File size: 25,190 Bytes
f29f2eb |
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 |
2023-10-12 17:47:39,484 ----------------------------------------------------------------------------------------------------
2023-10-12 17:47:39,486 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()
)"
2023-10-12 17:47:39,486 ----------------------------------------------------------------------------------------------------
2023-10-12 17:47:39,486 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
2023-10-12 17:47:39,487 ----------------------------------------------------------------------------------------------------
2023-10-12 17:47:39,487 Train: 5777 sentences
2023-10-12 17:47:39,487 (train_with_dev=False, train_with_test=False)
2023-10-12 17:47:39,487 ----------------------------------------------------------------------------------------------------
2023-10-12 17:47:39,487 Training Params:
2023-10-12 17:47:39,487 - learning_rate: "0.00015"
2023-10-12 17:47:39,487 - mini_batch_size: "8"
2023-10-12 17:47:39,487 - max_epochs: "10"
2023-10-12 17:47:39,487 - shuffle: "True"
2023-10-12 17:47:39,487 ----------------------------------------------------------------------------------------------------
2023-10-12 17:47:39,487 Plugins:
2023-10-12 17:47:39,487 - TensorboardLogger
2023-10-12 17:47:39,487 - LinearScheduler | warmup_fraction: '0.1'
2023-10-12 17:47:39,487 ----------------------------------------------------------------------------------------------------
2023-10-12 17:47:39,488 Final evaluation on model from best epoch (best-model.pt)
2023-10-12 17:47:39,488 - metric: "('micro avg', 'f1-score')"
2023-10-12 17:47:39,488 ----------------------------------------------------------------------------------------------------
2023-10-12 17:47:39,488 Computation:
2023-10-12 17:47:39,488 - compute on device: cuda:0
2023-10-12 17:47:39,488 - embedding storage: none
2023-10-12 17:47:39,488 ----------------------------------------------------------------------------------------------------
2023-10-12 17:47:39,488 Model training base path: "hmbench-icdar/nl-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs8-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-5"
2023-10-12 17:47:39,488 ----------------------------------------------------------------------------------------------------
2023-10-12 17:47:39,488 ----------------------------------------------------------------------------------------------------
2023-10-12 17:47:39,488 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-12 17:48:18,952 epoch 1 - iter 72/723 - loss 2.54426887 - time (sec): 39.46 - samples/sec: 457.63 - lr: 0.000015 - momentum: 0.000000
2023-10-12 17:48:58,124 epoch 1 - iter 144/723 - loss 2.48027812 - time (sec): 78.63 - samples/sec: 459.08 - lr: 0.000030 - momentum: 0.000000
2023-10-12 17:49:36,310 epoch 1 - iter 216/723 - loss 2.32386957 - time (sec): 116.82 - samples/sec: 448.35 - lr: 0.000045 - momentum: 0.000000
2023-10-12 17:50:14,644 epoch 1 - iter 288/723 - loss 2.11630297 - time (sec): 155.15 - samples/sec: 451.57 - lr: 0.000060 - momentum: 0.000000
2023-10-12 17:50:53,387 epoch 1 - iter 360/723 - loss 1.89885572 - time (sec): 193.90 - samples/sec: 452.96 - lr: 0.000074 - momentum: 0.000000
2023-10-12 17:51:31,686 epoch 1 - iter 432/723 - loss 1.68800148 - time (sec): 232.20 - samples/sec: 450.91 - lr: 0.000089 - momentum: 0.000000
2023-10-12 17:52:11,717 epoch 1 - iter 504/723 - loss 1.48803245 - time (sec): 272.23 - samples/sec: 449.72 - lr: 0.000104 - momentum: 0.000000
2023-10-12 17:52:51,394 epoch 1 - iter 576/723 - loss 1.34139358 - time (sec): 311.90 - samples/sec: 446.33 - lr: 0.000119 - momentum: 0.000000
2023-10-12 17:53:31,923 epoch 1 - iter 648/723 - loss 1.21633560 - time (sec): 352.43 - samples/sec: 445.27 - lr: 0.000134 - momentum: 0.000000
2023-10-12 17:54:12,600 epoch 1 - iter 720/723 - loss 1.10698770 - time (sec): 393.11 - samples/sec: 446.34 - lr: 0.000149 - momentum: 0.000000
2023-10-12 17:54:13,966 ----------------------------------------------------------------------------------------------------
2023-10-12 17:54:13,967 EPOCH 1 done: loss 1.1029 - lr: 0.000149
2023-10-12 17:54:34,128 DEV : loss 0.23375172913074493 - f1-score (micro avg) 0.0
2023-10-12 17:54:34,158 ----------------------------------------------------------------------------------------------------
2023-10-12 17:55:12,547 epoch 2 - iter 72/723 - loss 0.16731668 - time (sec): 38.39 - samples/sec: 454.44 - lr: 0.000148 - momentum: 0.000000
2023-10-12 17:55:51,979 epoch 2 - iter 144/723 - loss 0.15482755 - time (sec): 77.82 - samples/sec: 455.74 - lr: 0.000147 - momentum: 0.000000
2023-10-12 17:56:29,875 epoch 2 - iter 216/723 - loss 0.15195012 - time (sec): 115.71 - samples/sec: 449.25 - lr: 0.000145 - momentum: 0.000000
2023-10-12 17:57:09,651 epoch 2 - iter 288/723 - loss 0.14875477 - time (sec): 155.49 - samples/sec: 448.78 - lr: 0.000143 - momentum: 0.000000
2023-10-12 17:57:48,585 epoch 2 - iter 360/723 - loss 0.14422756 - time (sec): 194.42 - samples/sec: 448.71 - lr: 0.000142 - momentum: 0.000000
2023-10-12 17:58:27,126 epoch 2 - iter 432/723 - loss 0.14054029 - time (sec): 232.97 - samples/sec: 449.87 - lr: 0.000140 - momentum: 0.000000
2023-10-12 17:59:05,553 epoch 2 - iter 504/723 - loss 0.14009191 - time (sec): 271.39 - samples/sec: 449.88 - lr: 0.000138 - momentum: 0.000000
2023-10-12 17:59:44,287 epoch 2 - iter 576/723 - loss 0.13710606 - time (sec): 310.13 - samples/sec: 451.15 - lr: 0.000137 - momentum: 0.000000
2023-10-12 18:00:23,907 epoch 2 - iter 648/723 - loss 0.13415472 - time (sec): 349.75 - samples/sec: 452.30 - lr: 0.000135 - momentum: 0.000000
2023-10-12 18:01:03,365 epoch 2 - iter 720/723 - loss 0.13018803 - time (sec): 389.20 - samples/sec: 451.19 - lr: 0.000133 - momentum: 0.000000
2023-10-12 18:01:04,560 ----------------------------------------------------------------------------------------------------
2023-10-12 18:01:04,560 EPOCH 2 done: loss 0.1300 - lr: 0.000133
2023-10-12 18:01:25,127 DEV : loss 0.11187870055437088 - f1-score (micro avg) 0.7651
2023-10-12 18:01:25,159 saving best model
2023-10-12 18:01:26,051 ----------------------------------------------------------------------------------------------------
2023-10-12 18:02:05,492 epoch 3 - iter 72/723 - loss 0.08592561 - time (sec): 39.44 - samples/sec: 453.97 - lr: 0.000132 - momentum: 0.000000
2023-10-12 18:02:44,351 epoch 3 - iter 144/723 - loss 0.08297721 - time (sec): 78.30 - samples/sec: 459.33 - lr: 0.000130 - momentum: 0.000000
2023-10-12 18:03:22,483 epoch 3 - iter 216/723 - loss 0.08326043 - time (sec): 116.43 - samples/sec: 459.71 - lr: 0.000128 - momentum: 0.000000
2023-10-12 18:04:00,111 epoch 3 - iter 288/723 - loss 0.08082376 - time (sec): 154.06 - samples/sec: 461.41 - lr: 0.000127 - momentum: 0.000000
2023-10-12 18:04:38,650 epoch 3 - iter 360/723 - loss 0.08052324 - time (sec): 192.60 - samples/sec: 462.31 - lr: 0.000125 - momentum: 0.000000
2023-10-12 18:05:17,244 epoch 3 - iter 432/723 - loss 0.08037304 - time (sec): 231.19 - samples/sec: 466.15 - lr: 0.000123 - momentum: 0.000000
2023-10-12 18:05:56,024 epoch 3 - iter 504/723 - loss 0.07975153 - time (sec): 269.97 - samples/sec: 463.63 - lr: 0.000122 - momentum: 0.000000
2023-10-12 18:06:34,552 epoch 3 - iter 576/723 - loss 0.07988037 - time (sec): 308.50 - samples/sec: 460.86 - lr: 0.000120 - momentum: 0.000000
2023-10-12 18:07:12,746 epoch 3 - iter 648/723 - loss 0.07895931 - time (sec): 346.69 - samples/sec: 457.82 - lr: 0.000118 - momentum: 0.000000
2023-10-12 18:07:50,890 epoch 3 - iter 720/723 - loss 0.07794918 - time (sec): 384.84 - samples/sec: 456.08 - lr: 0.000117 - momentum: 0.000000
2023-10-12 18:07:52,195 ----------------------------------------------------------------------------------------------------
2023-10-12 18:07:52,195 EPOCH 3 done: loss 0.0781 - lr: 0.000117
2023-10-12 18:08:14,169 DEV : loss 0.0785035490989685 - f1-score (micro avg) 0.8324
2023-10-12 18:08:14,200 saving best model
2023-10-12 18:08:16,769 ----------------------------------------------------------------------------------------------------
2023-10-12 18:08:58,498 epoch 4 - iter 72/723 - loss 0.05759562 - time (sec): 41.72 - samples/sec: 429.84 - lr: 0.000115 - momentum: 0.000000
2023-10-12 18:09:39,681 epoch 4 - iter 144/723 - loss 0.05741867 - time (sec): 82.91 - samples/sec: 419.13 - lr: 0.000113 - momentum: 0.000000
2023-10-12 18:10:20,375 epoch 4 - iter 216/723 - loss 0.05415096 - time (sec): 123.60 - samples/sec: 424.18 - lr: 0.000112 - momentum: 0.000000
2023-10-12 18:11:00,926 epoch 4 - iter 288/723 - loss 0.05445295 - time (sec): 164.15 - samples/sec: 435.34 - lr: 0.000110 - momentum: 0.000000
2023-10-12 18:11:39,771 epoch 4 - iter 360/723 - loss 0.05261383 - time (sec): 203.00 - samples/sec: 439.52 - lr: 0.000108 - momentum: 0.000000
2023-10-12 18:12:18,198 epoch 4 - iter 432/723 - loss 0.05135342 - time (sec): 241.42 - samples/sec: 439.86 - lr: 0.000107 - momentum: 0.000000
2023-10-12 18:12:57,140 epoch 4 - iter 504/723 - loss 0.05062735 - time (sec): 280.37 - samples/sec: 439.37 - lr: 0.000105 - momentum: 0.000000
2023-10-12 18:13:36,398 epoch 4 - iter 576/723 - loss 0.04994851 - time (sec): 319.62 - samples/sec: 442.45 - lr: 0.000103 - momentum: 0.000000
2023-10-12 18:14:15,270 epoch 4 - iter 648/723 - loss 0.05141664 - time (sec): 358.50 - samples/sec: 443.12 - lr: 0.000102 - momentum: 0.000000
2023-10-12 18:14:53,270 epoch 4 - iter 720/723 - loss 0.05052236 - time (sec): 396.50 - samples/sec: 443.46 - lr: 0.000100 - momentum: 0.000000
2023-10-12 18:14:54,369 ----------------------------------------------------------------------------------------------------
2023-10-12 18:14:54,369 EPOCH 4 done: loss 0.0506 - lr: 0.000100
2023-10-12 18:15:15,356 DEV : loss 0.07858217507600784 - f1-score (micro avg) 0.8518
2023-10-12 18:15:15,388 saving best model
2023-10-12 18:15:17,971 ----------------------------------------------------------------------------------------------------
2023-10-12 18:15:59,176 epoch 5 - iter 72/723 - loss 0.03874655 - time (sec): 41.20 - samples/sec: 458.15 - lr: 0.000098 - momentum: 0.000000
2023-10-12 18:16:39,098 epoch 5 - iter 144/723 - loss 0.03347685 - time (sec): 81.12 - samples/sec: 447.35 - lr: 0.000097 - momentum: 0.000000
2023-10-12 18:17:16,677 epoch 5 - iter 216/723 - loss 0.03299884 - time (sec): 118.70 - samples/sec: 436.84 - lr: 0.000095 - momentum: 0.000000
2023-10-12 18:17:54,188 epoch 5 - iter 288/723 - loss 0.03250430 - time (sec): 156.21 - samples/sec: 435.65 - lr: 0.000093 - momentum: 0.000000
2023-10-12 18:18:33,812 epoch 5 - iter 360/723 - loss 0.03475104 - time (sec): 195.84 - samples/sec: 442.33 - lr: 0.000092 - momentum: 0.000000
2023-10-12 18:19:12,566 epoch 5 - iter 432/723 - loss 0.03403611 - time (sec): 234.59 - samples/sec: 443.37 - lr: 0.000090 - momentum: 0.000000
2023-10-12 18:19:52,453 epoch 5 - iter 504/723 - loss 0.03407641 - time (sec): 274.48 - samples/sec: 446.40 - lr: 0.000088 - momentum: 0.000000
2023-10-12 18:20:31,367 epoch 5 - iter 576/723 - loss 0.03399255 - time (sec): 313.39 - samples/sec: 447.69 - lr: 0.000087 - momentum: 0.000000
2023-10-12 18:21:10,192 epoch 5 - iter 648/723 - loss 0.03424309 - time (sec): 352.22 - samples/sec: 448.30 - lr: 0.000085 - momentum: 0.000000
2023-10-12 18:21:48,915 epoch 5 - iter 720/723 - loss 0.03511790 - time (sec): 390.94 - samples/sec: 448.58 - lr: 0.000083 - momentum: 0.000000
2023-10-12 18:21:50,307 ----------------------------------------------------------------------------------------------------
2023-10-12 18:21:50,308 EPOCH 5 done: loss 0.0353 - lr: 0.000083
2023-10-12 18:22:12,540 DEV : loss 0.0762009546160698 - f1-score (micro avg) 0.8552
2023-10-12 18:22:12,572 saving best model
2023-10-12 18:22:15,194 ----------------------------------------------------------------------------------------------------
2023-10-12 18:22:54,690 epoch 6 - iter 72/723 - loss 0.02292044 - time (sec): 39.49 - samples/sec: 435.70 - lr: 0.000082 - momentum: 0.000000
2023-10-12 18:23:33,953 epoch 6 - iter 144/723 - loss 0.02437413 - time (sec): 78.75 - samples/sec: 438.41 - lr: 0.000080 - momentum: 0.000000
2023-10-12 18:24:13,910 epoch 6 - iter 216/723 - loss 0.02690559 - time (sec): 118.71 - samples/sec: 441.44 - lr: 0.000078 - momentum: 0.000000
2023-10-12 18:24:54,865 epoch 6 - iter 288/723 - loss 0.02616528 - time (sec): 159.67 - samples/sec: 440.99 - lr: 0.000077 - momentum: 0.000000
2023-10-12 18:25:34,380 epoch 6 - iter 360/723 - loss 0.02600927 - time (sec): 199.18 - samples/sec: 442.46 - lr: 0.000075 - momentum: 0.000000
2023-10-12 18:26:14,694 epoch 6 - iter 432/723 - loss 0.02342678 - time (sec): 239.50 - samples/sec: 444.95 - lr: 0.000073 - momentum: 0.000000
2023-10-12 18:26:55,637 epoch 6 - iter 504/723 - loss 0.02526975 - time (sec): 280.44 - samples/sec: 443.58 - lr: 0.000072 - momentum: 0.000000
2023-10-12 18:27:35,655 epoch 6 - iter 576/723 - loss 0.02479113 - time (sec): 320.46 - samples/sec: 439.85 - lr: 0.000070 - momentum: 0.000000
2023-10-12 18:28:15,107 epoch 6 - iter 648/723 - loss 0.02441460 - time (sec): 359.91 - samples/sec: 438.05 - lr: 0.000068 - momentum: 0.000000
2023-10-12 18:28:56,224 epoch 6 - iter 720/723 - loss 0.02520439 - time (sec): 401.03 - samples/sec: 438.06 - lr: 0.000067 - momentum: 0.000000
2023-10-12 18:28:57,424 ----------------------------------------------------------------------------------------------------
2023-10-12 18:28:57,425 EPOCH 6 done: loss 0.0252 - lr: 0.000067
2023-10-12 18:29:18,375 DEV : loss 0.08811366558074951 - f1-score (micro avg) 0.8673
2023-10-12 18:29:18,405 saving best model
2023-10-12 18:29:20,978 ----------------------------------------------------------------------------------------------------
2023-10-12 18:30:00,020 epoch 7 - iter 72/723 - loss 0.02375220 - time (sec): 39.04 - samples/sec: 451.86 - lr: 0.000065 - momentum: 0.000000
2023-10-12 18:30:39,390 epoch 7 - iter 144/723 - loss 0.02317751 - time (sec): 78.41 - samples/sec: 453.74 - lr: 0.000063 - momentum: 0.000000
2023-10-12 18:31:17,750 epoch 7 - iter 216/723 - loss 0.02191170 - time (sec): 116.77 - samples/sec: 446.14 - lr: 0.000062 - momentum: 0.000000
2023-10-12 18:31:56,342 epoch 7 - iter 288/723 - loss 0.02214116 - time (sec): 155.36 - samples/sec: 441.87 - lr: 0.000060 - momentum: 0.000000
2023-10-12 18:32:35,819 epoch 7 - iter 360/723 - loss 0.02354976 - time (sec): 194.84 - samples/sec: 445.48 - lr: 0.000058 - momentum: 0.000000
2023-10-12 18:33:15,315 epoch 7 - iter 432/723 - loss 0.02157574 - time (sec): 234.33 - samples/sec: 444.74 - lr: 0.000057 - momentum: 0.000000
2023-10-12 18:33:54,392 epoch 7 - iter 504/723 - loss 0.02134366 - time (sec): 273.41 - samples/sec: 445.97 - lr: 0.000055 - momentum: 0.000000
2023-10-12 18:34:33,498 epoch 7 - iter 576/723 - loss 0.02118785 - time (sec): 312.51 - samples/sec: 444.59 - lr: 0.000053 - momentum: 0.000000
2023-10-12 18:35:12,523 epoch 7 - iter 648/723 - loss 0.02048925 - time (sec): 351.54 - samples/sec: 445.23 - lr: 0.000052 - momentum: 0.000000
2023-10-12 18:35:52,164 epoch 7 - iter 720/723 - loss 0.02077841 - time (sec): 391.18 - samples/sec: 448.61 - lr: 0.000050 - momentum: 0.000000
2023-10-12 18:35:53,549 ----------------------------------------------------------------------------------------------------
2023-10-12 18:35:53,550 EPOCH 7 done: loss 0.0207 - lr: 0.000050
2023-10-12 18:36:15,769 DEV : loss 0.09639902412891388 - f1-score (micro avg) 0.8597
2023-10-12 18:36:15,806 ----------------------------------------------------------------------------------------------------
2023-10-12 18:36:57,015 epoch 8 - iter 72/723 - loss 0.01788093 - time (sec): 41.21 - samples/sec: 450.42 - lr: 0.000048 - momentum: 0.000000
2023-10-12 18:37:36,612 epoch 8 - iter 144/723 - loss 0.01763665 - time (sec): 80.80 - samples/sec: 444.40 - lr: 0.000047 - momentum: 0.000000
2023-10-12 18:38:16,755 epoch 8 - iter 216/723 - loss 0.01670632 - time (sec): 120.95 - samples/sec: 436.75 - lr: 0.000045 - momentum: 0.000000
2023-10-12 18:38:57,813 epoch 8 - iter 288/723 - loss 0.01563682 - time (sec): 162.00 - samples/sec: 444.00 - lr: 0.000043 - momentum: 0.000000
2023-10-12 18:39:37,387 epoch 8 - iter 360/723 - loss 0.01520499 - time (sec): 201.58 - samples/sec: 443.01 - lr: 0.000042 - momentum: 0.000000
2023-10-12 18:40:16,403 epoch 8 - iter 432/723 - loss 0.01554144 - time (sec): 240.59 - samples/sec: 439.58 - lr: 0.000040 - momentum: 0.000000
2023-10-12 18:40:55,597 epoch 8 - iter 504/723 - loss 0.01575127 - time (sec): 279.79 - samples/sec: 439.79 - lr: 0.000038 - momentum: 0.000000
2023-10-12 18:41:34,231 epoch 8 - iter 576/723 - loss 0.01547145 - time (sec): 318.42 - samples/sec: 438.17 - lr: 0.000037 - momentum: 0.000000
2023-10-12 18:42:14,603 epoch 8 - iter 648/723 - loss 0.01719346 - time (sec): 358.79 - samples/sec: 440.09 - lr: 0.000035 - momentum: 0.000000
2023-10-12 18:42:54,038 epoch 8 - iter 720/723 - loss 0.01672576 - time (sec): 398.23 - samples/sec: 441.26 - lr: 0.000033 - momentum: 0.000000
2023-10-12 18:42:55,161 ----------------------------------------------------------------------------------------------------
2023-10-12 18:42:55,161 EPOCH 8 done: loss 0.0167 - lr: 0.000033
2023-10-12 18:43:18,087 DEV : loss 0.10770395398139954 - f1-score (micro avg) 0.855
2023-10-12 18:43:18,120 ----------------------------------------------------------------------------------------------------
2023-10-12 18:43:58,068 epoch 9 - iter 72/723 - loss 0.00592187 - time (sec): 39.95 - samples/sec: 460.70 - lr: 0.000032 - momentum: 0.000000
2023-10-12 18:44:37,416 epoch 9 - iter 144/723 - loss 0.01701284 - time (sec): 79.29 - samples/sec: 469.13 - lr: 0.000030 - momentum: 0.000000
2023-10-12 18:45:16,392 epoch 9 - iter 216/723 - loss 0.01704092 - time (sec): 118.27 - samples/sec: 465.48 - lr: 0.000028 - momentum: 0.000000
2023-10-12 18:45:54,208 epoch 9 - iter 288/723 - loss 0.01622595 - time (sec): 156.09 - samples/sec: 455.77 - lr: 0.000027 - momentum: 0.000000
2023-10-12 18:46:32,450 epoch 9 - iter 360/723 - loss 0.01556455 - time (sec): 194.33 - samples/sec: 447.69 - lr: 0.000025 - momentum: 0.000000
2023-10-12 18:47:11,398 epoch 9 - iter 432/723 - loss 0.01494819 - time (sec): 233.28 - samples/sec: 449.75 - lr: 0.000023 - momentum: 0.000000
2023-10-12 18:47:50,162 epoch 9 - iter 504/723 - loss 0.01504915 - time (sec): 272.04 - samples/sec: 450.43 - lr: 0.000022 - momentum: 0.000000
2023-10-12 18:48:29,909 epoch 9 - iter 576/723 - loss 0.01463468 - time (sec): 311.79 - samples/sec: 453.71 - lr: 0.000020 - momentum: 0.000000
2023-10-12 18:49:08,405 epoch 9 - iter 648/723 - loss 0.01388108 - time (sec): 350.28 - samples/sec: 452.53 - lr: 0.000018 - momentum: 0.000000
2023-10-12 18:49:50,026 epoch 9 - iter 720/723 - loss 0.01368723 - time (sec): 391.90 - samples/sec: 448.25 - lr: 0.000017 - momentum: 0.000000
2023-10-12 18:49:51,219 ----------------------------------------------------------------------------------------------------
2023-10-12 18:49:51,220 EPOCH 9 done: loss 0.0136 - lr: 0.000017
2023-10-12 18:50:13,188 DEV : loss 0.11267418414354324 - f1-score (micro avg) 0.8553
2023-10-12 18:50:13,220 ----------------------------------------------------------------------------------------------------
2023-10-12 18:50:52,654 epoch 10 - iter 72/723 - loss 0.00826221 - time (sec): 39.43 - samples/sec: 456.87 - lr: 0.000015 - momentum: 0.000000
2023-10-12 18:51:29,673 epoch 10 - iter 144/723 - loss 0.00855540 - time (sec): 76.45 - samples/sec: 441.22 - lr: 0.000013 - momentum: 0.000000
2023-10-12 18:52:07,710 epoch 10 - iter 216/723 - loss 0.00947687 - time (sec): 114.49 - samples/sec: 442.76 - lr: 0.000012 - momentum: 0.000000
2023-10-12 18:52:47,939 epoch 10 - iter 288/723 - loss 0.01177807 - time (sec): 154.71 - samples/sec: 446.72 - lr: 0.000010 - momentum: 0.000000
2023-10-12 18:53:27,103 epoch 10 - iter 360/723 - loss 0.01109256 - time (sec): 193.88 - samples/sec: 445.15 - lr: 0.000008 - momentum: 0.000000
2023-10-12 18:54:07,126 epoch 10 - iter 432/723 - loss 0.01009327 - time (sec): 233.90 - samples/sec: 447.66 - lr: 0.000007 - momentum: 0.000000
2023-10-12 18:54:48,429 epoch 10 - iter 504/723 - loss 0.01055010 - time (sec): 275.20 - samples/sec: 447.94 - lr: 0.000005 - momentum: 0.000000
2023-10-12 18:55:28,401 epoch 10 - iter 576/723 - loss 0.01002322 - time (sec): 315.18 - samples/sec: 444.09 - lr: 0.000003 - momentum: 0.000000
2023-10-12 18:56:07,902 epoch 10 - iter 648/723 - loss 0.01138945 - time (sec): 354.68 - samples/sec: 444.59 - lr: 0.000002 - momentum: 0.000000
2023-10-12 18:56:48,588 epoch 10 - iter 720/723 - loss 0.01160600 - time (sec): 395.36 - samples/sec: 444.50 - lr: 0.000000 - momentum: 0.000000
2023-10-12 18:56:49,706 ----------------------------------------------------------------------------------------------------
2023-10-12 18:56:49,707 EPOCH 10 done: loss 0.0116 - lr: 0.000000
2023-10-12 18:57:10,758 DEV : loss 0.11588139832019806 - f1-score (micro avg) 0.8573
2023-10-12 18:57:11,656 ----------------------------------------------------------------------------------------------------
2023-10-12 18:57:11,658 Loading model from best epoch ...
2023-10-12 18:57:15,259 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
2023-10-12 18:57:35,839
Results:
- F-score (micro) 0.858
- F-score (macro) 0.7492
- Accuracy 0.7592
By class:
precision recall f1-score support
PER 0.8376 0.8880 0.8620 482
LOC 0.9273 0.8908 0.9087 458
ORG 0.5082 0.4493 0.4769 69
micro avg 0.8567 0.8593 0.8580 1009
macro avg 0.7577 0.7427 0.7492 1009
weighted avg 0.8558 0.8593 0.8569 1009
2023-10-12 18:57:35,840 ----------------------------------------------------------------------------------------------------
|