File size: 25,188 Bytes
bcc5022 |
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 |
2023-10-11 14:05:30,951 ----------------------------------------------------------------------------------------------------
2023-10-11 14:05:30,953 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): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
(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): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
(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): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
(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): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
(final_layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=1472, out_features=13, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-11 14:05:30,953 ----------------------------------------------------------------------------------------------------
2023-10-11 14:05:30,954 MultiCorpus: 5777 train + 722 dev + 723 test sentences
- NER_ICDAR_EUROPEANA Corpus: 5777 train + 722 dev + 723 test sentences - /root/.flair/datasets/ner_icdar_europeana/nl
2023-10-11 14:05:30,954 ----------------------------------------------------------------------------------------------------
2023-10-11 14:05:30,954 Train: 5777 sentences
2023-10-11 14:05:30,954 (train_with_dev=False, train_with_test=False)
2023-10-11 14:05:30,954 ----------------------------------------------------------------------------------------------------
2023-10-11 14:05:30,954 Training Params:
2023-10-11 14:05:30,954 - learning_rate: "0.00015"
2023-10-11 14:05:30,954 - mini_batch_size: "8"
2023-10-11 14:05:30,954 - max_epochs: "10"
2023-10-11 14:05:30,954 - shuffle: "True"
2023-10-11 14:05:30,954 ----------------------------------------------------------------------------------------------------
2023-10-11 14:05:30,954 Plugins:
2023-10-11 14:05:30,954 - TensorboardLogger
2023-10-11 14:05:30,954 - LinearScheduler | warmup_fraction: '0.1'
2023-10-11 14:05:30,955 ----------------------------------------------------------------------------------------------------
2023-10-11 14:05:30,955 Final evaluation on model from best epoch (best-model.pt)
2023-10-11 14:05:30,955 - metric: "('micro avg', 'f1-score')"
2023-10-11 14:05:30,955 ----------------------------------------------------------------------------------------------------
2023-10-11 14:05:30,955 Computation:
2023-10-11 14:05:30,955 - compute on device: cuda:0
2023-10-11 14:05:30,955 - embedding storage: none
2023-10-11 14:05:30,955 ----------------------------------------------------------------------------------------------------
2023-10-11 14:05:30,955 Model training base path: "hmbench-icdar/nl-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs8-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-1"
2023-10-11 14:05:30,955 ----------------------------------------------------------------------------------------------------
2023-10-11 14:05:30,955 ----------------------------------------------------------------------------------------------------
2023-10-11 14:05:30,956 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-11 14:06:12,814 epoch 1 - iter 72/723 - loss 2.58705621 - time (sec): 41.86 - samples/sec: 420.87 - lr: 0.000015 - momentum: 0.000000
2023-10-11 14:06:55,356 epoch 1 - iter 144/723 - loss 2.55125919 - time (sec): 84.40 - samples/sec: 440.37 - lr: 0.000030 - momentum: 0.000000
2023-10-11 14:07:32,736 epoch 1 - iter 216/723 - loss 2.42208641 - time (sec): 121.78 - samples/sec: 439.68 - lr: 0.000045 - momentum: 0.000000
2023-10-11 14:08:11,674 epoch 1 - iter 288/723 - loss 2.21128263 - time (sec): 160.72 - samples/sec: 445.55 - lr: 0.000060 - momentum: 0.000000
2023-10-11 14:08:52,057 epoch 1 - iter 360/723 - loss 1.96849538 - time (sec): 201.10 - samples/sec: 453.32 - lr: 0.000074 - momentum: 0.000000
2023-10-11 14:09:30,023 epoch 1 - iter 432/723 - loss 1.75669262 - time (sec): 239.07 - samples/sec: 452.37 - lr: 0.000089 - momentum: 0.000000
2023-10-11 14:10:08,540 epoch 1 - iter 504/723 - loss 1.56586690 - time (sec): 277.58 - samples/sec: 452.47 - lr: 0.000104 - momentum: 0.000000
2023-10-11 14:10:46,586 epoch 1 - iter 576/723 - loss 1.41230121 - time (sec): 315.63 - samples/sec: 451.25 - lr: 0.000119 - momentum: 0.000000
2023-10-11 14:11:23,583 epoch 1 - iter 648/723 - loss 1.29317716 - time (sec): 352.63 - samples/sec: 448.46 - lr: 0.000134 - momentum: 0.000000
2023-10-11 14:12:02,646 epoch 1 - iter 720/723 - loss 1.18252757 - time (sec): 391.69 - samples/sec: 448.88 - lr: 0.000149 - momentum: 0.000000
2023-10-11 14:12:03,721 ----------------------------------------------------------------------------------------------------
2023-10-11 14:12:03,721 EPOCH 1 done: loss 1.1808 - lr: 0.000149
2023-10-11 14:12:23,449 DEV : loss 0.2481938898563385 - f1-score (micro avg) 0.0
2023-10-11 14:12:23,483 ----------------------------------------------------------------------------------------------------
2023-10-11 14:13:02,418 epoch 2 - iter 72/723 - loss 0.19665532 - time (sec): 38.93 - samples/sec: 463.18 - lr: 0.000148 - momentum: 0.000000
2023-10-11 14:13:43,718 epoch 2 - iter 144/723 - loss 0.17259289 - time (sec): 80.23 - samples/sec: 442.69 - lr: 0.000147 - momentum: 0.000000
2023-10-11 14:14:22,701 epoch 2 - iter 216/723 - loss 0.17115371 - time (sec): 119.22 - samples/sec: 433.56 - lr: 0.000145 - momentum: 0.000000
2023-10-11 14:15:03,146 epoch 2 - iter 288/723 - loss 0.16331197 - time (sec): 159.66 - samples/sec: 434.40 - lr: 0.000143 - momentum: 0.000000
2023-10-11 14:15:43,402 epoch 2 - iter 360/723 - loss 0.16009303 - time (sec): 199.92 - samples/sec: 437.18 - lr: 0.000142 - momentum: 0.000000
2023-10-11 14:16:23,005 epoch 2 - iter 432/723 - loss 0.15669479 - time (sec): 239.52 - samples/sec: 437.77 - lr: 0.000140 - momentum: 0.000000
2023-10-11 14:17:02,620 epoch 2 - iter 504/723 - loss 0.14903751 - time (sec): 279.14 - samples/sec: 441.76 - lr: 0.000138 - momentum: 0.000000
2023-10-11 14:17:43,533 epoch 2 - iter 576/723 - loss 0.14173836 - time (sec): 320.05 - samples/sec: 446.76 - lr: 0.000137 - momentum: 0.000000
2023-10-11 14:18:21,648 epoch 2 - iter 648/723 - loss 0.13875105 - time (sec): 358.16 - samples/sec: 444.18 - lr: 0.000135 - momentum: 0.000000
2023-10-11 14:19:03,265 epoch 2 - iter 720/723 - loss 0.13671767 - time (sec): 399.78 - samples/sec: 438.93 - lr: 0.000133 - momentum: 0.000000
2023-10-11 14:19:04,918 ----------------------------------------------------------------------------------------------------
2023-10-11 14:19:04,919 EPOCH 2 done: loss 0.1366 - lr: 0.000133
2023-10-11 14:19:26,427 DEV : loss 0.11942420899868011 - f1-score (micro avg) 0.7416
2023-10-11 14:19:26,458 saving best model
2023-10-11 14:19:27,827 ----------------------------------------------------------------------------------------------------
2023-10-11 14:20:12,005 epoch 3 - iter 72/723 - loss 0.08259658 - time (sec): 44.18 - samples/sec: 390.95 - lr: 0.000132 - momentum: 0.000000
2023-10-11 14:20:53,017 epoch 3 - iter 144/723 - loss 0.09033580 - time (sec): 85.19 - samples/sec: 404.11 - lr: 0.000130 - momentum: 0.000000
2023-10-11 14:21:34,442 epoch 3 - iter 216/723 - loss 0.08477045 - time (sec): 126.61 - samples/sec: 410.89 - lr: 0.000128 - momentum: 0.000000
2023-10-11 14:22:14,549 epoch 3 - iter 288/723 - loss 0.09192864 - time (sec): 166.72 - samples/sec: 411.77 - lr: 0.000127 - momentum: 0.000000
2023-10-11 14:22:55,726 epoch 3 - iter 360/723 - loss 0.08964205 - time (sec): 207.90 - samples/sec: 419.60 - lr: 0.000125 - momentum: 0.000000
2023-10-11 14:23:36,234 epoch 3 - iter 432/723 - loss 0.08993680 - time (sec): 248.40 - samples/sec: 418.75 - lr: 0.000123 - momentum: 0.000000
2023-10-11 14:24:14,710 epoch 3 - iter 504/723 - loss 0.08843282 - time (sec): 286.88 - samples/sec: 423.56 - lr: 0.000122 - momentum: 0.000000
2023-10-11 14:24:54,374 epoch 3 - iter 576/723 - loss 0.08469597 - time (sec): 326.55 - samples/sec: 427.94 - lr: 0.000120 - momentum: 0.000000
2023-10-11 14:25:33,840 epoch 3 - iter 648/723 - loss 0.08177926 - time (sec): 366.01 - samples/sec: 431.84 - lr: 0.000118 - momentum: 0.000000
2023-10-11 14:26:15,057 epoch 3 - iter 720/723 - loss 0.08098319 - time (sec): 407.23 - samples/sec: 431.32 - lr: 0.000117 - momentum: 0.000000
2023-10-11 14:26:16,272 ----------------------------------------------------------------------------------------------------
2023-10-11 14:26:16,272 EPOCH 3 done: loss 0.0810 - lr: 0.000117
2023-10-11 14:26:39,318 DEV : loss 0.0933394655585289 - f1-score (micro avg) 0.7843
2023-10-11 14:26:39,365 saving best model
2023-10-11 14:26:42,488 ----------------------------------------------------------------------------------------------------
2023-10-11 14:27:23,951 epoch 4 - iter 72/723 - loss 0.06794431 - time (sec): 41.46 - samples/sec: 436.19 - lr: 0.000115 - momentum: 0.000000
2023-10-11 14:28:04,485 epoch 4 - iter 144/723 - loss 0.05806820 - time (sec): 81.99 - samples/sec: 427.75 - lr: 0.000113 - momentum: 0.000000
2023-10-11 14:28:44,946 epoch 4 - iter 216/723 - loss 0.05532765 - time (sec): 122.45 - samples/sec: 431.87 - lr: 0.000112 - momentum: 0.000000
2023-10-11 14:29:25,877 epoch 4 - iter 288/723 - loss 0.05495337 - time (sec): 163.38 - samples/sec: 430.00 - lr: 0.000110 - momentum: 0.000000
2023-10-11 14:30:06,043 epoch 4 - iter 360/723 - loss 0.05464785 - time (sec): 203.55 - samples/sec: 428.59 - lr: 0.000108 - momentum: 0.000000
2023-10-11 14:30:43,057 epoch 4 - iter 432/723 - loss 0.05467701 - time (sec): 240.56 - samples/sec: 431.40 - lr: 0.000107 - momentum: 0.000000
2023-10-11 14:31:22,681 epoch 4 - iter 504/723 - loss 0.05477446 - time (sec): 280.19 - samples/sec: 431.78 - lr: 0.000105 - momentum: 0.000000
2023-10-11 14:32:04,286 epoch 4 - iter 576/723 - loss 0.05566990 - time (sec): 321.79 - samples/sec: 432.98 - lr: 0.000103 - momentum: 0.000000
2023-10-11 14:32:42,989 epoch 4 - iter 648/723 - loss 0.05417209 - time (sec): 360.49 - samples/sec: 436.05 - lr: 0.000102 - momentum: 0.000000
2023-10-11 14:33:21,891 epoch 4 - iter 720/723 - loss 0.05222244 - time (sec): 399.40 - samples/sec: 440.11 - lr: 0.000100 - momentum: 0.000000
2023-10-11 14:33:22,951 ----------------------------------------------------------------------------------------------------
2023-10-11 14:33:22,952 EPOCH 4 done: loss 0.0522 - lr: 0.000100
2023-10-11 14:33:43,757 DEV : loss 0.0753728598356247 - f1-score (micro avg) 0.8404
2023-10-11 14:33:43,792 saving best model
2023-10-11 14:33:46,481 ----------------------------------------------------------------------------------------------------
2023-10-11 14:34:26,467 epoch 5 - iter 72/723 - loss 0.02836811 - time (sec): 39.98 - samples/sec: 446.76 - lr: 0.000098 - momentum: 0.000000
2023-10-11 14:35:04,451 epoch 5 - iter 144/723 - loss 0.03014743 - time (sec): 77.96 - samples/sec: 443.42 - lr: 0.000097 - momentum: 0.000000
2023-10-11 14:35:43,263 epoch 5 - iter 216/723 - loss 0.03649224 - time (sec): 116.77 - samples/sec: 439.91 - lr: 0.000095 - momentum: 0.000000
2023-10-11 14:36:24,885 epoch 5 - iter 288/723 - loss 0.03442032 - time (sec): 158.40 - samples/sec: 436.97 - lr: 0.000093 - momentum: 0.000000
2023-10-11 14:37:11,469 epoch 5 - iter 360/723 - loss 0.03751789 - time (sec): 204.98 - samples/sec: 428.52 - lr: 0.000092 - momentum: 0.000000
2023-10-11 14:37:56,871 epoch 5 - iter 432/723 - loss 0.03539090 - time (sec): 250.38 - samples/sec: 416.37 - lr: 0.000090 - momentum: 0.000000
2023-10-11 14:38:39,139 epoch 5 - iter 504/723 - loss 0.03487207 - time (sec): 292.65 - samples/sec: 416.32 - lr: 0.000088 - momentum: 0.000000
2023-10-11 14:39:21,228 epoch 5 - iter 576/723 - loss 0.03534212 - time (sec): 334.74 - samples/sec: 420.39 - lr: 0.000087 - momentum: 0.000000
2023-10-11 14:40:01,428 epoch 5 - iter 648/723 - loss 0.03542313 - time (sec): 374.94 - samples/sec: 420.10 - lr: 0.000085 - momentum: 0.000000
2023-10-11 14:40:41,071 epoch 5 - iter 720/723 - loss 0.03559847 - time (sec): 414.58 - samples/sec: 423.82 - lr: 0.000083 - momentum: 0.000000
2023-10-11 14:40:42,254 ----------------------------------------------------------------------------------------------------
2023-10-11 14:40:42,254 EPOCH 5 done: loss 0.0356 - lr: 0.000083
2023-10-11 14:41:04,297 DEV : loss 0.08206792920827866 - f1-score (micro avg) 0.8407
2023-10-11 14:41:04,332 saving best model
2023-10-11 14:41:06,999 ----------------------------------------------------------------------------------------------------
2023-10-11 14:41:44,849 epoch 6 - iter 72/723 - loss 0.02906877 - time (sec): 37.84 - samples/sec: 444.50 - lr: 0.000082 - momentum: 0.000000
2023-10-11 14:42:25,331 epoch 6 - iter 144/723 - loss 0.02818642 - time (sec): 78.32 - samples/sec: 434.40 - lr: 0.000080 - momentum: 0.000000
2023-10-11 14:43:06,681 epoch 6 - iter 216/723 - loss 0.02990584 - time (sec): 119.67 - samples/sec: 433.58 - lr: 0.000078 - momentum: 0.000000
2023-10-11 14:43:47,544 epoch 6 - iter 288/723 - loss 0.02714357 - time (sec): 160.53 - samples/sec: 436.14 - lr: 0.000077 - momentum: 0.000000
2023-10-11 14:44:29,330 epoch 6 - iter 360/723 - loss 0.02747624 - time (sec): 202.32 - samples/sec: 436.90 - lr: 0.000075 - momentum: 0.000000
2023-10-11 14:45:10,018 epoch 6 - iter 432/723 - loss 0.02687441 - time (sec): 243.01 - samples/sec: 433.75 - lr: 0.000073 - momentum: 0.000000
2023-10-11 14:45:52,721 epoch 6 - iter 504/723 - loss 0.02594408 - time (sec): 285.71 - samples/sec: 427.66 - lr: 0.000072 - momentum: 0.000000
2023-10-11 14:46:37,946 epoch 6 - iter 576/723 - loss 0.02542891 - time (sec): 330.94 - samples/sec: 422.42 - lr: 0.000070 - momentum: 0.000000
2023-10-11 14:47:23,370 epoch 6 - iter 648/723 - loss 0.02627547 - time (sec): 376.36 - samples/sec: 423.51 - lr: 0.000068 - momentum: 0.000000
2023-10-11 14:48:05,278 epoch 6 - iter 720/723 - loss 0.02645835 - time (sec): 418.27 - samples/sec: 420.22 - lr: 0.000067 - momentum: 0.000000
2023-10-11 14:48:06,364 ----------------------------------------------------------------------------------------------------
2023-10-11 14:48:06,364 EPOCH 6 done: loss 0.0266 - lr: 0.000067
2023-10-11 14:48:26,664 DEV : loss 0.08414550870656967 - f1-score (micro avg) 0.871
2023-10-11 14:48:26,696 saving best model
2023-10-11 14:48:29,329 ----------------------------------------------------------------------------------------------------
2023-10-11 14:49:10,549 epoch 7 - iter 72/723 - loss 0.02559871 - time (sec): 41.22 - samples/sec: 434.37 - lr: 0.000065 - momentum: 0.000000
2023-10-11 14:49:50,528 epoch 7 - iter 144/723 - loss 0.02202443 - time (sec): 81.19 - samples/sec: 429.79 - lr: 0.000063 - momentum: 0.000000
2023-10-11 14:50:33,114 epoch 7 - iter 216/723 - loss 0.01970189 - time (sec): 123.78 - samples/sec: 418.63 - lr: 0.000062 - momentum: 0.000000
2023-10-11 14:51:15,905 epoch 7 - iter 288/723 - loss 0.02107156 - time (sec): 166.57 - samples/sec: 413.39 - lr: 0.000060 - momentum: 0.000000
2023-10-11 14:51:59,500 epoch 7 - iter 360/723 - loss 0.02138550 - time (sec): 210.17 - samples/sec: 411.59 - lr: 0.000058 - momentum: 0.000000
2023-10-11 14:52:41,711 epoch 7 - iter 432/723 - loss 0.02195180 - time (sec): 252.38 - samples/sec: 413.71 - lr: 0.000057 - momentum: 0.000000
2023-10-11 14:53:22,689 epoch 7 - iter 504/723 - loss 0.02270591 - time (sec): 293.36 - samples/sec: 417.57 - lr: 0.000055 - momentum: 0.000000
2023-10-11 14:54:02,112 epoch 7 - iter 576/723 - loss 0.02192802 - time (sec): 332.78 - samples/sec: 420.35 - lr: 0.000053 - momentum: 0.000000
2023-10-11 14:54:41,762 epoch 7 - iter 648/723 - loss 0.02326251 - time (sec): 372.43 - samples/sec: 422.01 - lr: 0.000052 - momentum: 0.000000
2023-10-11 14:55:21,919 epoch 7 - iter 720/723 - loss 0.02251924 - time (sec): 412.59 - samples/sec: 425.38 - lr: 0.000050 - momentum: 0.000000
2023-10-11 14:55:23,238 ----------------------------------------------------------------------------------------------------
2023-10-11 14:55:23,239 EPOCH 7 done: loss 0.0225 - lr: 0.000050
2023-10-11 14:55:45,123 DEV : loss 0.09741368144750595 - f1-score (micro avg) 0.86
2023-10-11 14:55:45,161 ----------------------------------------------------------------------------------------------------
2023-10-11 14:56:25,043 epoch 8 - iter 72/723 - loss 0.01566425 - time (sec): 39.88 - samples/sec: 470.64 - lr: 0.000048 - momentum: 0.000000
2023-10-11 14:57:03,296 epoch 8 - iter 144/723 - loss 0.01425224 - time (sec): 78.13 - samples/sec: 460.33 - lr: 0.000047 - momentum: 0.000000
2023-10-11 14:57:41,996 epoch 8 - iter 216/723 - loss 0.01421036 - time (sec): 116.83 - samples/sec: 454.02 - lr: 0.000045 - momentum: 0.000000
2023-10-11 14:58:20,720 epoch 8 - iter 288/723 - loss 0.01334895 - time (sec): 155.56 - samples/sec: 449.16 - lr: 0.000043 - momentum: 0.000000
2023-10-11 14:59:01,048 epoch 8 - iter 360/723 - loss 0.01534408 - time (sec): 195.89 - samples/sec: 448.40 - lr: 0.000042 - momentum: 0.000000
2023-10-11 14:59:42,233 epoch 8 - iter 432/723 - loss 0.01520775 - time (sec): 237.07 - samples/sec: 447.96 - lr: 0.000040 - momentum: 0.000000
2023-10-11 15:00:22,379 epoch 8 - iter 504/723 - loss 0.01628644 - time (sec): 277.22 - samples/sec: 447.92 - lr: 0.000038 - momentum: 0.000000
2023-10-11 15:01:03,684 epoch 8 - iter 576/723 - loss 0.01780027 - time (sec): 318.52 - samples/sec: 447.22 - lr: 0.000037 - momentum: 0.000000
2023-10-11 15:01:46,996 epoch 8 - iter 648/723 - loss 0.01751548 - time (sec): 361.83 - samples/sec: 437.74 - lr: 0.000035 - momentum: 0.000000
2023-10-11 15:02:30,881 epoch 8 - iter 720/723 - loss 0.01788380 - time (sec): 405.72 - samples/sec: 432.99 - lr: 0.000033 - momentum: 0.000000
2023-10-11 15:02:32,251 ----------------------------------------------------------------------------------------------------
2023-10-11 15:02:32,251 EPOCH 8 done: loss 0.0179 - lr: 0.000033
2023-10-11 15:02:54,685 DEV : loss 0.11365482956171036 - f1-score (micro avg) 0.8415
2023-10-11 15:02:54,716 ----------------------------------------------------------------------------------------------------
2023-10-11 15:03:41,064 epoch 9 - iter 72/723 - loss 0.02258103 - time (sec): 46.35 - samples/sec: 404.98 - lr: 0.000032 - momentum: 0.000000
2023-10-11 15:04:22,261 epoch 9 - iter 144/723 - loss 0.01587706 - time (sec): 87.54 - samples/sec: 402.96 - lr: 0.000030 - momentum: 0.000000
2023-10-11 15:05:06,366 epoch 9 - iter 216/723 - loss 0.01536952 - time (sec): 131.65 - samples/sec: 409.42 - lr: 0.000028 - momentum: 0.000000
2023-10-11 15:05:51,100 epoch 9 - iter 288/723 - loss 0.01464115 - time (sec): 176.38 - samples/sec: 404.84 - lr: 0.000027 - momentum: 0.000000
2023-10-11 15:06:36,197 epoch 9 - iter 360/723 - loss 0.01451980 - time (sec): 221.48 - samples/sec: 400.55 - lr: 0.000025 - momentum: 0.000000
2023-10-11 15:07:21,999 epoch 9 - iter 432/723 - loss 0.01468506 - time (sec): 267.28 - samples/sec: 397.94 - lr: 0.000023 - momentum: 0.000000
2023-10-11 15:08:04,206 epoch 9 - iter 504/723 - loss 0.01473622 - time (sec): 309.49 - samples/sec: 403.27 - lr: 0.000022 - momentum: 0.000000
2023-10-11 15:08:45,094 epoch 9 - iter 576/723 - loss 0.01522224 - time (sec): 350.38 - samples/sec: 403.83 - lr: 0.000020 - momentum: 0.000000
2023-10-11 15:09:24,875 epoch 9 - iter 648/723 - loss 0.01442489 - time (sec): 390.16 - samples/sec: 406.21 - lr: 0.000018 - momentum: 0.000000
2023-10-11 15:10:04,792 epoch 9 - iter 720/723 - loss 0.01450651 - time (sec): 430.07 - samples/sec: 408.82 - lr: 0.000017 - momentum: 0.000000
2023-10-11 15:10:05,942 ----------------------------------------------------------------------------------------------------
2023-10-11 15:10:05,942 EPOCH 9 done: loss 0.0145 - lr: 0.000017
2023-10-11 15:10:26,729 DEV : loss 0.11227083206176758 - f1-score (micro avg) 0.8546
2023-10-11 15:10:26,766 ----------------------------------------------------------------------------------------------------
2023-10-11 15:11:05,747 epoch 10 - iter 72/723 - loss 0.00745596 - time (sec): 38.98 - samples/sec: 430.39 - lr: 0.000015 - momentum: 0.000000
2023-10-11 15:11:47,734 epoch 10 - iter 144/723 - loss 0.01171360 - time (sec): 80.96 - samples/sec: 435.46 - lr: 0.000013 - momentum: 0.000000
2023-10-11 15:12:31,352 epoch 10 - iter 216/723 - loss 0.01187809 - time (sec): 124.58 - samples/sec: 420.80 - lr: 0.000012 - momentum: 0.000000
2023-10-11 15:13:13,766 epoch 10 - iter 288/723 - loss 0.01086224 - time (sec): 167.00 - samples/sec: 414.49 - lr: 0.000010 - momentum: 0.000000
2023-10-11 15:13:58,505 epoch 10 - iter 360/723 - loss 0.01143704 - time (sec): 211.74 - samples/sec: 414.27 - lr: 0.000008 - momentum: 0.000000
2023-10-11 15:14:39,915 epoch 10 - iter 432/723 - loss 0.01105902 - time (sec): 253.15 - samples/sec: 418.84 - lr: 0.000007 - momentum: 0.000000
2023-10-11 15:15:20,796 epoch 10 - iter 504/723 - loss 0.01259970 - time (sec): 294.03 - samples/sec: 422.08 - lr: 0.000005 - momentum: 0.000000
2023-10-11 15:16:00,222 epoch 10 - iter 576/723 - loss 0.01185845 - time (sec): 333.45 - samples/sec: 422.00 - lr: 0.000003 - momentum: 0.000000
2023-10-11 15:16:40,161 epoch 10 - iter 648/723 - loss 0.01209669 - time (sec): 373.39 - samples/sec: 424.20 - lr: 0.000002 - momentum: 0.000000
2023-10-11 15:17:19,030 epoch 10 - iter 720/723 - loss 0.01209148 - time (sec): 412.26 - samples/sec: 425.91 - lr: 0.000000 - momentum: 0.000000
2023-10-11 15:17:20,270 ----------------------------------------------------------------------------------------------------
2023-10-11 15:17:20,270 EPOCH 10 done: loss 0.0121 - lr: 0.000000
2023-10-11 15:17:41,620 DEV : loss 0.12011167407035828 - f1-score (micro avg) 0.8502
2023-10-11 15:17:42,629 ----------------------------------------------------------------------------------------------------
2023-10-11 15:17:42,632 Loading model from best epoch ...
2023-10-11 15:17:48,592 SequenceTagger predicts: Dictionary with 13 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
2023-10-11 15:18:08,765
Results:
- F-score (micro) 0.8516
- F-score (macro) 0.7648
- Accuracy 0.7513
By class:
precision recall f1-score support
PER 0.8173 0.8817 0.8483 482
LOC 0.9240 0.8755 0.8991 458
ORG 0.5932 0.5072 0.5469 69
micro avg 0.8500 0.8533 0.8516 1009
macro avg 0.7782 0.7548 0.7648 1009
weighted avg 0.8504 0.8533 0.8507 1009
2023-10-11 15:18:08,765 ----------------------------------------------------------------------------------------------------
|