2023-10-16 18:54:34,035 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:54:34,035 Model: "SequenceTagger( (embeddings): TransformerWordEmbeddings( (model): BertModel( (embeddings): BertEmbeddings( (word_embeddings): Embedding(32001, 768) (position_embeddings): Embedding(512, 768) (token_type_embeddings): Embedding(2, 768) (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) (encoder): BertEncoder( (layer): ModuleList( (0-11): 12 x BertLayer( (attention): BertAttention( (self): BertSelfAttention( (query): Linear(in_features=768, out_features=768, bias=True) (key): Linear(in_features=768, out_features=768, bias=True) (value): Linear(in_features=768, out_features=768, bias=True) (dropout): Dropout(p=0.1, inplace=False) ) (output): BertSelfOutput( (dense): Linear(in_features=768, out_features=768, bias=True) (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) ) (intermediate): BertIntermediate( (dense): Linear(in_features=768, out_features=3072, bias=True) (intermediate_act_fn): GELUActivation() ) (output): BertOutput( (dense): Linear(in_features=3072, out_features=768, bias=True) (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True) (dropout): Dropout(p=0.1, inplace=False) ) ) ) ) (pooler): BertPooler( (dense): Linear(in_features=768, out_features=768, bias=True) (activation): Tanh() ) ) ) (locked_dropout): LockedDropout(p=0.5) (linear): Linear(in_features=768, out_features=17, bias=True) (loss_function): CrossEntropyLoss() )" 2023-10-16 18:54:34,036 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:54:34,036 MultiCorpus: 1166 train + 165 dev + 415 test sentences - NER_HIPE_2022 Corpus: 1166 train + 165 dev + 415 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/fi/with_doc_seperator 2023-10-16 18:54:34,036 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:54:34,036 Train: 1166 sentences 2023-10-16 18:54:34,036 (train_with_dev=False, train_with_test=False) 2023-10-16 18:54:34,036 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:54:34,036 Training Params: 2023-10-16 18:54:34,036 - learning_rate: "3e-05" 2023-10-16 18:54:34,036 - mini_batch_size: "8" 2023-10-16 18:54:34,036 - max_epochs: "10" 2023-10-16 18:54:34,036 - shuffle: "True" 2023-10-16 18:54:34,036 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:54:34,036 Plugins: 2023-10-16 18:54:34,036 - LinearScheduler | warmup_fraction: '0.1' 2023-10-16 18:54:34,036 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:54:34,036 Final evaluation on model from best epoch (best-model.pt) 2023-10-16 18:54:34,036 - metric: "('micro avg', 'f1-score')" 2023-10-16 18:54:34,036 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:54:34,036 Computation: 2023-10-16 18:54:34,036 - compute on device: cuda:0 2023-10-16 18:54:34,036 - embedding storage: none 2023-10-16 18:54:34,036 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:54:34,036 Model training base path: "hmbench-newseye/fi-dbmdz/bert-base-historic-multilingual-cased-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-5" 2023-10-16 18:54:34,036 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:54:34,036 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:54:35,681 epoch 1 - iter 14/146 - loss 2.89280552 - time (sec): 1.64 - samples/sec: 2610.50 - lr: 0.000003 - momentum: 0.000000 2023-10-16 18:54:37,076 epoch 1 - iter 28/146 - loss 2.73874156 - time (sec): 3.04 - samples/sec: 2888.75 - lr: 0.000006 - momentum: 0.000000 2023-10-16 18:54:38,584 epoch 1 - iter 42/146 - loss 2.29234143 - time (sec): 4.55 - samples/sec: 2965.93 - lr: 0.000008 - momentum: 0.000000 2023-10-16 18:54:39,920 epoch 1 - iter 56/146 - loss 1.88317374 - time (sec): 5.88 - samples/sec: 3007.81 - lr: 0.000011 - momentum: 0.000000 2023-10-16 18:54:41,047 epoch 1 - iter 70/146 - loss 1.66977161 - time (sec): 7.01 - samples/sec: 3025.32 - lr: 0.000014 - momentum: 0.000000 2023-10-16 18:54:42,406 epoch 1 - iter 84/146 - loss 1.48284175 - time (sec): 8.37 - samples/sec: 3068.98 - lr: 0.000017 - momentum: 0.000000 2023-10-16 18:54:43,755 epoch 1 - iter 98/146 - loss 1.33440979 - time (sec): 9.72 - samples/sec: 3101.76 - lr: 0.000020 - momentum: 0.000000 2023-10-16 18:54:45,240 epoch 1 - iter 112/146 - loss 1.22069852 - time (sec): 11.20 - samples/sec: 3069.37 - lr: 0.000023 - momentum: 0.000000 2023-10-16 18:54:46,565 epoch 1 - iter 126/146 - loss 1.12851814 - time (sec): 12.53 - samples/sec: 3056.37 - lr: 0.000026 - momentum: 0.000000 2023-10-16 18:54:47,875 epoch 1 - iter 140/146 - loss 1.04787068 - time (sec): 13.84 - samples/sec: 3059.53 - lr: 0.000029 - momentum: 0.000000 2023-10-16 18:54:48,571 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:54:48,571 EPOCH 1 done: loss 1.0198 - lr: 0.000029 2023-10-16 18:54:49,351 DEV : loss 0.236769899725914 - f1-score (micro avg) 0.4108 2023-10-16 18:54:49,355 saving best model 2023-10-16 18:54:49,712 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:54:51,269 epoch 2 - iter 14/146 - loss 0.27287495 - time (sec): 1.56 - samples/sec: 3346.29 - lr: 0.000030 - momentum: 0.000000 2023-10-16 18:54:52,612 epoch 2 - iter 28/146 - loss 0.23811369 - time (sec): 2.90 - samples/sec: 3279.52 - lr: 0.000029 - momentum: 0.000000 2023-10-16 18:54:54,155 epoch 2 - iter 42/146 - loss 0.23885534 - time (sec): 4.44 - samples/sec: 3039.18 - lr: 0.000029 - momentum: 0.000000 2023-10-16 18:54:55,553 epoch 2 - iter 56/146 - loss 0.24920038 - time (sec): 5.84 - samples/sec: 3078.63 - lr: 0.000029 - momentum: 0.000000 2023-10-16 18:54:57,214 epoch 2 - iter 70/146 - loss 0.25812295 - time (sec): 7.50 - samples/sec: 3058.61 - lr: 0.000028 - momentum: 0.000000 2023-10-16 18:54:58,622 epoch 2 - iter 84/146 - loss 0.26043063 - time (sec): 8.91 - samples/sec: 3077.57 - lr: 0.000028 - momentum: 0.000000 2023-10-16 18:54:59,716 epoch 2 - iter 98/146 - loss 0.25029883 - time (sec): 10.00 - samples/sec: 3104.15 - lr: 0.000028 - momentum: 0.000000 2023-10-16 18:55:00,911 epoch 2 - iter 112/146 - loss 0.24933642 - time (sec): 11.20 - samples/sec: 3086.48 - lr: 0.000027 - momentum: 0.000000 2023-10-16 18:55:02,263 epoch 2 - iter 126/146 - loss 0.23808665 - time (sec): 12.55 - samples/sec: 3113.09 - lr: 0.000027 - momentum: 0.000000 2023-10-16 18:55:03,515 epoch 2 - iter 140/146 - loss 0.23137554 - time (sec): 13.80 - samples/sec: 3103.82 - lr: 0.000027 - momentum: 0.000000 2023-10-16 18:55:04,018 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:55:04,018 EPOCH 2 done: loss 0.2292 - lr: 0.000027 2023-10-16 18:55:05,412 DEV : loss 0.15563829243183136 - f1-score (micro avg) 0.5482 2023-10-16 18:55:05,416 saving best model 2023-10-16 18:55:05,890 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:55:07,247 epoch 3 - iter 14/146 - loss 0.17712828 - time (sec): 1.36 - samples/sec: 2920.16 - lr: 0.000026 - momentum: 0.000000 2023-10-16 18:55:08,550 epoch 3 - iter 28/146 - loss 0.18540699 - time (sec): 2.66 - samples/sec: 3131.81 - lr: 0.000026 - momentum: 0.000000 2023-10-16 18:55:10,053 epoch 3 - iter 42/146 - loss 0.15430173 - time (sec): 4.16 - samples/sec: 3064.67 - lr: 0.000026 - momentum: 0.000000 2023-10-16 18:55:11,470 epoch 3 - iter 56/146 - loss 0.15241278 - time (sec): 5.58 - samples/sec: 3051.84 - lr: 0.000025 - momentum: 0.000000 2023-10-16 18:55:13,005 epoch 3 - iter 70/146 - loss 0.14300126 - time (sec): 7.11 - samples/sec: 3061.86 - lr: 0.000025 - momentum: 0.000000 2023-10-16 18:55:14,469 epoch 3 - iter 84/146 - loss 0.13630069 - time (sec): 8.58 - samples/sec: 3035.39 - lr: 0.000025 - momentum: 0.000000 2023-10-16 18:55:15,955 epoch 3 - iter 98/146 - loss 0.14274446 - time (sec): 10.06 - samples/sec: 2994.63 - lr: 0.000024 - momentum: 0.000000 2023-10-16 18:55:17,298 epoch 3 - iter 112/146 - loss 0.14023351 - time (sec): 11.41 - samples/sec: 2996.09 - lr: 0.000024 - momentum: 0.000000 2023-10-16 18:55:18,689 epoch 3 - iter 126/146 - loss 0.13527969 - time (sec): 12.80 - samples/sec: 3000.38 - lr: 0.000024 - momentum: 0.000000 2023-10-16 18:55:19,901 epoch 3 - iter 140/146 - loss 0.13010389 - time (sec): 14.01 - samples/sec: 3004.41 - lr: 0.000024 - momentum: 0.000000 2023-10-16 18:55:20,702 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:55:20,702 EPOCH 3 done: loss 0.1321 - lr: 0.000024 2023-10-16 18:55:21,960 DEV : loss 0.12500031292438507 - f1-score (micro avg) 0.6379 2023-10-16 18:55:21,965 saving best model 2023-10-16 18:55:22,411 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:55:23,772 epoch 4 - iter 14/146 - loss 0.10475455 - time (sec): 1.36 - samples/sec: 3129.71 - lr: 0.000023 - momentum: 0.000000 2023-10-16 18:55:25,231 epoch 4 - iter 28/146 - loss 0.08547014 - time (sec): 2.82 - samples/sec: 2900.87 - lr: 0.000023 - momentum: 0.000000 2023-10-16 18:55:26,586 epoch 4 - iter 42/146 - loss 0.08510852 - time (sec): 4.17 - samples/sec: 2925.21 - lr: 0.000022 - momentum: 0.000000 2023-10-16 18:55:27,936 epoch 4 - iter 56/146 - loss 0.08462186 - time (sec): 5.52 - samples/sec: 2879.49 - lr: 0.000022 - momentum: 0.000000 2023-10-16 18:55:29,630 epoch 4 - iter 70/146 - loss 0.08581286 - time (sec): 7.22 - samples/sec: 2980.98 - lr: 0.000022 - momentum: 0.000000 2023-10-16 18:55:30,941 epoch 4 - iter 84/146 - loss 0.08601370 - time (sec): 8.53 - samples/sec: 2968.75 - lr: 0.000021 - momentum: 0.000000 2023-10-16 18:55:32,216 epoch 4 - iter 98/146 - loss 0.08409253 - time (sec): 9.80 - samples/sec: 2992.89 - lr: 0.000021 - momentum: 0.000000 2023-10-16 18:55:33,639 epoch 4 - iter 112/146 - loss 0.08729243 - time (sec): 11.23 - samples/sec: 3004.44 - lr: 0.000021 - momentum: 0.000000 2023-10-16 18:55:35,173 epoch 4 - iter 126/146 - loss 0.08748336 - time (sec): 12.76 - samples/sec: 2998.96 - lr: 0.000021 - momentum: 0.000000 2023-10-16 18:55:36,841 epoch 4 - iter 140/146 - loss 0.08579317 - time (sec): 14.43 - samples/sec: 2973.70 - lr: 0.000020 - momentum: 0.000000 2023-10-16 18:55:37,315 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:55:37,315 EPOCH 4 done: loss 0.0855 - lr: 0.000020 2023-10-16 18:55:38,531 DEV : loss 0.1333526223897934 - f1-score (micro avg) 0.6763 2023-10-16 18:55:38,535 saving best model 2023-10-16 18:55:38,990 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:55:40,503 epoch 5 - iter 14/146 - loss 0.06206087 - time (sec): 1.51 - samples/sec: 2656.04 - lr: 0.000020 - momentum: 0.000000 2023-10-16 18:55:41,840 epoch 5 - iter 28/146 - loss 0.05012670 - time (sec): 2.85 - samples/sec: 2861.11 - lr: 0.000019 - momentum: 0.000000 2023-10-16 18:55:43,584 epoch 5 - iter 42/146 - loss 0.06215053 - time (sec): 4.59 - samples/sec: 2766.92 - lr: 0.000019 - momentum: 0.000000 2023-10-16 18:55:44,940 epoch 5 - iter 56/146 - loss 0.05842795 - time (sec): 5.95 - samples/sec: 2912.88 - lr: 0.000019 - momentum: 0.000000 2023-10-16 18:55:46,219 epoch 5 - iter 70/146 - loss 0.06284010 - time (sec): 7.23 - samples/sec: 2977.43 - lr: 0.000018 - momentum: 0.000000 2023-10-16 18:55:47,319 epoch 5 - iter 84/146 - loss 0.06350029 - time (sec): 8.33 - samples/sec: 3032.44 - lr: 0.000018 - momentum: 0.000000 2023-10-16 18:55:48,804 epoch 5 - iter 98/146 - loss 0.06178057 - time (sec): 9.81 - samples/sec: 3048.15 - lr: 0.000018 - momentum: 0.000000 2023-10-16 18:55:50,202 epoch 5 - iter 112/146 - loss 0.05904463 - time (sec): 11.21 - samples/sec: 3045.78 - lr: 0.000018 - momentum: 0.000000 2023-10-16 18:55:51,581 epoch 5 - iter 126/146 - loss 0.05893695 - time (sec): 12.59 - samples/sec: 3040.24 - lr: 0.000017 - momentum: 0.000000 2023-10-16 18:55:53,253 epoch 5 - iter 140/146 - loss 0.05809481 - time (sec): 14.26 - samples/sec: 3008.75 - lr: 0.000017 - momentum: 0.000000 2023-10-16 18:55:53,721 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:55:53,721 EPOCH 5 done: loss 0.0587 - lr: 0.000017 2023-10-16 18:55:55,378 DEV : loss 0.12143861502408981 - f1-score (micro avg) 0.7161 2023-10-16 18:55:55,385 saving best model 2023-10-16 18:55:55,933 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:55:57,461 epoch 6 - iter 14/146 - loss 0.03607615 - time (sec): 1.53 - samples/sec: 2876.46 - lr: 0.000016 - momentum: 0.000000 2023-10-16 18:55:58,944 epoch 6 - iter 28/146 - loss 0.04013210 - time (sec): 3.01 - samples/sec: 2893.84 - lr: 0.000016 - momentum: 0.000000 2023-10-16 18:56:00,407 epoch 6 - iter 42/146 - loss 0.03731182 - time (sec): 4.47 - samples/sec: 2934.51 - lr: 0.000016 - momentum: 0.000000 2023-10-16 18:56:01,782 epoch 6 - iter 56/146 - loss 0.03654630 - time (sec): 5.85 - samples/sec: 2938.20 - lr: 0.000015 - momentum: 0.000000 2023-10-16 18:56:03,614 epoch 6 - iter 70/146 - loss 0.03587700 - time (sec): 7.68 - samples/sec: 2902.37 - lr: 0.000015 - momentum: 0.000000 2023-10-16 18:56:05,056 epoch 6 - iter 84/146 - loss 0.03669627 - time (sec): 9.12 - samples/sec: 2903.38 - lr: 0.000015 - momentum: 0.000000 2023-10-16 18:56:06,226 epoch 6 - iter 98/146 - loss 0.03589081 - time (sec): 10.29 - samples/sec: 2922.68 - lr: 0.000015 - momentum: 0.000000 2023-10-16 18:56:07,783 epoch 6 - iter 112/146 - loss 0.03981007 - time (sec): 11.85 - samples/sec: 2936.13 - lr: 0.000014 - momentum: 0.000000 2023-10-16 18:56:08,944 epoch 6 - iter 126/146 - loss 0.03963558 - time (sec): 13.01 - samples/sec: 2928.91 - lr: 0.000014 - momentum: 0.000000 2023-10-16 18:56:10,418 epoch 6 - iter 140/146 - loss 0.04046337 - time (sec): 14.48 - samples/sec: 2941.68 - lr: 0.000014 - momentum: 0.000000 2023-10-16 18:56:11,044 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:56:11,045 EPOCH 6 done: loss 0.0402 - lr: 0.000014 2023-10-16 18:56:12,269 DEV : loss 0.1350891888141632 - f1-score (micro avg) 0.7328 2023-10-16 18:56:12,274 saving best model 2023-10-16 18:56:12,725 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:56:14,584 epoch 7 - iter 14/146 - loss 0.04059843 - time (sec): 1.85 - samples/sec: 3017.29 - lr: 0.000013 - momentum: 0.000000 2023-10-16 18:56:15,941 epoch 7 - iter 28/146 - loss 0.03439852 - time (sec): 3.21 - samples/sec: 3017.46 - lr: 0.000013 - momentum: 0.000000 2023-10-16 18:56:17,252 epoch 7 - iter 42/146 - loss 0.03464459 - time (sec): 4.52 - samples/sec: 3035.40 - lr: 0.000012 - momentum: 0.000000 2023-10-16 18:56:18,668 epoch 7 - iter 56/146 - loss 0.03616284 - time (sec): 5.94 - samples/sec: 3026.48 - lr: 0.000012 - momentum: 0.000000 2023-10-16 18:56:20,057 epoch 7 - iter 70/146 - loss 0.03702150 - time (sec): 7.33 - samples/sec: 2918.68 - lr: 0.000012 - momentum: 0.000000 2023-10-16 18:56:21,626 epoch 7 - iter 84/146 - loss 0.03595560 - time (sec): 8.90 - samples/sec: 2929.39 - lr: 0.000012 - momentum: 0.000000 2023-10-16 18:56:22,843 epoch 7 - iter 98/146 - loss 0.03386155 - time (sec): 10.11 - samples/sec: 2956.83 - lr: 0.000011 - momentum: 0.000000 2023-10-16 18:56:24,376 epoch 7 - iter 112/146 - loss 0.03576602 - time (sec): 11.65 - samples/sec: 2931.94 - lr: 0.000011 - momentum: 0.000000 2023-10-16 18:56:25,581 epoch 7 - iter 126/146 - loss 0.03449554 - time (sec): 12.85 - samples/sec: 2986.90 - lr: 0.000011 - momentum: 0.000000 2023-10-16 18:56:27,198 epoch 7 - iter 140/146 - loss 0.03287051 - time (sec): 14.47 - samples/sec: 2963.53 - lr: 0.000010 - momentum: 0.000000 2023-10-16 18:56:27,799 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:56:27,799 EPOCH 7 done: loss 0.0323 - lr: 0.000010 2023-10-16 18:56:29,039 DEV : loss 0.1452094465494156 - f1-score (micro avg) 0.7069 2023-10-16 18:56:29,044 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:56:30,420 epoch 8 - iter 14/146 - loss 0.03251634 - time (sec): 1.37 - samples/sec: 3243.02 - lr: 0.000010 - momentum: 0.000000 2023-10-16 18:56:31,782 epoch 8 - iter 28/146 - loss 0.02633798 - time (sec): 2.74 - samples/sec: 3119.42 - lr: 0.000009 - momentum: 0.000000 2023-10-16 18:56:33,117 epoch 8 - iter 42/146 - loss 0.02260259 - time (sec): 4.07 - samples/sec: 2996.68 - lr: 0.000009 - momentum: 0.000000 2023-10-16 18:56:34,622 epoch 8 - iter 56/146 - loss 0.02145170 - time (sec): 5.58 - samples/sec: 3035.64 - lr: 0.000009 - momentum: 0.000000 2023-10-16 18:56:36,008 epoch 8 - iter 70/146 - loss 0.02157167 - time (sec): 6.96 - samples/sec: 2995.86 - lr: 0.000009 - momentum: 0.000000 2023-10-16 18:56:37,531 epoch 8 - iter 84/146 - loss 0.02074857 - time (sec): 8.49 - samples/sec: 3027.40 - lr: 0.000008 - momentum: 0.000000 2023-10-16 18:56:39,133 epoch 8 - iter 98/146 - loss 0.02173471 - time (sec): 10.09 - samples/sec: 2933.21 - lr: 0.000008 - momentum: 0.000000 2023-10-16 18:56:40,348 epoch 8 - iter 112/146 - loss 0.02670667 - time (sec): 11.30 - samples/sec: 2958.94 - lr: 0.000008 - momentum: 0.000000 2023-10-16 18:56:42,102 epoch 8 - iter 126/146 - loss 0.02625262 - time (sec): 13.06 - samples/sec: 2921.14 - lr: 0.000007 - momentum: 0.000000 2023-10-16 18:56:43,519 epoch 8 - iter 140/146 - loss 0.02732306 - time (sec): 14.47 - samples/sec: 2943.76 - lr: 0.000007 - momentum: 0.000000 2023-10-16 18:56:44,116 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:56:44,116 EPOCH 8 done: loss 0.0267 - lr: 0.000007 2023-10-16 18:56:45,516 DEV : loss 0.15337024629116058 - f1-score (micro avg) 0.7076 2023-10-16 18:56:45,521 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:56:47,100 epoch 9 - iter 14/146 - loss 0.00916726 - time (sec): 1.58 - samples/sec: 3041.55 - lr: 0.000006 - momentum: 0.000000 2023-10-16 18:56:48,428 epoch 9 - iter 28/146 - loss 0.00863155 - time (sec): 2.91 - samples/sec: 3022.22 - lr: 0.000006 - momentum: 0.000000 2023-10-16 18:56:49,833 epoch 9 - iter 42/146 - loss 0.01507339 - time (sec): 4.31 - samples/sec: 3022.94 - lr: 0.000006 - momentum: 0.000000 2023-10-16 18:56:51,260 epoch 9 - iter 56/146 - loss 0.01738903 - time (sec): 5.74 - samples/sec: 3053.70 - lr: 0.000006 - momentum: 0.000000 2023-10-16 18:56:52,983 epoch 9 - iter 70/146 - loss 0.01812648 - time (sec): 7.46 - samples/sec: 3057.65 - lr: 0.000005 - momentum: 0.000000 2023-10-16 18:56:54,276 epoch 9 - iter 84/146 - loss 0.01855582 - time (sec): 8.75 - samples/sec: 3027.00 - lr: 0.000005 - momentum: 0.000000 2023-10-16 18:56:55,586 epoch 9 - iter 98/146 - loss 0.01935023 - time (sec): 10.06 - samples/sec: 2994.77 - lr: 0.000005 - momentum: 0.000000 2023-10-16 18:56:57,016 epoch 9 - iter 112/146 - loss 0.01757770 - time (sec): 11.49 - samples/sec: 3000.50 - lr: 0.000004 - momentum: 0.000000 2023-10-16 18:56:58,360 epoch 9 - iter 126/146 - loss 0.01929319 - time (sec): 12.84 - samples/sec: 3006.26 - lr: 0.000004 - momentum: 0.000000 2023-10-16 18:56:59,890 epoch 9 - iter 140/146 - loss 0.02150382 - time (sec): 14.37 - samples/sec: 2987.86 - lr: 0.000004 - momentum: 0.000000 2023-10-16 18:57:00,424 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:57:00,424 EPOCH 9 done: loss 0.0221 - lr: 0.000004 2023-10-16 18:57:01,640 DEV : loss 0.1633269190788269 - f1-score (micro avg) 0.7034 2023-10-16 18:57:01,644 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:57:03,014 epoch 10 - iter 14/146 - loss 0.00942726 - time (sec): 1.37 - samples/sec: 2886.55 - lr: 0.000003 - momentum: 0.000000 2023-10-16 18:57:04,499 epoch 10 - iter 28/146 - loss 0.01186661 - time (sec): 2.85 - samples/sec: 2953.45 - lr: 0.000003 - momentum: 0.000000 2023-10-16 18:57:06,043 epoch 10 - iter 42/146 - loss 0.01639072 - time (sec): 4.40 - samples/sec: 2945.84 - lr: 0.000003 - momentum: 0.000000 2023-10-16 18:57:07,550 epoch 10 - iter 56/146 - loss 0.01613294 - time (sec): 5.91 - samples/sec: 3068.01 - lr: 0.000002 - momentum: 0.000000 2023-10-16 18:57:09,143 epoch 10 - iter 70/146 - loss 0.01704983 - time (sec): 7.50 - samples/sec: 3022.87 - lr: 0.000002 - momentum: 0.000000 2023-10-16 18:57:10,531 epoch 10 - iter 84/146 - loss 0.01632056 - time (sec): 8.89 - samples/sec: 3029.73 - lr: 0.000002 - momentum: 0.000000 2023-10-16 18:57:11,894 epoch 10 - iter 98/146 - loss 0.01654833 - time (sec): 10.25 - samples/sec: 2976.32 - lr: 0.000001 - momentum: 0.000000 2023-10-16 18:57:13,276 epoch 10 - iter 112/146 - loss 0.01649142 - time (sec): 11.63 - samples/sec: 2988.32 - lr: 0.000001 - momentum: 0.000000 2023-10-16 18:57:14,483 epoch 10 - iter 126/146 - loss 0.01757757 - time (sec): 12.84 - samples/sec: 3008.33 - lr: 0.000001 - momentum: 0.000000 2023-10-16 18:57:15,888 epoch 10 - iter 140/146 - loss 0.01985176 - time (sec): 14.24 - samples/sec: 3007.60 - lr: 0.000000 - momentum: 0.000000 2023-10-16 18:57:16,400 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:57:16,400 EPOCH 10 done: loss 0.0195 - lr: 0.000000 2023-10-16 18:57:17,645 DEV : loss 0.16719266772270203 - f1-score (micro avg) 0.7106 2023-10-16 18:57:18,010 ---------------------------------------------------------------------------------------------------- 2023-10-16 18:57:18,011 Loading model from best epoch ... 2023-10-16 18:57:19,453 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 2023-10-16 18:57:21,781 Results: - F-score (micro) 0.751 - F-score (macro) 0.685 - Accuracy 0.6206 By class: precision recall f1-score support PER 0.8106 0.8362 0.8232 348 LOC 0.6495 0.8238 0.7264 261 ORG 0.3774 0.3846 0.3810 52 HumanProd 0.8500 0.7727 0.8095 22 micro avg 0.7117 0.7950 0.7510 683 macro avg 0.6719 0.7043 0.6850 683 weighted avg 0.7173 0.7950 0.7521 683 2023-10-16 18:57:21,781 ----------------------------------------------------------------------------------------------------