2023-10-10 18:16:34,319 ---------------------------------------------------------------------------------------------------- 2023-10-10 18:16:34,322 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=17, bias=True) (loss_function): CrossEntropyLoss() )" 2023-10-10 18:16:34,322 ---------------------------------------------------------------------------------------------------- 2023-10-10 18:16:34,322 MultiCorpus: 20847 train + 1123 dev + 3350 test sentences - NER_HIPE_2022 Corpus: 20847 train + 1123 dev + 3350 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/de/with_doc_seperator 2023-10-10 18:16:34,322 ---------------------------------------------------------------------------------------------------- 2023-10-10 18:16:34,322 Train: 20847 sentences 2023-10-10 18:16:34,322 (train_with_dev=False, train_with_test=False) 2023-10-10 18:16:34,322 ---------------------------------------------------------------------------------------------------- 2023-10-10 18:16:34,322 Training Params: 2023-10-10 18:16:34,322 - learning_rate: "0.00015" 2023-10-10 18:16:34,323 - mini_batch_size: "4" 2023-10-10 18:16:34,323 - max_epochs: "10" 2023-10-10 18:16:34,323 - shuffle: "True" 2023-10-10 18:16:34,323 ---------------------------------------------------------------------------------------------------- 2023-10-10 18:16:34,323 Plugins: 2023-10-10 18:16:34,323 - TensorboardLogger 2023-10-10 18:16:34,323 - LinearScheduler | warmup_fraction: '0.1' 2023-10-10 18:16:34,323 ---------------------------------------------------------------------------------------------------- 2023-10-10 18:16:34,323 Final evaluation on model from best epoch (best-model.pt) 2023-10-10 18:16:34,323 - metric: "('micro avg', 'f1-score')" 2023-10-10 18:16:34,323 ---------------------------------------------------------------------------------------------------- 2023-10-10 18:16:34,323 Computation: 2023-10-10 18:16:34,323 - compute on device: cuda:0 2023-10-10 18:16:34,323 - embedding storage: none 2023-10-10 18:16:34,323 ---------------------------------------------------------------------------------------------------- 2023-10-10 18:16:34,324 Model training base path: "hmbench-newseye/de-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs4-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-2" 2023-10-10 18:16:34,324 ---------------------------------------------------------------------------------------------------- 2023-10-10 18:16:34,324 ---------------------------------------------------------------------------------------------------- 2023-10-10 18:16:34,324 Logging anything other than scalars to TensorBoard is currently not supported. 2023-10-10 18:19:07,343 epoch 1 - iter 521/5212 - loss 2.81016252 - time (sec): 153.02 - samples/sec: 234.74 - lr: 0.000015 - momentum: 0.000000 2023-10-10 18:21:43,109 epoch 1 - iter 1042/5212 - loss 2.37850019 - time (sec): 308.78 - samples/sec: 233.59 - lr: 0.000030 - momentum: 0.000000 2023-10-10 18:24:21,241 epoch 1 - iter 1563/5212 - loss 1.84782296 - time (sec): 466.92 - samples/sec: 234.30 - lr: 0.000045 - momentum: 0.000000 2023-10-10 18:27:01,083 epoch 1 - iter 2084/5212 - loss 1.48063193 - time (sec): 626.76 - samples/sec: 237.27 - lr: 0.000060 - momentum: 0.000000 2023-10-10 18:29:38,764 epoch 1 - iter 2605/5212 - loss 1.27253415 - time (sec): 784.44 - samples/sec: 236.54 - lr: 0.000075 - momentum: 0.000000 2023-10-10 18:32:17,153 epoch 1 - iter 3126/5212 - loss 1.11681269 - time (sec): 942.83 - samples/sec: 237.00 - lr: 0.000090 - momentum: 0.000000 2023-10-10 18:34:54,795 epoch 1 - iter 3647/5212 - loss 1.00835432 - time (sec): 1100.47 - samples/sec: 236.27 - lr: 0.000105 - momentum: 0.000000 2023-10-10 18:37:32,394 epoch 1 - iter 4168/5212 - loss 0.91865937 - time (sec): 1258.07 - samples/sec: 234.97 - lr: 0.000120 - momentum: 0.000000 2023-10-10 18:40:00,428 epoch 1 - iter 4689/5212 - loss 0.84845181 - time (sec): 1406.10 - samples/sec: 234.64 - lr: 0.000135 - momentum: 0.000000 2023-10-10 18:42:26,316 epoch 1 - iter 5210/5212 - loss 0.78330933 - time (sec): 1551.99 - samples/sec: 236.66 - lr: 0.000150 - momentum: 0.000000 2023-10-10 18:42:26,799 ---------------------------------------------------------------------------------------------------- 2023-10-10 18:42:26,800 EPOCH 1 done: loss 0.7830 - lr: 0.000150 2023-10-10 18:43:02,191 DEV : loss 0.14865536987781525 - f1-score (micro avg) 0.2591 2023-10-10 18:43:02,244 saving best model 2023-10-10 18:43:03,204 ---------------------------------------------------------------------------------------------------- 2023-10-10 18:45:23,109 epoch 2 - iter 521/5212 - loss 0.18267050 - time (sec): 139.90 - samples/sec: 250.67 - lr: 0.000148 - momentum: 0.000000 2023-10-10 18:47:50,051 epoch 2 - iter 1042/5212 - loss 0.18153459 - time (sec): 286.84 - samples/sec: 247.64 - lr: 0.000147 - momentum: 0.000000 2023-10-10 18:50:09,700 epoch 2 - iter 1563/5212 - loss 0.17717219 - time (sec): 426.49 - samples/sec: 252.57 - lr: 0.000145 - momentum: 0.000000 2023-10-10 18:52:31,360 epoch 2 - iter 2084/5212 - loss 0.17309531 - time (sec): 568.15 - samples/sec: 251.18 - lr: 0.000143 - momentum: 0.000000 2023-10-10 18:54:52,430 epoch 2 - iter 2605/5212 - loss 0.16981704 - time (sec): 709.22 - samples/sec: 253.68 - lr: 0.000142 - momentum: 0.000000 2023-10-10 18:57:12,245 epoch 2 - iter 3126/5212 - loss 0.16298321 - time (sec): 849.04 - samples/sec: 256.48 - lr: 0.000140 - momentum: 0.000000 2023-10-10 18:59:31,681 epoch 2 - iter 3647/5212 - loss 0.16081033 - time (sec): 988.47 - samples/sec: 258.82 - lr: 0.000138 - momentum: 0.000000 2023-10-10 19:01:48,864 epoch 2 - iter 4168/5212 - loss 0.15845681 - time (sec): 1125.66 - samples/sec: 258.75 - lr: 0.000137 - momentum: 0.000000 2023-10-10 19:04:06,270 epoch 2 - iter 4689/5212 - loss 0.15711708 - time (sec): 1263.06 - samples/sec: 258.81 - lr: 0.000135 - momentum: 0.000000 2023-10-10 19:06:27,101 epoch 2 - iter 5210/5212 - loss 0.15390781 - time (sec): 1403.89 - samples/sec: 261.55 - lr: 0.000133 - momentum: 0.000000 2023-10-10 19:06:27,658 ---------------------------------------------------------------------------------------------------- 2023-10-10 19:06:27,658 EPOCH 2 done: loss 0.1539 - lr: 0.000133 2023-10-10 19:07:07,419 DEV : loss 0.12877270579338074 - f1-score (micro avg) 0.3352 2023-10-10 19:07:07,472 saving best model 2023-10-10 19:07:10,082 ---------------------------------------------------------------------------------------------------- 2023-10-10 19:09:26,104 epoch 3 - iter 521/5212 - loss 0.09359735 - time (sec): 136.02 - samples/sec: 262.05 - lr: 0.000132 - momentum: 0.000000 2023-10-10 19:11:43,039 epoch 3 - iter 1042/5212 - loss 0.09901175 - time (sec): 272.95 - samples/sec: 266.45 - lr: 0.000130 - momentum: 0.000000 2023-10-10 19:14:01,365 epoch 3 - iter 1563/5212 - loss 0.10198032 - time (sec): 411.28 - samples/sec: 267.17 - lr: 0.000128 - momentum: 0.000000 2023-10-10 19:16:17,180 epoch 3 - iter 2084/5212 - loss 0.10644730 - time (sec): 547.09 - samples/sec: 266.24 - lr: 0.000127 - momentum: 0.000000 2023-10-10 19:18:36,955 epoch 3 - iter 2605/5212 - loss 0.10979467 - time (sec): 686.87 - samples/sec: 270.23 - lr: 0.000125 - momentum: 0.000000 2023-10-10 19:20:57,796 epoch 3 - iter 3126/5212 - loss 0.10703931 - time (sec): 827.71 - samples/sec: 269.96 - lr: 0.000123 - momentum: 0.000000 2023-10-10 19:23:16,304 epoch 3 - iter 3647/5212 - loss 0.10648081 - time (sec): 966.22 - samples/sec: 269.63 - lr: 0.000122 - momentum: 0.000000 2023-10-10 19:25:35,494 epoch 3 - iter 4168/5212 - loss 0.10571107 - time (sec): 1105.41 - samples/sec: 268.40 - lr: 0.000120 - momentum: 0.000000 2023-10-10 19:27:52,717 epoch 3 - iter 4689/5212 - loss 0.10470171 - time (sec): 1242.63 - samples/sec: 265.47 - lr: 0.000118 - momentum: 0.000000 2023-10-10 19:30:15,588 epoch 3 - iter 5210/5212 - loss 0.10374270 - time (sec): 1385.50 - samples/sec: 265.05 - lr: 0.000117 - momentum: 0.000000 2023-10-10 19:30:16,133 ---------------------------------------------------------------------------------------------------- 2023-10-10 19:30:16,133 EPOCH 3 done: loss 0.1037 - lr: 0.000117 2023-10-10 19:30:54,988 DEV : loss 0.17644649744033813 - f1-score (micro avg) 0.3625 2023-10-10 19:30:55,039 saving best model 2023-10-10 19:30:57,643 ---------------------------------------------------------------------------------------------------- 2023-10-10 19:33:24,242 epoch 4 - iter 521/5212 - loss 0.06203084 - time (sec): 146.59 - samples/sec: 251.99 - lr: 0.000115 - momentum: 0.000000 2023-10-10 19:35:52,103 epoch 4 - iter 1042/5212 - loss 0.07415039 - time (sec): 294.45 - samples/sec: 260.15 - lr: 0.000113 - momentum: 0.000000 2023-10-10 19:38:15,428 epoch 4 - iter 1563/5212 - loss 0.07137797 - time (sec): 437.78 - samples/sec: 257.60 - lr: 0.000112 - momentum: 0.000000 2023-10-10 19:40:37,400 epoch 4 - iter 2084/5212 - loss 0.07121646 - time (sec): 579.75 - samples/sec: 256.34 - lr: 0.000110 - momentum: 0.000000 2023-10-10 19:43:03,829 epoch 4 - iter 2605/5212 - loss 0.06855636 - time (sec): 726.18 - samples/sec: 255.84 - lr: 0.000108 - momentum: 0.000000 2023-10-10 19:45:28,588 epoch 4 - iter 3126/5212 - loss 0.07215692 - time (sec): 870.94 - samples/sec: 255.84 - lr: 0.000107 - momentum: 0.000000 2023-10-10 19:47:53,752 epoch 4 - iter 3647/5212 - loss 0.07245324 - time (sec): 1016.10 - samples/sec: 256.59 - lr: 0.000105 - momentum: 0.000000 2023-10-10 19:50:17,550 epoch 4 - iter 4168/5212 - loss 0.07195821 - time (sec): 1159.90 - samples/sec: 254.60 - lr: 0.000103 - momentum: 0.000000 2023-10-10 19:52:43,827 epoch 4 - iter 4689/5212 - loss 0.07354350 - time (sec): 1306.18 - samples/sec: 254.26 - lr: 0.000102 - momentum: 0.000000 2023-10-10 19:55:06,401 epoch 4 - iter 5210/5212 - loss 0.07389688 - time (sec): 1448.75 - samples/sec: 253.55 - lr: 0.000100 - momentum: 0.000000 2023-10-10 19:55:06,857 ---------------------------------------------------------------------------------------------------- 2023-10-10 19:55:06,857 EPOCH 4 done: loss 0.0739 - lr: 0.000100 2023-10-10 19:55:48,160 DEV : loss 0.3035148084163666 - f1-score (micro avg) 0.3425 2023-10-10 19:55:48,214 ---------------------------------------------------------------------------------------------------- 2023-10-10 19:58:10,368 epoch 5 - iter 521/5212 - loss 0.04380137 - time (sec): 142.15 - samples/sec: 241.95 - lr: 0.000098 - momentum: 0.000000 2023-10-10 20:00:32,233 epoch 5 - iter 1042/5212 - loss 0.04814301 - time (sec): 284.02 - samples/sec: 247.92 - lr: 0.000097 - momentum: 0.000000 2023-10-10 20:02:56,247 epoch 5 - iter 1563/5212 - loss 0.04738308 - time (sec): 428.03 - samples/sec: 254.16 - lr: 0.000095 - momentum: 0.000000 2023-10-10 20:05:18,346 epoch 5 - iter 2084/5212 - loss 0.04616747 - time (sec): 570.13 - samples/sec: 254.93 - lr: 0.000093 - momentum: 0.000000 2023-10-10 20:07:43,212 epoch 5 - iter 2605/5212 - loss 0.04857278 - time (sec): 715.00 - samples/sec: 255.31 - lr: 0.000092 - momentum: 0.000000 2023-10-10 20:10:05,909 epoch 5 - iter 3126/5212 - loss 0.04886862 - time (sec): 857.69 - samples/sec: 255.88 - lr: 0.000090 - momentum: 0.000000 2023-10-10 20:12:25,629 epoch 5 - iter 3647/5212 - loss 0.04941233 - time (sec): 997.41 - samples/sec: 255.51 - lr: 0.000088 - momentum: 0.000000 2023-10-10 20:15:05,698 epoch 5 - iter 4168/5212 - loss 0.04979465 - time (sec): 1157.48 - samples/sec: 253.62 - lr: 0.000087 - momentum: 0.000000 2023-10-10 20:17:41,182 epoch 5 - iter 4689/5212 - loss 0.04912451 - time (sec): 1312.97 - samples/sec: 252.86 - lr: 0.000085 - momentum: 0.000000 2023-10-10 20:20:07,092 epoch 5 - iter 5210/5212 - loss 0.04860995 - time (sec): 1458.88 - samples/sec: 251.66 - lr: 0.000083 - momentum: 0.000000 2023-10-10 20:20:07,750 ---------------------------------------------------------------------------------------------------- 2023-10-10 20:20:07,750 EPOCH 5 done: loss 0.0486 - lr: 0.000083 2023-10-10 20:20:50,848 DEV : loss 0.3323441743850708 - f1-score (micro avg) 0.3643 2023-10-10 20:20:50,907 saving best model 2023-10-10 20:20:53,520 ---------------------------------------------------------------------------------------------------- 2023-10-10 20:23:09,350 epoch 6 - iter 521/5212 - loss 0.03801063 - time (sec): 135.83 - samples/sec: 258.05 - lr: 0.000082 - momentum: 0.000000 2023-10-10 20:25:25,462 epoch 6 - iter 1042/5212 - loss 0.03880207 - time (sec): 271.94 - samples/sec: 257.64 - lr: 0.000080 - momentum: 0.000000 2023-10-10 20:27:44,831 epoch 6 - iter 1563/5212 - loss 0.03989806 - time (sec): 411.31 - samples/sec: 263.80 - lr: 0.000078 - momentum: 0.000000 2023-10-10 20:30:04,726 epoch 6 - iter 2084/5212 - loss 0.04174066 - time (sec): 551.20 - samples/sec: 266.80 - lr: 0.000077 - momentum: 0.000000 2023-10-10 20:32:24,885 epoch 6 - iter 2605/5212 - loss 0.03948956 - time (sec): 691.36 - samples/sec: 269.01 - lr: 0.000075 - momentum: 0.000000 2023-10-10 20:34:43,523 epoch 6 - iter 3126/5212 - loss 0.03830296 - time (sec): 830.00 - samples/sec: 267.05 - lr: 0.000073 - momentum: 0.000000 2023-10-10 20:37:01,312 epoch 6 - iter 3647/5212 - loss 0.03849237 - time (sec): 967.79 - samples/sec: 266.77 - lr: 0.000072 - momentum: 0.000000 2023-10-10 20:39:20,384 epoch 6 - iter 4168/5212 - loss 0.03899918 - time (sec): 1106.86 - samples/sec: 266.65 - lr: 0.000070 - momentum: 0.000000 2023-10-10 20:41:38,051 epoch 6 - iter 4689/5212 - loss 0.03784107 - time (sec): 1244.53 - samples/sec: 264.93 - lr: 0.000068 - momentum: 0.000000 2023-10-10 20:43:57,783 epoch 6 - iter 5210/5212 - loss 0.03843205 - time (sec): 1384.26 - samples/sec: 265.36 - lr: 0.000067 - momentum: 0.000000 2023-10-10 20:43:58,230 ---------------------------------------------------------------------------------------------------- 2023-10-10 20:43:58,230 EPOCH 6 done: loss 0.0384 - lr: 0.000067 2023-10-10 20:44:38,095 DEV : loss 0.3565690815448761 - f1-score (micro avg) 0.3714 2023-10-10 20:44:38,147 saving best model 2023-10-10 20:44:40,782 ---------------------------------------------------------------------------------------------------- 2023-10-10 20:47:08,584 epoch 7 - iter 521/5212 - loss 0.02908204 - time (sec): 147.80 - samples/sec: 240.48 - lr: 0.000065 - momentum: 0.000000 2023-10-10 20:49:40,233 epoch 7 - iter 1042/5212 - loss 0.02884625 - time (sec): 299.45 - samples/sec: 238.89 - lr: 0.000063 - momentum: 0.000000 2023-10-10 20:52:06,698 epoch 7 - iter 1563/5212 - loss 0.02640536 - time (sec): 445.91 - samples/sec: 242.78 - lr: 0.000062 - momentum: 0.000000 2023-10-10 20:54:30,081 epoch 7 - iter 2084/5212 - loss 0.02749239 - time (sec): 589.29 - samples/sec: 245.46 - lr: 0.000060 - momentum: 0.000000 2023-10-10 20:56:53,383 epoch 7 - iter 2605/5212 - loss 0.02821275 - time (sec): 732.59 - samples/sec: 249.90 - lr: 0.000058 - momentum: 0.000000 2023-10-10 20:59:13,941 epoch 7 - iter 3126/5212 - loss 0.02790788 - time (sec): 873.15 - samples/sec: 251.10 - lr: 0.000057 - momentum: 0.000000 2023-10-10 21:01:41,509 epoch 7 - iter 3647/5212 - loss 0.02797348 - time (sec): 1020.72 - samples/sec: 250.96 - lr: 0.000055 - momentum: 0.000000 2023-10-10 21:04:04,892 epoch 7 - iter 4168/5212 - loss 0.02744857 - time (sec): 1164.10 - samples/sec: 249.39 - lr: 0.000053 - momentum: 0.000000 2023-10-10 21:06:29,203 epoch 7 - iter 4689/5212 - loss 0.02707634 - time (sec): 1308.42 - samples/sec: 250.75 - lr: 0.000052 - momentum: 0.000000 2023-10-10 21:08:56,201 epoch 7 - iter 5210/5212 - loss 0.02676611 - time (sec): 1455.41 - samples/sec: 252.36 - lr: 0.000050 - momentum: 0.000000 2023-10-10 21:08:56,684 ---------------------------------------------------------------------------------------------------- 2023-10-10 21:08:56,684 EPOCH 7 done: loss 0.0268 - lr: 0.000050 2023-10-10 21:09:37,357 DEV : loss 0.3953941762447357 - f1-score (micro avg) 0.3828 2023-10-10 21:09:37,410 saving best model 2023-10-10 21:09:40,213 ---------------------------------------------------------------------------------------------------- 2023-10-10 21:12:07,582 epoch 8 - iter 521/5212 - loss 0.01910833 - time (sec): 147.37 - samples/sec: 271.50 - lr: 0.000048 - momentum: 0.000000 2023-10-10 21:14:32,536 epoch 8 - iter 1042/5212 - loss 0.01909368 - time (sec): 292.32 - samples/sec: 264.69 - lr: 0.000047 - momentum: 0.000000 2023-10-10 21:16:51,098 epoch 8 - iter 1563/5212 - loss 0.01709210 - time (sec): 430.88 - samples/sec: 260.58 - lr: 0.000045 - momentum: 0.000000 2023-10-10 21:19:14,800 epoch 8 - iter 2084/5212 - loss 0.01718362 - time (sec): 574.58 - samples/sec: 261.23 - lr: 0.000043 - momentum: 0.000000 2023-10-10 21:21:38,710 epoch 8 - iter 2605/5212 - loss 0.01688276 - time (sec): 718.49 - samples/sec: 258.02 - lr: 0.000042 - momentum: 0.000000 2023-10-10 21:24:04,659 epoch 8 - iter 3126/5212 - loss 0.01713091 - time (sec): 864.44 - samples/sec: 258.08 - lr: 0.000040 - momentum: 0.000000 2023-10-10 21:26:29,475 epoch 8 - iter 3647/5212 - loss 0.01686813 - time (sec): 1009.26 - samples/sec: 256.90 - lr: 0.000038 - momentum: 0.000000 2023-10-10 21:28:50,236 epoch 8 - iter 4168/5212 - loss 0.01687351 - time (sec): 1150.02 - samples/sec: 255.80 - lr: 0.000037 - momentum: 0.000000 2023-10-10 21:31:09,057 epoch 8 - iter 4689/5212 - loss 0.01695639 - time (sec): 1288.84 - samples/sec: 254.16 - lr: 0.000035 - momentum: 0.000000 2023-10-10 21:33:31,826 epoch 8 - iter 5210/5212 - loss 0.01809776 - time (sec): 1431.61 - samples/sec: 256.63 - lr: 0.000033 - momentum: 0.000000 2023-10-10 21:33:32,225 ---------------------------------------------------------------------------------------------------- 2023-10-10 21:33:32,225 EPOCH 8 done: loss 0.0181 - lr: 0.000033 2023-10-10 21:34:11,286 DEV : loss 0.43120914697647095 - f1-score (micro avg) 0.3696 2023-10-10 21:34:11,342 ---------------------------------------------------------------------------------------------------- 2023-10-10 21:36:31,101 epoch 9 - iter 521/5212 - loss 0.01561923 - time (sec): 139.76 - samples/sec: 266.39 - lr: 0.000032 - momentum: 0.000000 2023-10-10 21:38:54,563 epoch 9 - iter 1042/5212 - loss 0.01332527 - time (sec): 283.22 - samples/sec: 273.46 - lr: 0.000030 - momentum: 0.000000 2023-10-10 21:41:13,645 epoch 9 - iter 1563/5212 - loss 0.01209082 - time (sec): 422.30 - samples/sec: 268.95 - lr: 0.000028 - momentum: 0.000000 2023-10-10 21:43:28,845 epoch 9 - iter 2084/5212 - loss 0.01241173 - time (sec): 557.50 - samples/sec: 263.32 - lr: 0.000027 - momentum: 0.000000 2023-10-10 21:45:49,437 epoch 9 - iter 2605/5212 - loss 0.01364082 - time (sec): 698.09 - samples/sec: 265.94 - lr: 0.000025 - momentum: 0.000000 2023-10-10 21:48:05,337 epoch 9 - iter 3126/5212 - loss 0.01364299 - time (sec): 833.99 - samples/sec: 264.53 - lr: 0.000023 - momentum: 0.000000 2023-10-10 21:50:21,661 epoch 9 - iter 3647/5212 - loss 0.01332216 - time (sec): 970.32 - samples/sec: 263.44 - lr: 0.000022 - momentum: 0.000000 2023-10-10 21:52:40,256 epoch 9 - iter 4168/5212 - loss 0.01347305 - time (sec): 1108.91 - samples/sec: 263.38 - lr: 0.000020 - momentum: 0.000000 2023-10-10 21:54:57,896 epoch 9 - iter 4689/5212 - loss 0.01371078 - time (sec): 1246.55 - samples/sec: 263.47 - lr: 0.000018 - momentum: 0.000000 2023-10-10 21:57:17,103 epoch 9 - iter 5210/5212 - loss 0.01386029 - time (sec): 1385.76 - samples/sec: 265.11 - lr: 0.000017 - momentum: 0.000000 2023-10-10 21:57:17,512 ---------------------------------------------------------------------------------------------------- 2023-10-10 21:57:17,513 EPOCH 9 done: loss 0.0139 - lr: 0.000017 2023-10-10 21:57:57,009 DEV : loss 0.4479060769081116 - f1-score (micro avg) 0.3711 2023-10-10 21:57:57,061 ---------------------------------------------------------------------------------------------------- 2023-10-10 22:00:15,416 epoch 10 - iter 521/5212 - loss 0.00629895 - time (sec): 138.35 - samples/sec: 276.50 - lr: 0.000015 - momentum: 0.000000 2023-10-10 22:02:36,634 epoch 10 - iter 1042/5212 - loss 0.00672524 - time (sec): 279.57 - samples/sec: 275.18 - lr: 0.000013 - momentum: 0.000000 2023-10-10 22:04:56,189 epoch 10 - iter 1563/5212 - loss 0.00663359 - time (sec): 419.13 - samples/sec: 269.04 - lr: 0.000012 - momentum: 0.000000 2023-10-10 22:07:12,353 epoch 10 - iter 2084/5212 - loss 0.00656745 - time (sec): 555.29 - samples/sec: 261.90 - lr: 0.000010 - momentum: 0.000000 2023-10-10 22:09:33,505 epoch 10 - iter 2605/5212 - loss 0.00650200 - time (sec): 696.44 - samples/sec: 262.57 - lr: 0.000008 - momentum: 0.000000 2023-10-10 22:11:53,364 epoch 10 - iter 3126/5212 - loss 0.00681067 - time (sec): 836.30 - samples/sec: 261.75 - lr: 0.000007 - momentum: 0.000000 2023-10-10 22:14:13,626 epoch 10 - iter 3647/5212 - loss 0.00705344 - time (sec): 976.56 - samples/sec: 262.86 - lr: 0.000005 - momentum: 0.000000 2023-10-10 22:16:35,112 epoch 10 - iter 4168/5212 - loss 0.00721008 - time (sec): 1118.05 - samples/sec: 264.27 - lr: 0.000003 - momentum: 0.000000 2023-10-10 22:18:53,877 epoch 10 - iter 4689/5212 - loss 0.00709267 - time (sec): 1256.81 - samples/sec: 263.70 - lr: 0.000002 - momentum: 0.000000 2023-10-10 22:21:11,081 epoch 10 - iter 5210/5212 - loss 0.00738418 - time (sec): 1394.02 - samples/sec: 263.35 - lr: 0.000000 - momentum: 0.000000 2023-10-10 22:21:11,710 ---------------------------------------------------------------------------------------------------- 2023-10-10 22:21:11,710 EPOCH 10 done: loss 0.0074 - lr: 0.000000 2023-10-10 22:21:50,043 DEV : loss 0.47310149669647217 - f1-score (micro avg) 0.3745 2023-10-10 22:21:51,032 ---------------------------------------------------------------------------------------------------- 2023-10-10 22:21:51,034 Loading model from best epoch ... 2023-10-10 22:21:56,334 SequenceTagger predicts: Dictionary with 17 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, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd 2023-10-10 22:23:37,448 Results: - F-score (micro) 0.4359 - F-score (macro) 0.2967 - Accuracy 0.2824 By class: precision recall f1-score support LOC 0.4896 0.5066 0.4980 1214 PER 0.4020 0.4567 0.4276 808 ORG 0.2487 0.2748 0.2611 353 HumanProd 0.0000 0.0000 0.0000 15 micro avg 0.4206 0.4523 0.4359 2390 macro avg 0.2851 0.3095 0.2967 2390 weighted avg 0.4213 0.4523 0.4361 2390 2023-10-10 22:23:37,448 ----------------------------------------------------------------------------------------------------