(in_features=384, out_features=1472, bias=False)
)
(layer_norm): T5LayerNorm()
(dropout): Dropout(p=0.1, inplace=False)
)
(1): T5LayerFF(
(DenseReluDense): T5DenseGatedActDense(
(wi_0): Linear(in_features=1472, out_features=3584, bias=False)
(wi_1): Linear(in_features=1472, out_features=3584, bias=False)
(wo): Linear(in_features=3584, out_features=1472, bias=False)
(dropout): Dropout(p=0.1, inplace=False)
(act): NewGELUActivation()
)
(layer_norm): T5LayerNorm()
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
(final_layer_norm): T5LayerNorm()
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=1472, out_features=25, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-06 11:40:49,469 ----------------------------------------------------------------------------------------------------
2023-10-06 11:40:49,470 MultiCorpus: 1214 train + 266 dev + 251 test sentences
- NER_HIPE_2022 Corpus: 1214 train + 266 dev + 251 test sentences - /app/.flair/datasets/ner_hipe_2022/v2.1/ajmc/en/with_doc_seperator
2023-10-06 11:40:49,470 ----------------------------------------------------------------------------------------------------
2023-10-06 11:40:49,470 Train: 1214 sentences
2023-10-06 11:40:49,470 (train_with_dev=False, train_with_test=False)
2023-10-06 11:40:49,470 ----------------------------------------------------------------------------------------------------
2023-10-06 11:40:49,470 Training Params:
2023-10-06 11:40:49,470 - learning_rate: "0.00015"
2023-10-06 11:40:49,470 - mini_batch_size: "8"
2023-10-06 11:40:49,470 - max_epochs: "10"
2023-10-06 11:40:49,470 - shuffle: "True"
2023-10-06 11:40:49,470 ----------------------------------------------------------------------------------------------------
2023-10-06 11:40:49,470 Plugins:
2023-10-06 11:40:49,470 - TensorboardLogger
2023-10-06 11:40:49,470 - LinearScheduler | warmup_fraction: '0.1'
2023-10-06 11:40:49,471 ----------------------------------------------------------------------------------------------------
2023-10-06 11:40:49,471 Final evaluation on model from best epoch (best-model.pt)
2023-10-06 11:40:49,471 - metric: "('micro avg', 'f1-score')"
2023-10-06 11:40:49,471 ----------------------------------------------------------------------------------------------------
2023-10-06 11:40:49,471 Computation:
2023-10-06 11:40:49,471 - compute on device: cuda:0
2023-10-06 11:40:49,471 - embedding storage: none
2023-10-06 11:40:49,471 ----------------------------------------------------------------------------------------------------
2023-10-06 11:40:49,471 Model training base path: "hmbench-ajmc/en-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs8-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-2"
2023-10-06 11:40:49,471 ----------------------------------------------------------------------------------------------------
2023-10-06 11:40:49,471 ----------------------------------------------------------------------------------------------------
2023-10-06 11:40:49,471 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-06 11:41:00,549 epoch 1 - iter 15/152 - loss 3.23650866 - time (sec): 11.08 - samples/sec: 279.23 - lr: 0.000014 - momentum: 0.000000
2023-10-06 11:41:11,564 epoch 1 - iter 30/152 - loss 3.23109334 - time (sec): 22.09 - samples/sec: 279.07 - lr: 0.000029 - momentum: 0.000000
2023-10-06 11:41:22,776 epoch 1 - iter 45/152 - loss 3.21953383 - time (sec): 33.30 - samples/sec: 281.32 - lr: 0.000043 - momentum: 0.000000
2023-10-06 11:41:34,625 epoch 1 - iter 60/152 - loss 3.20155482 - time (sec): 45.15 - samples/sec: 282.16 - lr: 0.000058 - momentum: 0.000000
2023-10-06 11:41:45,889 epoch 1 - iter 75/152 - loss 3.17002476 - time (sec): 56.42 - samples/sec: 280.68 - lr: 0.000073 - momentum: 0.000000
2023-10-06 11:41:57,404 epoch 1 - iter 90/152 - loss 3.10578049 - time (sec): 67.93 - samples/sec: 279.94 - lr: 0.000088 - momentum: 0.000000
2023-10-06 11:42:08,171 epoch 1 - iter 105/152 - loss 3.03046549 - time (sec): 78.70 - samples/sec: 277.27 - lr: 0.000103 - momentum: 0.000000
2023-10-06 11:42:19,288 epoch 1 - iter 120/152 - loss 2.94097459 - time (sec): 89.82 - samples/sec: 276.65 - lr: 0.000117 - momentum: 0.000000
2023-10-06 11:42:29,686 epoch 1 - iter 135/152 - loss 2.85356329 - time (sec): 100.21 - samples/sec: 275.04 - lr: 0.000132 - momentum: 0.000000
2023-10-06 11:42:40,876 epoch 1 - iter 150/152 - loss 2.74658756 - time (sec): 111.40 - samples/sec: 275.49 - lr: 0.000147 - momentum: 0.000000
2023-10-06 11:42:42,072 ----------------------------------------------------------------------------------------------------
2023-10-06 11:42:42,072 EPOCH 1 done: loss 2.7375 - lr: 0.000147
2023-10-06 11:42:49,696 DEV : loss 1.626281976699829 - f1-score (micro avg) 0.0
2023-10-06 11:42:49,703 ----------------------------------------------------------------------------------------------------
2023-10-06 11:43:00,883 epoch 2 - iter 15/152 - loss 1.58388197 - time (sec): 11.18 - samples/sec: 275.18 - lr: 0.000148 - momentum: 0.000000
2023-10-06 11:43:11,945 epoch 2 - iter 30/152 - loss 1.42058436 - time (sec): 22.24 - samples/sec: 272.07 - lr: 0.000147 - momentum: 0.000000
2023-10-06 11:43:23,461 epoch 2 - iter 45/152 - loss 1.32006778 - time (sec): 33.76 - samples/sec: 272.69 - lr: 0.000145 - momentum: 0.000000
2023-10-06 11:43:34,246 epoch 2 - iter 60/152 - loss 1.22909384 - time (sec): 44.54 - samples/sec: 274.46 - lr: 0.000144 - momentum: 0.000000
2023-10-06 11:43:44,957 epoch 2 - iter 75/152 - loss 1.15585388 - time (sec): 55.25 - samples/sec: 272.98 - lr: 0.000142 - momentum: 0.000000
2023-10-06 11:43:56,489 epoch 2 - iter 90/152 - loss 1.08103450 - time (sec): 66.78 - samples/sec: 274.73 - lr: 0.000140 - momentum: 0.000000
2023-10-06 11:44:07,713 epoch 2 - iter 105/152 - loss 1.01665926 - time (sec): 78.01 - samples/sec: 274.71 - lr: 0.000139 - momentum: 0.000000
2023-10-06 11:44:18,325 epoch 2 - iter 120/152 - loss 0.97879470 - time (sec): 88.62 - samples/sec: 273.78 - lr: 0.000137 - momentum: 0.000000
2023-10-06 11:44:29,862 epoch 2 - iter 135/152 - loss 0.94440775 - time (sec): 100.16 - samples/sec: 274.31 - lr: 0.000135 - momentum: 0.000000
2023-10-06 11:44:40,800 epoch 2 - iter 150/152 - loss 0.91769126 - time (sec): 111.10 - samples/sec: 275.17 - lr: 0.000134 - momentum: 0.000000
2023-10-06 11:44:42,166 ----------------------------------------------------------------------------------------------------
2023-10-06 11:44:42,166 EPOCH 2 done: loss 0.9140 - lr: 0.000134
2023-10-06 11:44:50,048 DEV : loss 0.6008118987083435 - f1-score (micro avg) 0.0
2023-10-06 11:44:50,056 ----------------------------------------------------------------------------------------------------
2023-10-06 11:45:01,236 epoch 3 - iter 15/152 - loss 0.54124878 - time (sec): 11.18 - samples/sec: 276.16 - lr: 0.000132 - momentum: 0.000000
2023-10-06 11:45:11,760 epoch 3 - iter 30/152 - loss 0.48867766 - time (sec): 21.70 - samples/sec: 275.18 - lr: 0.000130 - momentum: 0.000000
2023-10-06 11:45:22,405 epoch 3 - iter 45/152 - loss 0.47061111 - time (sec): 32.35 - samples/sec: 274.80 - lr: 0.000129 - momentum: 0.000000
2023-10-06 11:45:34,805 epoch 3 - iter 60/152 - loss 0.46014139 - time (sec): 44.75 - samples/sec: 279.32 - lr: 0.000127 - momentum: 0.000000
2023-10-06 11:45:46,070 epoch 3 - iter 75/152 - loss 0.45667027 - time (sec): 56.01 - samples/sec: 278.22 - lr: 0.000125 - momentum: 0.000000
2023-10-06 11:45:57,360 epoch 3 - iter 90/152 - loss 0.42982243 - time (sec): 67.30 - samples/sec: 277.73 - lr: 0.000124 - momentum: 0.000000
2023-10-06 11:46:08,146 epoch 3 - iter 105/152 - loss 0.41230990 - time (sec): 78.09 - samples/sec: 276.48 - lr: 0.000122 - momentum: 0.000000
2023-10-06 11:46:19,262 epoch 3 - iter 120/152 - loss 0.40429627 - time (sec): 89.20 - samples/sec: 277.46 - lr: 0.000120 - momentum: 0.000000
2023-10-06 11:46:30,403 epoch 3 - iter 135/152 - loss 0.39232877 - time (sec): 100.35 - samples/sec: 276.66 - lr: 0.000119 - momentum: 0.000000
2023-10-06 11:46:41,323 epoch 3 - iter 150/152 - loss 0.38949799 - time (sec): 111.26 - samples/sec: 275.33 - lr: 0.000117 - momentum: 0.000000
2023-10-06 11:46:42,635 ----------------------------------------------------------------------------------------------------
2023-10-06 11:46:42,636 EPOCH 3 done: loss 0.3892 - lr: 0.000117
2023-10-06 11:46:50,526 DEV : loss 0.339983731508255 - f1-score (micro avg) 0.5211
2023-10-06 11:46:50,533 saving best model
2023-10-06 11:46:51,379 ----------------------------------------------------------------------------------------------------
2023-10-06 11:47:01,879 epoch 4 - iter 15/152 - loss 0.27249716 - time (sec): 10.50 - samples/sec: 268.42 - lr: 0.000115 - momentum: 0.000000
2023-10-06 11:47:13,041 epoch 4 - iter 30/152 - loss 0.27502400 - time (sec): 21.66 - samples/sec: 264.12 - lr: 0.000114 - momentum: 0.000000
2023-10-06 11:47:24,522 epoch 4 - iter 45/152 - loss 0.26296682 - time (sec): 33.14 - samples/sec: 269.21 - lr: 0.000112 - momentum: 0.000000
2023-10-06 11:47:35,706 epoch 4 - iter 60/152 - loss 0.26165441 - time (sec): 44.33 - samples/sec: 269.59 - lr: 0.000110 - momentum: 0.000000
2023-10-06 11:47:47,542 epoch 4 - iter 75/152 - loss 0.25680579 - time (sec): 56.16 - samples/sec: 273.69 - lr: 0.000109 - momentum: 0.000000
2023-10-06 11:47:59,396 epoch 4 - iter 90/152 - loss 0.25394861 - time (sec): 68.02 - samples/sec: 275.88 - lr: 0.000107 - momentum: 0.000000
2023-10-06 11:48:10,598 epoch 4 - iter 105/152 - loss 0.24979844 - time (sec): 79.22 - samples/sec: 275.77 - lr: 0.000105 - momentum: 0.000000
2023-10-06 11:48:21,274 epoch 4 - iter 120/152 - loss 0.24458649 - time (sec): 89.89 - samples/sec: 275.11 - lr: 0.000104 - momentum: 0.000000
2023-10-06 11:48:32,019 epoch 4 - iter 135/152 - loss 0.24155398 - time (sec): 100.64 - samples/sec: 274.59 - lr: 0.000102 - momentum: 0.000000
2023-10-06 11:48:42,875 epoch 4 - iter 150/152 - loss 0.23636777 - time (sec): 111.50 - samples/sec: 273.90 - lr: 0.000101 - momentum: 0.000000
2023-10-06 11:48:44,397 ----------------------------------------------------------------------------------------------------
2023-10-06 11:48:44,398 EPOCH 4 done: loss 0.2378 - lr: 0.000101
2023-10-06 11:48:52,277 DEV : loss 0.24833999574184418 - f1-score (micro avg) 0.6874
2023-10-06 11:48:52,284 saving best model
2023-10-06 11:48:56,646 ----------------------------------------------------------------------------------------------------
2023-10-06 11:49:07,702 epoch 5 - iter 15/152 - loss 0.18384846 - time (sec): 11.05 - samples/sec: 269.57 - lr: 0.000099 - momentum: 0.000000
2023-10-06 11:49:18,526 epoch 5 - iter 30/152 - loss 0.17133826 - time (sec): 21.88 - samples/sec: 266.38 - lr: 0.000097 - momentum: 0.000000
2023-10-06 11:49:30,508 epoch 5 - iter 45/152 - loss 0.16088395 - time (sec): 33.86 - samples/sec: 274.24 - lr: 0.000095 - momentum: 0.000000
2023-10-06 11:49:41,771 epoch 5 - iter 60/152 - loss 0.16433942 - time (sec): 45.12 - samples/sec: 275.64 - lr: 0.000094 - momentum: 0.000000
2023-10-06 11:49:53,169 epoch 5 - iter 75/152 - loss 0.16163685 - time (sec): 56.52 - samples/sec: 276.30 - lr: 0.000092 - momentum: 0.000000
2023-10-06 11:50:04,680 epoch 5 - iter 90/152 - loss 0.16120892 - time (sec): 68.03 - samples/sec: 275.65 - lr: 0.000091 - momentum: 0.000000
2023-10-06 11:50:14,958 epoch 5 - iter 105/152 - loss 0.15841038 - time (sec): 78.31 - samples/sec: 272.54 - lr: 0.000089 - momentum: 0.000000
2023-10-06 11:50:25,602 epoch 5 - iter 120/152 - loss 0.15944779 - time (sec): 88.95 - samples/sec: 271.77 - lr: 0.000087 - momentum: 0.000000
2023-10-06 11:50:36,931 epoch 5 - iter 135/152 - loss 0.15939219 - time (sec): 100.28 - samples/sec: 272.97 - lr: 0.000086 - momentum: 0.000000
2023-10-06 11:50:48,188 epoch 5 - iter 150/152 - loss 0.16161550 - time (sec): 111.54 - samples/sec: 274.44 - lr: 0.000084 - momentum: 0.000000
2023-10-06 11:50:49,578 ----------------------------------------------------------------------------------------------------
2023-10-06 11:50:49,579 EPOCH 5 done: loss 0.1608 - lr: 0.000084
2023-10-06 11:50:57,499 DEV : loss 0.18905217945575714 - f1-score (micro avg) 0.7007
2023-10-06 11:50:57,506 saving best model
2023-10-06 11:51:01,895 ----------------------------------------------------------------------------------------------------
2023-10-06 11:51:12,689 epoch 6 - iter 15/152 - loss 0.14202084 - time (sec): 10.79 - samples/sec: 273.16 - lr: 0.000082 - momentum: 0.000000
2023-10-06 11:51:23,857 epoch 6 - iter 30/152 - loss 0.13186308 - time (sec): 21.96 - samples/sec: 272.95 - lr: 0.000080 - momentum: 0.000000
2023-10-06 11:51:34,639 epoch 6 - iter 45/152 - loss 0.12753401 - time (sec): 32.74 - samples/sec: 272.95 - lr: 0.000079 - momentum: 0.000000
2023-10-06 11:51:45,662 epoch 6 - iter 60/152 - loss 0.13110037 - time (sec): 43.77 - samples/sec: 272.73 - lr: 0.000077 - momentum: 0.000000
2023-10-06 11:51:56,695 epoch 6 - iter 75/152 - loss 0.12500236 - time (sec): 54.80 - samples/sec: 274.24 - lr: 0.000076 - momentum: 0.000000
2023-10-06 11:52:08,333 epoch 6 - iter 90/152 - loss 0.12450013 - time (sec): 66.44 - samples/sec: 276.25 - lr: 0.000074 - momentum: 0.000000
2023-10-06 11:52:19,166 epoch 6 - iter 105/152 - loss 0.12628717 - time (sec): 77.27 - samples/sec: 275.49 - lr: 0.000072 - momentum: 0.000000
2023-10-06 11:52:31,180 epoch 6 - iter 120/152 - loss 0.12035512 - time (sec): 89.28 - samples/sec: 276.16 - lr: 0.000071 - momentum: 0.000000
2023-10-06 11:52:42,487 epoch 6 - iter 135/152 - loss 0.11594502 - time (sec): 100.59 - samples/sec: 275.16 - lr: 0.000069 - momentum: 0.000000
2023-10-06 11:52:53,613 epoch 6 - iter 150/152 - loss 0.11549344 - time (sec): 111.72 - samples/sec: 274.87 - lr: 0.000067 - momentum: 0.000000
2023-10-06 11:52:54,788 ----------------------------------------------------------------------------------------------------
2023-10-06 11:52:54,788 EPOCH 6 done: loss 0.1152 - lr: 0.000067
2023-10-06 11:53:02,686 DEV : loss 0.16070891916751862 - f1-score (micro avg) 0.7552
2023-10-06 11:53:02,693 saving best model
2023-10-06 11:53:07,043 ----------------------------------------------------------------------------------------------------
2023-10-06 11:53:18,238 epoch 7 - iter 15/152 - loss 0.10658676 - time (sec): 11.19 - samples/sec: 275.80 - lr: 0.000066 - momentum: 0.000000
2023-10-06 11:53:29,396 epoch 7 - iter 30/152 - loss 0.07982796 - time (sec): 22.35 - samples/sec: 278.02 - lr: 0.000064 - momentum: 0.000000
2023-10-06 11:53:40,897 epoch 7 - iter 45/152 - loss 0.08533241 - time (sec): 33.85 - samples/sec: 277.91 - lr: 0.000062 - momentum: 0.000000
2023-10-06 11:53:52,246 epoch 7 - iter 60/152 - loss 0.09230543 - time (sec): 45.20 - samples/sec: 277.75 - lr: 0.000061 - momentum: 0.000000
2023-10-06 11:54:02,737 epoch 7 - iter 75/152 - loss 0.09459485 - time (sec): 55.69 - samples/sec: 275.94 - lr: 0.000059 - momentum: 0.000000
2023-10-06 11:54:13,562 epoch 7 - iter 90/152 - loss 0.09220023 - time (sec): 66.52 - samples/sec: 275.24 - lr: 0.000057 - momentum: 0.000000
2023-10-06 11:54:24,717 epoch 7 - iter 105/152 - loss 0.08957111 - time (sec): 77.67 - samples/sec: 275.39 - lr: 0.000056 - momentum: 0.000000
2023-10-06 11:54:35,526 epoch 7 - iter 120/152 - loss 0.08819805 - time (sec): 88.48 - samples/sec: 274.87 - lr: 0.000054 - momentum: 0.000000
2023-10-06 11:54:47,380 epoch 7 - iter 135/152 - loss 0.08655385 - time (sec): 100.34 - samples/sec: 274.67 - lr: 0.000052 - momentum: 0.000000
2023-10-06 11:54:58,481 epoch 7 - iter 150/152 - loss 0.08719061 - time (sec): 111.44 - samples/sec: 274.63 - lr: 0.000051 - momentum: 0.000000
2023-10-06 11:54:59,869 ----------------------------------------------------------------------------------------------------
2023-10-06 11:54:59,870 EPOCH 7 done: loss 0.0865 - lr: 0.000051
2023-10-06 11:55:07,764 DEV : loss 0.1486317217350006 - f1-score (micro avg) 0.8097
2023-10-06 11:55:07,771 saving best model
2023-10-06 11:55:12,119 ----------------------------------------------------------------------------------------------------
2023-10-06 11:55:23,182 epoch 8 - iter 15/152 - loss 0.07126019 - time (sec): 11.06 - samples/sec: 261.37 - lr: 0.000049 - momentum: 0.000000
2023-10-06 11:55:34,480 epoch 8 - iter 30/152 - loss 0.07210258 - time (sec): 22.36 - samples/sec: 269.42 - lr: 0.000047 - momentum: 0.000000
2023-10-06 11:55:45,567 epoch 8 - iter 45/152 - loss 0.06136012 - time (sec): 33.45 - samples/sec: 272.50 - lr: 0.000046 - momentum: 0.000000
2023-10-06 11:55:57,313 epoch 8 - iter 60/152 - loss 0.06922305 - time (sec): 45.19 - samples/sec: 275.41 - lr: 0.000044 - momentum: 0.000000
2023-10-06 11:56:08,310 epoch 8 - iter 75/152 - loss 0.07252529 - time (sec): 56.19 - samples/sec: 275.68 - lr: 0.000042 - momentum: 0.000000
2023-10-06 11:56:19,141 epoch 8 - iter 90/152 - loss 0.07195863 - time (sec): 67.02 - samples/sec: 273.72 - lr: 0.000041 - momentum: 0.000000
2023-10-06 11:56:30,593 epoch 8 - iter 105/152 - loss 0.06975463 - time (sec): 78.47 - samples/sec: 274.67 - lr: 0.000039 - momentum: 0.000000
2023-10-06 11:56:41,684 epoch 8 - iter 120/152 - loss 0.07045719 - time (sec): 89.56 - samples/sec: 274.61 - lr: 0.000037 - momentum: 0.000000
2023-10-06 11:56:52,427 epoch 8 - iter 135/152 - loss 0.07062132 - time (sec): 100.31 - samples/sec: 273.51 - lr: 0.000036 - momentum: 0.000000
2023-10-06 11:57:03,970 epoch 8 - iter 150/152 - loss 0.06850640 - time (sec): 111.85 - samples/sec: 274.11 - lr: 0.000034 - momentum: 0.000000
2023-10-06 11:57:05,170 ----------------------------------------------------------------------------------------------------
2023-10-06 11:57:05,170 EPOCH 8 done: loss 0.0687 - lr: 0.000034
2023-10-06 11:57:13,130 DEV : loss 0.14806878566741943 - f1-score (micro avg) 0.8158
2023-10-06 11:57:13,137 saving best model
2023-10-06 11:57:17,481 ----------------------------------------------------------------------------------------------------
2023-10-06 11:57:28,825 epoch 9 - iter 15/152 - loss 0.04140531 - time (sec): 11.34 - samples/sec: 270.39 - lr: 0.000032 - momentum: 0.000000
2023-10-06 11:57:39,503 epoch 9 - iter 30/152 - loss 0.05246784 - time (sec): 22.02 - samples/sec: 272.61 - lr: 0.000031 - momentum: 0.000000
2023-10-06 11:57:50,467 epoch 9 - iter 45/152 - loss 0.04910361 - time (sec): 32.98 - samples/sec: 270.21 - lr: 0.000029 - momentum: 0.000000
2023-10-06 11:58:02,104 epoch 9 - iter 60/152 - loss 0.05462582 - time (sec): 44.62 - samples/sec: 273.21 - lr: 0.000027 - momentum: 0.000000
2023-10-06 11:58:13,149 epoch 9 - iter 75/152 - loss 0.05853267 - time (sec): 55.67 - samples/sec: 274.02 - lr: 0.000026 - momentum: 0.000000
2023-10-06 11:58:24,892 epoch 9 - iter 90/152 - loss 0.06375465 - time (sec): 67.41 - samples/sec: 277.39 - lr: 0.000024 - momentum: 0.000000
2023-10-06 11:58:35,543 epoch 9 - iter 105/152 - loss 0.06134915 - time (sec): 78.06 - samples/sec: 275.85 - lr: 0.000022 - momentum: 0.000000
2023-10-06 11:58:47,079 epoch 9 - iter 120/152 - loss 0.06167699 - time (sec): 89.60 - samples/sec: 275.34 - lr: 0.000021 - momentum: 0.000000
2023-10-06 11:58:57,868 epoch 9 - iter 135/152 - loss 0.06111079 - time (sec): 100.39 - samples/sec: 274.50 - lr: 0.000019 - momentum: 0.000000
2023-10-06 11:59:08,947 epoch 9 - iter 150/152 - loss 0.05921126 - time (sec): 111.46 - samples/sec: 274.63 - lr: 0.000018 - momentum: 0.000000
2023-10-06 11:59:10,385 ----------------------------------------------------------------------------------------------------
2023-10-06 11:59:10,386 EPOCH 9 done: loss 0.0591 - lr: 0.000018
2023-10-06 11:59:18,389 DEV : loss 0.14643555879592896 - f1-score (micro avg) 0.8242
2023-10-06 11:59:18,397 saving best model
2023-10-06 11:59:22,732 ----------------------------------------------------------------------------------------------------
2023-10-06 11:59:34,011 epoch 10 - iter 15/152 - loss 0.07490235 - time (sec): 11.28 - samples/sec: 269.22 - lr: 0.000016 - momentum: 0.000000
2023-10-06 11:59:44,906 epoch 10 - iter 30/152 - loss 0.05329799 - time (sec): 22.17 - samples/sec: 268.85 - lr: 0.000014 - momentum: 0.000000
2023-10-06 11:59:56,562 epoch 10 - iter 45/152 - loss 0.04675656 - time (sec): 33.83 - samples/sec: 273.12 - lr: 0.000012 - momentum: 0.000000
2023-10-06 12:00:07,483 epoch 10 - iter 60/152 - loss 0.05173666 - time (sec): 44.75 - samples/sec: 274.71 - lr: 0.000011 - momentum: 0.000000
2023-10-06 12:00:19,503 epoch 10 - iter 75/152 - loss 0.05175693 - time (sec): 56.77 - samples/sec: 276.08 - lr: 0.000009 - momentum: 0.000000
2023-10-06 12:00:30,825 epoch 10 - iter 90/152 - loss 0.05105866 - time (sec): 68.09 - samples/sec: 276.12 - lr: 0.000008 - momentum: 0.000000
2023-10-06 12:00:41,699 epoch 10 - iter 105/152 - loss 0.05368124 - time (sec): 78.97 - samples/sec: 276.30 - lr: 0.000006 - momentum: 0.000000
2023-10-06 12:00:52,416 epoch 10 - iter 120/152 - loss 0.05547270 - time (sec): 89.68 - samples/sec: 275.17 - lr: 0.000004 - momentum: 0.000000
2023-10-06 12:01:03,172 epoch 10 - iter 135/152 - loss 0.05575050 - time (sec): 100.44 - samples/sec: 274.45 - lr: 0.000003 - momentum: 0.000000
2023-10-06 12:01:14,256 epoch 10 - iter 150/152 - loss 0.05479050 - time (sec): 111.52 - samples/sec: 274.36 - lr: 0.000001 - momentum: 0.000000
2023-10-06 12:01:15,634 ----------------------------------------------------------------------------------------------------
2023-10-06 12:01:15,634 EPOCH 10 done: loss 0.0545 - lr: 0.000001
2023-10-06 12:01:23,704 DEV : loss 0.14744949340820312 - f1-score (micro avg) 0.8233
2023-10-06 12:01:24,608 ----------------------------------------------------------------------------------------------------
2023-10-06 12:01:24,617 Loading model from best epoch ...
2023-10-06 12:01:28,740 SequenceTagger predicts: Dictionary with 25 tags: O, S-scope, B-scope, E-scope, I-scope, S-pers, B-pers, E-pers, I-pers, S-work, B-work, E-work, I-work, S-loc, B-loc, E-loc, I-loc, S-date, B-date, E-date, I-date, S-object, B-object, E-object, I-object
2023-10-06 12:01:35,956
Results:
- F-score (micro) 0.8049
- F-score (macro) 0.4917
- Accuracy 0.6813
By class:
precision recall f1-score support
scope 0.7391 0.7881 0.7628 151
pers 0.8070 0.9583 0.8762 96
work 0.7636 0.8842 0.8195 95
loc 0.0000 0.0000 0.0000 3
date 0.0000 0.0000 0.0000 3
micro avg 0.7662 0.8477 0.8049 348
macro avg 0.4620 0.5261 0.4917 348
weighted avg 0.7518 0.8477 0.7964 348
2023-10-06 12:01:35,956 ----------------------------------------------------------------------------------------------------
|