File size: 25,484 Bytes
e2ee494 |
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 |
2023-10-13 10:08:42,076 ----------------------------------------------------------------------------------------------------
2023-10-13 10:08:42,079 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()
)"
2023-10-13 10:08:42,079 ----------------------------------------------------------------------------------------------------
2023-10-13 10:08:42,079 MultiCorpus: 14465 train + 1392 dev + 2432 test sentences
- NER_HIPE_2022 Corpus: 14465 train + 1392 dev + 2432 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/letemps/fr/with_doc_seperator
2023-10-13 10:08:42,079 ----------------------------------------------------------------------------------------------------
2023-10-13 10:08:42,079 Train: 14465 sentences
2023-10-13 10:08:42,079 (train_with_dev=False, train_with_test=False)
2023-10-13 10:08:42,079 ----------------------------------------------------------------------------------------------------
2023-10-13 10:08:42,079 Training Params:
2023-10-13 10:08:42,079 - learning_rate: "0.00015"
2023-10-13 10:08:42,080 - mini_batch_size: "8"
2023-10-13 10:08:42,080 - max_epochs: "10"
2023-10-13 10:08:42,080 - shuffle: "True"
2023-10-13 10:08:42,080 ----------------------------------------------------------------------------------------------------
2023-10-13 10:08:42,080 Plugins:
2023-10-13 10:08:42,080 - TensorboardLogger
2023-10-13 10:08:42,080 - LinearScheduler | warmup_fraction: '0.1'
2023-10-13 10:08:42,080 ----------------------------------------------------------------------------------------------------
2023-10-13 10:08:42,080 Final evaluation on model from best epoch (best-model.pt)
2023-10-13 10:08:42,080 - metric: "('micro avg', 'f1-score')"
2023-10-13 10:08:42,080 ----------------------------------------------------------------------------------------------------
2023-10-13 10:08:42,080 Computation:
2023-10-13 10:08:42,080 - compute on device: cuda:0
2023-10-13 10:08:42,080 - embedding storage: none
2023-10-13 10:08:42,080 ----------------------------------------------------------------------------------------------------
2023-10-13 10:08:42,081 Model training base path: "hmbench-letemps/fr-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs8-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-2"
2023-10-13 10:08:42,081 ----------------------------------------------------------------------------------------------------
2023-10-13 10:08:42,081 ----------------------------------------------------------------------------------------------------
2023-10-13 10:08:42,081 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-13 10:10:16,670 epoch 1 - iter 180/1809 - loss 2.55193452 - time (sec): 94.59 - samples/sec: 393.52 - lr: 0.000015 - momentum: 0.000000
2023-10-13 10:11:50,746 epoch 1 - iter 360/1809 - loss 2.32664425 - time (sec): 188.66 - samples/sec: 395.41 - lr: 0.000030 - momentum: 0.000000
2023-10-13 10:13:30,449 epoch 1 - iter 540/1809 - loss 1.97458754 - time (sec): 288.37 - samples/sec: 391.31 - lr: 0.000045 - momentum: 0.000000
2023-10-13 10:15:09,687 epoch 1 - iter 720/1809 - loss 1.63149756 - time (sec): 387.60 - samples/sec: 388.96 - lr: 0.000060 - momentum: 0.000000
2023-10-13 10:16:47,097 epoch 1 - iter 900/1809 - loss 1.36469605 - time (sec): 485.01 - samples/sec: 389.16 - lr: 0.000075 - momentum: 0.000000
2023-10-13 10:18:23,052 epoch 1 - iter 1080/1809 - loss 1.17974807 - time (sec): 580.97 - samples/sec: 389.57 - lr: 0.000089 - momentum: 0.000000
2023-10-13 10:19:58,384 epoch 1 - iter 1260/1809 - loss 1.03904632 - time (sec): 676.30 - samples/sec: 390.55 - lr: 0.000104 - momentum: 0.000000
2023-10-13 10:21:34,085 epoch 1 - iter 1440/1809 - loss 0.92828376 - time (sec): 772.00 - samples/sec: 391.23 - lr: 0.000119 - momentum: 0.000000
2023-10-13 10:23:11,299 epoch 1 - iter 1620/1809 - loss 0.84195771 - time (sec): 869.22 - samples/sec: 391.84 - lr: 0.000134 - momentum: 0.000000
2023-10-13 10:24:46,248 epoch 1 - iter 1800/1809 - loss 0.77324166 - time (sec): 964.17 - samples/sec: 392.14 - lr: 0.000149 - momentum: 0.000000
2023-10-13 10:24:50,774 ----------------------------------------------------------------------------------------------------
2023-10-13 10:24:50,774 EPOCH 1 done: loss 0.7705 - lr: 0.000149
2023-10-13 10:25:30,426 DEV : loss 0.14874930679798126 - f1-score (micro avg) 0.4029
2023-10-13 10:25:30,486 saving best model
2023-10-13 10:25:31,350 ----------------------------------------------------------------------------------------------------
2023-10-13 10:27:05,720 epoch 2 - iter 180/1809 - loss 0.13805801 - time (sec): 94.37 - samples/sec: 388.33 - lr: 0.000148 - momentum: 0.000000
2023-10-13 10:28:42,834 epoch 2 - iter 360/1809 - loss 0.12933060 - time (sec): 191.48 - samples/sec: 389.43 - lr: 0.000147 - momentum: 0.000000
2023-10-13 10:30:19,877 epoch 2 - iter 540/1809 - loss 0.12513132 - time (sec): 288.52 - samples/sec: 390.84 - lr: 0.000145 - momentum: 0.000000
2023-10-13 10:31:57,990 epoch 2 - iter 720/1809 - loss 0.12053367 - time (sec): 386.64 - samples/sec: 389.16 - lr: 0.000143 - momentum: 0.000000
2023-10-13 10:33:35,527 epoch 2 - iter 900/1809 - loss 0.11629545 - time (sec): 484.17 - samples/sec: 391.21 - lr: 0.000142 - momentum: 0.000000
2023-10-13 10:35:08,984 epoch 2 - iter 1080/1809 - loss 0.11333380 - time (sec): 577.63 - samples/sec: 391.35 - lr: 0.000140 - momentum: 0.000000
2023-10-13 10:36:43,056 epoch 2 - iter 1260/1809 - loss 0.10988685 - time (sec): 671.70 - samples/sec: 391.95 - lr: 0.000138 - momentum: 0.000000
2023-10-13 10:38:22,782 epoch 2 - iter 1440/1809 - loss 0.10663465 - time (sec): 771.43 - samples/sec: 392.15 - lr: 0.000137 - momentum: 0.000000
2023-10-13 10:40:03,308 epoch 2 - iter 1620/1809 - loss 0.10350687 - time (sec): 871.95 - samples/sec: 390.60 - lr: 0.000135 - momentum: 0.000000
2023-10-13 10:41:38,900 epoch 2 - iter 1800/1809 - loss 0.10224515 - time (sec): 967.55 - samples/sec: 390.97 - lr: 0.000133 - momentum: 0.000000
2023-10-13 10:41:43,136 ----------------------------------------------------------------------------------------------------
2023-10-13 10:41:43,137 EPOCH 2 done: loss 0.1022 - lr: 0.000133
2023-10-13 10:42:24,954 DEV : loss 0.09910175204277039 - f1-score (micro avg) 0.5719
2023-10-13 10:42:25,015 saving best model
2023-10-13 10:42:27,591 ----------------------------------------------------------------------------------------------------
2023-10-13 10:44:04,602 epoch 3 - iter 180/1809 - loss 0.06155697 - time (sec): 97.01 - samples/sec: 403.70 - lr: 0.000132 - momentum: 0.000000
2023-10-13 10:45:38,856 epoch 3 - iter 360/1809 - loss 0.06070254 - time (sec): 191.26 - samples/sec: 395.73 - lr: 0.000130 - momentum: 0.000000
2023-10-13 10:47:14,475 epoch 3 - iter 540/1809 - loss 0.06120311 - time (sec): 286.88 - samples/sec: 394.26 - lr: 0.000128 - momentum: 0.000000
2023-10-13 10:48:53,782 epoch 3 - iter 720/1809 - loss 0.06292844 - time (sec): 386.19 - samples/sec: 390.36 - lr: 0.000127 - momentum: 0.000000
2023-10-13 10:50:32,623 epoch 3 - iter 900/1809 - loss 0.06307044 - time (sec): 485.03 - samples/sec: 389.58 - lr: 0.000125 - momentum: 0.000000
2023-10-13 10:52:10,549 epoch 3 - iter 1080/1809 - loss 0.06402904 - time (sec): 582.95 - samples/sec: 386.68 - lr: 0.000123 - momentum: 0.000000
2023-10-13 10:53:49,156 epoch 3 - iter 1260/1809 - loss 0.06377227 - time (sec): 681.56 - samples/sec: 389.15 - lr: 0.000122 - momentum: 0.000000
2023-10-13 10:55:24,344 epoch 3 - iter 1440/1809 - loss 0.06426367 - time (sec): 776.75 - samples/sec: 388.15 - lr: 0.000120 - momentum: 0.000000
2023-10-13 10:57:00,411 epoch 3 - iter 1620/1809 - loss 0.06362259 - time (sec): 872.82 - samples/sec: 389.28 - lr: 0.000118 - momentum: 0.000000
2023-10-13 10:58:38,665 epoch 3 - iter 1800/1809 - loss 0.06327889 - time (sec): 971.07 - samples/sec: 389.10 - lr: 0.000117 - momentum: 0.000000
2023-10-13 10:58:43,368 ----------------------------------------------------------------------------------------------------
2023-10-13 10:58:43,368 EPOCH 3 done: loss 0.0632 - lr: 0.000117
2023-10-13 10:59:24,367 DEV : loss 0.11729110032320023 - f1-score (micro avg) 0.6357
2023-10-13 10:59:24,427 saving best model
2023-10-13 10:59:26,988 ----------------------------------------------------------------------------------------------------
2023-10-13 11:01:01,873 epoch 4 - iter 180/1809 - loss 0.03980255 - time (sec): 94.88 - samples/sec: 390.73 - lr: 0.000115 - momentum: 0.000000
2023-10-13 11:02:41,954 epoch 4 - iter 360/1809 - loss 0.04218853 - time (sec): 194.96 - samples/sec: 389.89 - lr: 0.000113 - momentum: 0.000000
2023-10-13 11:04:22,545 epoch 4 - iter 540/1809 - loss 0.04514616 - time (sec): 295.55 - samples/sec: 382.99 - lr: 0.000112 - momentum: 0.000000
2023-10-13 11:06:00,574 epoch 4 - iter 720/1809 - loss 0.04552944 - time (sec): 393.58 - samples/sec: 382.84 - lr: 0.000110 - momentum: 0.000000
2023-10-13 11:07:37,214 epoch 4 - iter 900/1809 - loss 0.04691891 - time (sec): 490.22 - samples/sec: 384.07 - lr: 0.000108 - momentum: 0.000000
2023-10-13 11:09:17,007 epoch 4 - iter 1080/1809 - loss 0.04597866 - time (sec): 590.01 - samples/sec: 382.76 - lr: 0.000107 - momentum: 0.000000
2023-10-13 11:10:57,169 epoch 4 - iter 1260/1809 - loss 0.04513610 - time (sec): 690.18 - samples/sec: 381.27 - lr: 0.000105 - momentum: 0.000000
2023-10-13 11:12:40,828 epoch 4 - iter 1440/1809 - loss 0.04438682 - time (sec): 793.83 - samples/sec: 379.78 - lr: 0.000103 - momentum: 0.000000
2023-10-13 11:14:18,442 epoch 4 - iter 1620/1809 - loss 0.04429229 - time (sec): 891.45 - samples/sec: 381.80 - lr: 0.000102 - momentum: 0.000000
2023-10-13 11:16:00,362 epoch 4 - iter 1800/1809 - loss 0.04572766 - time (sec): 993.37 - samples/sec: 380.72 - lr: 0.000100 - momentum: 0.000000
2023-10-13 11:16:04,771 ----------------------------------------------------------------------------------------------------
2023-10-13 11:16:04,772 EPOCH 4 done: loss 0.0457 - lr: 0.000100
2023-10-13 11:16:44,751 DEV : loss 0.16882555186748505 - f1-score (micro avg) 0.6361
2023-10-13 11:16:44,823 saving best model
2023-10-13 11:16:47,519 ----------------------------------------------------------------------------------------------------
2023-10-13 11:18:28,087 epoch 5 - iter 180/1809 - loss 0.02745004 - time (sec): 100.56 - samples/sec: 383.98 - lr: 0.000098 - momentum: 0.000000
2023-10-13 11:20:05,038 epoch 5 - iter 360/1809 - loss 0.02948707 - time (sec): 197.51 - samples/sec: 391.81 - lr: 0.000097 - momentum: 0.000000
2023-10-13 11:21:43,338 epoch 5 - iter 540/1809 - loss 0.02991506 - time (sec): 295.81 - samples/sec: 385.31 - lr: 0.000095 - momentum: 0.000000
2023-10-13 11:23:23,915 epoch 5 - iter 720/1809 - loss 0.03207680 - time (sec): 396.39 - samples/sec: 386.48 - lr: 0.000093 - momentum: 0.000000
2023-10-13 11:25:03,777 epoch 5 - iter 900/1809 - loss 0.03173525 - time (sec): 496.25 - samples/sec: 386.05 - lr: 0.000092 - momentum: 0.000000
2023-10-13 11:26:41,708 epoch 5 - iter 1080/1809 - loss 0.03291552 - time (sec): 594.18 - samples/sec: 383.07 - lr: 0.000090 - momentum: 0.000000
2023-10-13 11:28:20,067 epoch 5 - iter 1260/1809 - loss 0.03293994 - time (sec): 692.54 - samples/sec: 383.92 - lr: 0.000088 - momentum: 0.000000
2023-10-13 11:29:59,571 epoch 5 - iter 1440/1809 - loss 0.03244028 - time (sec): 792.05 - samples/sec: 383.50 - lr: 0.000087 - momentum: 0.000000
2023-10-13 11:31:39,767 epoch 5 - iter 1620/1809 - loss 0.03323533 - time (sec): 892.24 - samples/sec: 381.40 - lr: 0.000085 - momentum: 0.000000
2023-10-13 11:33:19,274 epoch 5 - iter 1800/1809 - loss 0.03363279 - time (sec): 991.75 - samples/sec: 381.47 - lr: 0.000083 - momentum: 0.000000
2023-10-13 11:33:23,698 ----------------------------------------------------------------------------------------------------
2023-10-13 11:33:23,698 EPOCH 5 done: loss 0.0337 - lr: 0.000083
2023-10-13 11:34:04,732 DEV : loss 0.22161424160003662 - f1-score (micro avg) 0.6488
2023-10-13 11:34:04,800 saving best model
2023-10-13 11:34:07,393 ----------------------------------------------------------------------------------------------------
2023-10-13 11:35:48,090 epoch 6 - iter 180/1809 - loss 0.01989408 - time (sec): 100.69 - samples/sec: 377.17 - lr: 0.000082 - momentum: 0.000000
2023-10-13 11:37:24,788 epoch 6 - iter 360/1809 - loss 0.02145466 - time (sec): 197.39 - samples/sec: 380.27 - lr: 0.000080 - momentum: 0.000000
2023-10-13 11:39:00,908 epoch 6 - iter 540/1809 - loss 0.02242469 - time (sec): 293.51 - samples/sec: 381.09 - lr: 0.000078 - momentum: 0.000000
2023-10-13 11:40:34,463 epoch 6 - iter 720/1809 - loss 0.02387583 - time (sec): 387.06 - samples/sec: 388.26 - lr: 0.000077 - momentum: 0.000000
2023-10-13 11:42:09,699 epoch 6 - iter 900/1809 - loss 0.02427821 - time (sec): 482.30 - samples/sec: 388.84 - lr: 0.000075 - momentum: 0.000000
2023-10-13 11:43:42,056 epoch 6 - iter 1080/1809 - loss 0.02375360 - time (sec): 574.66 - samples/sec: 391.77 - lr: 0.000073 - momentum: 0.000000
2023-10-13 11:45:16,686 epoch 6 - iter 1260/1809 - loss 0.02379900 - time (sec): 669.29 - samples/sec: 393.07 - lr: 0.000072 - momentum: 0.000000
2023-10-13 11:46:50,964 epoch 6 - iter 1440/1809 - loss 0.02386266 - time (sec): 763.57 - samples/sec: 395.39 - lr: 0.000070 - momentum: 0.000000
2023-10-13 11:48:23,979 epoch 6 - iter 1620/1809 - loss 0.02406347 - time (sec): 856.58 - samples/sec: 396.80 - lr: 0.000068 - momentum: 0.000000
2023-10-13 11:49:57,153 epoch 6 - iter 1800/1809 - loss 0.02485654 - time (sec): 949.75 - samples/sec: 398.12 - lr: 0.000067 - momentum: 0.000000
2023-10-13 11:50:01,456 ----------------------------------------------------------------------------------------------------
2023-10-13 11:50:01,456 EPOCH 6 done: loss 0.0248 - lr: 0.000067
2023-10-13 11:50:42,664 DEV : loss 0.26427823305130005 - f1-score (micro avg) 0.6499
2023-10-13 11:50:42,729 saving best model
2023-10-13 11:50:45,273 ----------------------------------------------------------------------------------------------------
2023-10-13 11:52:24,565 epoch 7 - iter 180/1809 - loss 0.01829401 - time (sec): 99.29 - samples/sec: 388.56 - lr: 0.000065 - momentum: 0.000000
2023-10-13 11:54:00,473 epoch 7 - iter 360/1809 - loss 0.01662527 - time (sec): 195.20 - samples/sec: 390.08 - lr: 0.000063 - momentum: 0.000000
2023-10-13 11:55:35,250 epoch 7 - iter 540/1809 - loss 0.01711330 - time (sec): 289.97 - samples/sec: 396.97 - lr: 0.000062 - momentum: 0.000000
2023-10-13 11:57:12,171 epoch 7 - iter 720/1809 - loss 0.01763868 - time (sec): 386.89 - samples/sec: 391.83 - lr: 0.000060 - momentum: 0.000000
2023-10-13 11:58:52,384 epoch 7 - iter 900/1809 - loss 0.01892285 - time (sec): 487.11 - samples/sec: 388.52 - lr: 0.000058 - momentum: 0.000000
2023-10-13 12:00:33,274 epoch 7 - iter 1080/1809 - loss 0.01907594 - time (sec): 588.00 - samples/sec: 389.53 - lr: 0.000057 - momentum: 0.000000
2023-10-13 12:02:10,151 epoch 7 - iter 1260/1809 - loss 0.01948090 - time (sec): 684.87 - samples/sec: 389.17 - lr: 0.000055 - momentum: 0.000000
2023-10-13 12:03:44,517 epoch 7 - iter 1440/1809 - loss 0.01959196 - time (sec): 779.24 - samples/sec: 389.11 - lr: 0.000053 - momentum: 0.000000
2023-10-13 12:05:18,459 epoch 7 - iter 1620/1809 - loss 0.01974823 - time (sec): 873.18 - samples/sec: 389.17 - lr: 0.000052 - momentum: 0.000000
2023-10-13 12:06:52,523 epoch 7 - iter 1800/1809 - loss 0.01891607 - time (sec): 967.24 - samples/sec: 390.92 - lr: 0.000050 - momentum: 0.000000
2023-10-13 12:06:56,738 ----------------------------------------------------------------------------------------------------
2023-10-13 12:06:56,738 EPOCH 7 done: loss 0.0190 - lr: 0.000050
2023-10-13 12:07:37,764 DEV : loss 0.3006477653980255 - f1-score (micro avg) 0.6484
2023-10-13 12:07:37,830 ----------------------------------------------------------------------------------------------------
2023-10-13 12:09:14,953 epoch 8 - iter 180/1809 - loss 0.01107605 - time (sec): 97.12 - samples/sec: 390.74 - lr: 0.000048 - momentum: 0.000000
2023-10-13 12:10:55,384 epoch 8 - iter 360/1809 - loss 0.01371757 - time (sec): 197.55 - samples/sec: 391.33 - lr: 0.000047 - momentum: 0.000000
2023-10-13 12:12:34,854 epoch 8 - iter 540/1809 - loss 0.01237565 - time (sec): 297.02 - samples/sec: 389.62 - lr: 0.000045 - momentum: 0.000000
2023-10-13 12:14:13,295 epoch 8 - iter 720/1809 - loss 0.01229570 - time (sec): 395.46 - samples/sec: 389.84 - lr: 0.000043 - momentum: 0.000000
2023-10-13 12:15:49,208 epoch 8 - iter 900/1809 - loss 0.01315215 - time (sec): 491.38 - samples/sec: 387.35 - lr: 0.000042 - momentum: 0.000000
2023-10-13 12:17:23,939 epoch 8 - iter 1080/1809 - loss 0.01311433 - time (sec): 586.11 - samples/sec: 391.13 - lr: 0.000040 - momentum: 0.000000
2023-10-13 12:18:59,994 epoch 8 - iter 1260/1809 - loss 0.01292896 - time (sec): 682.16 - samples/sec: 389.79 - lr: 0.000038 - momentum: 0.000000
2023-10-13 12:20:36,705 epoch 8 - iter 1440/1809 - loss 0.01275007 - time (sec): 778.87 - samples/sec: 389.16 - lr: 0.000037 - momentum: 0.000000
2023-10-13 12:22:13,261 epoch 8 - iter 1620/1809 - loss 0.01281415 - time (sec): 875.43 - samples/sec: 389.99 - lr: 0.000035 - momentum: 0.000000
2023-10-13 12:23:53,434 epoch 8 - iter 1800/1809 - loss 0.01361457 - time (sec): 975.60 - samples/sec: 387.96 - lr: 0.000033 - momentum: 0.000000
2023-10-13 12:23:57,584 ----------------------------------------------------------------------------------------------------
2023-10-13 12:23:57,584 EPOCH 8 done: loss 0.0136 - lr: 0.000033
2023-10-13 12:24:39,214 DEV : loss 0.3133712708950043 - f1-score (micro avg) 0.6443
2023-10-13 12:24:39,281 ----------------------------------------------------------------------------------------------------
2023-10-13 12:26:14,295 epoch 9 - iter 180/1809 - loss 0.00811211 - time (sec): 95.01 - samples/sec: 381.60 - lr: 0.000032 - momentum: 0.000000
2023-10-13 12:27:51,642 epoch 9 - iter 360/1809 - loss 0.01014665 - time (sec): 192.36 - samples/sec: 386.96 - lr: 0.000030 - momentum: 0.000000
2023-10-13 12:29:28,757 epoch 9 - iter 540/1809 - loss 0.01204379 - time (sec): 289.47 - samples/sec: 389.86 - lr: 0.000028 - momentum: 0.000000
2023-10-13 12:31:06,089 epoch 9 - iter 720/1809 - loss 0.01210666 - time (sec): 386.81 - samples/sec: 389.63 - lr: 0.000027 - momentum: 0.000000
2023-10-13 12:32:48,351 epoch 9 - iter 900/1809 - loss 0.01191859 - time (sec): 489.07 - samples/sec: 387.06 - lr: 0.000025 - momentum: 0.000000
2023-10-13 12:34:29,374 epoch 9 - iter 1080/1809 - loss 0.01103504 - time (sec): 590.09 - samples/sec: 385.32 - lr: 0.000023 - momentum: 0.000000
2023-10-13 12:36:06,450 epoch 9 - iter 1260/1809 - loss 0.01185981 - time (sec): 687.17 - samples/sec: 384.86 - lr: 0.000022 - momentum: 0.000000
2023-10-13 12:37:44,048 epoch 9 - iter 1440/1809 - loss 0.01164861 - time (sec): 784.76 - samples/sec: 383.08 - lr: 0.000020 - momentum: 0.000000
2023-10-13 12:39:21,038 epoch 9 - iter 1620/1809 - loss 0.01129939 - time (sec): 881.75 - samples/sec: 384.22 - lr: 0.000018 - momentum: 0.000000
2023-10-13 12:41:04,369 epoch 9 - iter 1800/1809 - loss 0.01127319 - time (sec): 985.09 - samples/sec: 383.82 - lr: 0.000017 - momentum: 0.000000
2023-10-13 12:41:09,282 ----------------------------------------------------------------------------------------------------
2023-10-13 12:41:09,283 EPOCH 9 done: loss 0.0112 - lr: 0.000017
2023-10-13 12:41:52,191 DEV : loss 0.33530837297439575 - f1-score (micro avg) 0.6476
2023-10-13 12:41:52,272 ----------------------------------------------------------------------------------------------------
2023-10-13 12:43:32,493 epoch 10 - iter 180/1809 - loss 0.00503398 - time (sec): 100.22 - samples/sec: 380.94 - lr: 0.000015 - momentum: 0.000000
2023-10-13 12:45:10,018 epoch 10 - iter 360/1809 - loss 0.00514473 - time (sec): 197.74 - samples/sec: 383.57 - lr: 0.000013 - momentum: 0.000000
2023-10-13 12:46:46,023 epoch 10 - iter 540/1809 - loss 0.00587847 - time (sec): 293.75 - samples/sec: 385.18 - lr: 0.000012 - momentum: 0.000000
2023-10-13 12:48:21,614 epoch 10 - iter 720/1809 - loss 0.00651019 - time (sec): 389.34 - samples/sec: 389.69 - lr: 0.000010 - momentum: 0.000000
2023-10-13 12:49:57,424 epoch 10 - iter 900/1809 - loss 0.00679671 - time (sec): 485.15 - samples/sec: 389.14 - lr: 0.000008 - momentum: 0.000000
2023-10-13 12:51:33,136 epoch 10 - iter 1080/1809 - loss 0.00712549 - time (sec): 580.86 - samples/sec: 389.66 - lr: 0.000007 - momentum: 0.000000
2023-10-13 12:53:07,935 epoch 10 - iter 1260/1809 - loss 0.00705147 - time (sec): 675.66 - samples/sec: 391.82 - lr: 0.000005 - momentum: 0.000000
2023-10-13 12:54:42,621 epoch 10 - iter 1440/1809 - loss 0.00701041 - time (sec): 770.35 - samples/sec: 394.49 - lr: 0.000003 - momentum: 0.000000
2023-10-13 12:56:16,729 epoch 10 - iter 1620/1809 - loss 0.00703049 - time (sec): 864.45 - samples/sec: 392.77 - lr: 0.000002 - momentum: 0.000000
2023-10-13 12:57:53,169 epoch 10 - iter 1800/1809 - loss 0.00712609 - time (sec): 960.89 - samples/sec: 393.85 - lr: 0.000000 - momentum: 0.000000
2023-10-13 12:57:57,308 ----------------------------------------------------------------------------------------------------
2023-10-13 12:57:57,308 EPOCH 10 done: loss 0.0071 - lr: 0.000000
2023-10-13 12:58:38,220 DEV : loss 0.3420470952987671 - f1-score (micro avg) 0.6541
2023-10-13 12:58:38,294 saving best model
2023-10-13 12:58:45,503 ----------------------------------------------------------------------------------------------------
2023-10-13 12:58:45,506 Loading model from best epoch ...
2023-10-13 12:58:51,051 SequenceTagger predicts: Dictionary with 13 tags: O, S-loc, B-loc, E-loc, I-loc, S-pers, B-pers, E-pers, I-pers, S-org, B-org, E-org, I-org
2023-10-13 12:59:49,682
Results:
- F-score (micro) 0.6291
- F-score (macro) 0.5022
- Accuracy 0.4688
By class:
precision recall f1-score support
loc 0.6234 0.7479 0.6800 591
pers 0.5742 0.6611 0.6146 357
org 0.2642 0.1772 0.2121 79
micro avg 0.5899 0.6738 0.6291 1027
macro avg 0.4873 0.5287 0.5022 1027
weighted avg 0.5787 0.6738 0.6213 1027
2023-10-13 12:59:49,682 ----------------------------------------------------------------------------------------------------
|