File size: 23,829 Bytes
88f09a5 |
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 |
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 ----------------------------------------------------------------------------------------------------
|