File size: 23,991 Bytes
df77d27 |
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 |
2023-10-25 21:02:12,929 ----------------------------------------------------------------------------------------------------
2023-10-25 21:02:12,930 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): BertModel(
(embeddings): BertEmbeddings(
(word_embeddings): Embedding(64001, 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()
)"
2023-10-25 21:02:12,930 ----------------------------------------------------------------------------------------------------
2023-10-25 21:02:12,930 MultiCorpus: 1085 train + 148 dev + 364 test sentences
- NER_HIPE_2022 Corpus: 1085 train + 148 dev + 364 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/sv/with_doc_seperator
2023-10-25 21:02:12,930 ----------------------------------------------------------------------------------------------------
2023-10-25 21:02:12,930 Train: 1085 sentences
2023-10-25 21:02:12,930 (train_with_dev=False, train_with_test=False)
2023-10-25 21:02:12,930 ----------------------------------------------------------------------------------------------------
2023-10-25 21:02:12,931 Training Params:
2023-10-25 21:02:12,931 - learning_rate: "3e-05"
2023-10-25 21:02:12,931 - mini_batch_size: "4"
2023-10-25 21:02:12,931 - max_epochs: "10"
2023-10-25 21:02:12,931 - shuffle: "True"
2023-10-25 21:02:12,931 ----------------------------------------------------------------------------------------------------
2023-10-25 21:02:12,931 Plugins:
2023-10-25 21:02:12,931 - TensorboardLogger
2023-10-25 21:02:12,931 - LinearScheduler | warmup_fraction: '0.1'
2023-10-25 21:02:12,931 ----------------------------------------------------------------------------------------------------
2023-10-25 21:02:12,931 Final evaluation on model from best epoch (best-model.pt)
2023-10-25 21:02:12,931 - metric: "('micro avg', 'f1-score')"
2023-10-25 21:02:12,931 ----------------------------------------------------------------------------------------------------
2023-10-25 21:02:12,931 Computation:
2023-10-25 21:02:12,931 - compute on device: cuda:0
2023-10-25 21:02:12,931 - embedding storage: none
2023-10-25 21:02:12,931 ----------------------------------------------------------------------------------------------------
2023-10-25 21:02:12,931 Model training base path: "hmbench-newseye/sv-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-2"
2023-10-25 21:02:12,931 ----------------------------------------------------------------------------------------------------
2023-10-25 21:02:12,931 ----------------------------------------------------------------------------------------------------
2023-10-25 21:02:12,931 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-25 21:02:14,503 epoch 1 - iter 27/272 - loss 2.87919023 - time (sec): 1.57 - samples/sec: 3275.13 - lr: 0.000003 - momentum: 0.000000
2023-10-25 21:02:16,095 epoch 1 - iter 54/272 - loss 2.21963696 - time (sec): 3.16 - samples/sec: 3380.08 - lr: 0.000006 - momentum: 0.000000
2023-10-25 21:02:17,664 epoch 1 - iter 81/272 - loss 1.64921279 - time (sec): 4.73 - samples/sec: 3353.42 - lr: 0.000009 - momentum: 0.000000
2023-10-25 21:02:19,260 epoch 1 - iter 108/272 - loss 1.33307817 - time (sec): 6.33 - samples/sec: 3425.48 - lr: 0.000012 - momentum: 0.000000
2023-10-25 21:02:20,809 epoch 1 - iter 135/272 - loss 1.15572546 - time (sec): 7.88 - samples/sec: 3388.58 - lr: 0.000015 - momentum: 0.000000
2023-10-25 21:02:22,406 epoch 1 - iter 162/272 - loss 1.01002583 - time (sec): 9.47 - samples/sec: 3368.78 - lr: 0.000018 - momentum: 0.000000
2023-10-25 21:02:23,977 epoch 1 - iter 189/272 - loss 0.89259816 - time (sec): 11.04 - samples/sec: 3388.81 - lr: 0.000021 - momentum: 0.000000
2023-10-25 21:02:25,526 epoch 1 - iter 216/272 - loss 0.81375199 - time (sec): 12.59 - samples/sec: 3374.87 - lr: 0.000024 - momentum: 0.000000
2023-10-25 21:02:26,975 epoch 1 - iter 243/272 - loss 0.75888383 - time (sec): 14.04 - samples/sec: 3348.25 - lr: 0.000027 - momentum: 0.000000
2023-10-25 21:02:28,446 epoch 1 - iter 270/272 - loss 0.70942154 - time (sec): 15.51 - samples/sec: 3341.63 - lr: 0.000030 - momentum: 0.000000
2023-10-25 21:02:28,540 ----------------------------------------------------------------------------------------------------
2023-10-25 21:02:28,541 EPOCH 1 done: loss 0.7075 - lr: 0.000030
2023-10-25 21:02:29,692 DEV : loss 0.15761442482471466 - f1-score (micro avg) 0.6439
2023-10-25 21:02:29,702 saving best model
2023-10-25 21:02:30,216 ----------------------------------------------------------------------------------------------------
2023-10-25 21:02:31,704 epoch 2 - iter 27/272 - loss 0.17595089 - time (sec): 1.49 - samples/sec: 3259.34 - lr: 0.000030 - momentum: 0.000000
2023-10-25 21:02:33,206 epoch 2 - iter 54/272 - loss 0.15919573 - time (sec): 2.99 - samples/sec: 3312.90 - lr: 0.000029 - momentum: 0.000000
2023-10-25 21:02:34,643 epoch 2 - iter 81/272 - loss 0.15943867 - time (sec): 4.42 - samples/sec: 3408.85 - lr: 0.000029 - momentum: 0.000000
2023-10-25 21:02:36,202 epoch 2 - iter 108/272 - loss 0.15319579 - time (sec): 5.98 - samples/sec: 3414.65 - lr: 0.000029 - momentum: 0.000000
2023-10-25 21:02:37,804 epoch 2 - iter 135/272 - loss 0.15029334 - time (sec): 7.59 - samples/sec: 3416.82 - lr: 0.000028 - momentum: 0.000000
2023-10-25 21:02:39,423 epoch 2 - iter 162/272 - loss 0.14192725 - time (sec): 9.20 - samples/sec: 3372.06 - lr: 0.000028 - momentum: 0.000000
2023-10-25 21:02:40,981 epoch 2 - iter 189/272 - loss 0.14097129 - time (sec): 10.76 - samples/sec: 3341.97 - lr: 0.000028 - momentum: 0.000000
2023-10-25 21:02:42,573 epoch 2 - iter 216/272 - loss 0.13630344 - time (sec): 12.36 - samples/sec: 3308.21 - lr: 0.000027 - momentum: 0.000000
2023-10-25 21:02:44,189 epoch 2 - iter 243/272 - loss 0.13343320 - time (sec): 13.97 - samples/sec: 3337.05 - lr: 0.000027 - momentum: 0.000000
2023-10-25 21:02:45,741 epoch 2 - iter 270/272 - loss 0.12992319 - time (sec): 15.52 - samples/sec: 3336.25 - lr: 0.000027 - momentum: 0.000000
2023-10-25 21:02:45,845 ----------------------------------------------------------------------------------------------------
2023-10-25 21:02:45,845 EPOCH 2 done: loss 0.1314 - lr: 0.000027
2023-10-25 21:02:46,993 DEV : loss 0.10909783095121384 - f1-score (micro avg) 0.7844
2023-10-25 21:02:46,999 saving best model
2023-10-25 21:02:47,820 ----------------------------------------------------------------------------------------------------
2023-10-25 21:02:49,170 epoch 3 - iter 27/272 - loss 0.06595269 - time (sec): 1.35 - samples/sec: 3922.55 - lr: 0.000026 - momentum: 0.000000
2023-10-25 21:02:50,566 epoch 3 - iter 54/272 - loss 0.06250690 - time (sec): 2.74 - samples/sec: 3850.02 - lr: 0.000026 - momentum: 0.000000
2023-10-25 21:02:51,996 epoch 3 - iter 81/272 - loss 0.06055673 - time (sec): 4.17 - samples/sec: 3799.59 - lr: 0.000026 - momentum: 0.000000
2023-10-25 21:02:53,490 epoch 3 - iter 108/272 - loss 0.06786572 - time (sec): 5.67 - samples/sec: 3705.33 - lr: 0.000025 - momentum: 0.000000
2023-10-25 21:02:54,961 epoch 3 - iter 135/272 - loss 0.06991093 - time (sec): 7.14 - samples/sec: 3583.88 - lr: 0.000025 - momentum: 0.000000
2023-10-25 21:02:56,499 epoch 3 - iter 162/272 - loss 0.07228279 - time (sec): 8.67 - samples/sec: 3506.91 - lr: 0.000025 - momentum: 0.000000
2023-10-25 21:02:58,031 epoch 3 - iter 189/272 - loss 0.08048214 - time (sec): 10.21 - samples/sec: 3464.11 - lr: 0.000024 - momentum: 0.000000
2023-10-25 21:02:59,619 epoch 3 - iter 216/272 - loss 0.08966991 - time (sec): 11.79 - samples/sec: 3476.65 - lr: 0.000024 - momentum: 0.000000
2023-10-25 21:03:01,102 epoch 3 - iter 243/272 - loss 0.08674602 - time (sec): 13.28 - samples/sec: 3458.55 - lr: 0.000024 - momentum: 0.000000
2023-10-25 21:03:02,581 epoch 3 - iter 270/272 - loss 0.08151681 - time (sec): 14.76 - samples/sec: 3497.60 - lr: 0.000023 - momentum: 0.000000
2023-10-25 21:03:02,705 ----------------------------------------------------------------------------------------------------
2023-10-25 21:03:02,705 EPOCH 3 done: loss 0.0820 - lr: 0.000023
2023-10-25 21:03:03,886 DEV : loss 0.11511397361755371 - f1-score (micro avg) 0.7905
2023-10-25 21:03:03,892 saving best model
2023-10-25 21:03:04,419 ----------------------------------------------------------------------------------------------------
2023-10-25 21:03:05,861 epoch 4 - iter 27/272 - loss 0.05514955 - time (sec): 1.44 - samples/sec: 4091.83 - lr: 0.000023 - momentum: 0.000000
2023-10-25 21:03:07,255 epoch 4 - iter 54/272 - loss 0.05253139 - time (sec): 2.83 - samples/sec: 3707.24 - lr: 0.000023 - momentum: 0.000000
2023-10-25 21:03:08,687 epoch 4 - iter 81/272 - loss 0.04688129 - time (sec): 4.27 - samples/sec: 3592.67 - lr: 0.000022 - momentum: 0.000000
2023-10-25 21:03:10,211 epoch 4 - iter 108/272 - loss 0.04168124 - time (sec): 5.79 - samples/sec: 3627.43 - lr: 0.000022 - momentum: 0.000000
2023-10-25 21:03:11,807 epoch 4 - iter 135/272 - loss 0.05340026 - time (sec): 7.39 - samples/sec: 3558.06 - lr: 0.000022 - momentum: 0.000000
2023-10-25 21:03:13,348 epoch 4 - iter 162/272 - loss 0.05297633 - time (sec): 8.93 - samples/sec: 3458.96 - lr: 0.000021 - momentum: 0.000000
2023-10-25 21:03:14,952 epoch 4 - iter 189/272 - loss 0.05246380 - time (sec): 10.53 - samples/sec: 3446.82 - lr: 0.000021 - momentum: 0.000000
2023-10-25 21:03:16,486 epoch 4 - iter 216/272 - loss 0.05066621 - time (sec): 12.06 - samples/sec: 3403.09 - lr: 0.000021 - momentum: 0.000000
2023-10-25 21:03:18,557 epoch 4 - iter 243/272 - loss 0.05014524 - time (sec): 14.14 - samples/sec: 3304.57 - lr: 0.000020 - momentum: 0.000000
2023-10-25 21:03:20,120 epoch 4 - iter 270/272 - loss 0.04972528 - time (sec): 15.70 - samples/sec: 3281.53 - lr: 0.000020 - momentum: 0.000000
2023-10-25 21:03:20,234 ----------------------------------------------------------------------------------------------------
2023-10-25 21:03:20,235 EPOCH 4 done: loss 0.0499 - lr: 0.000020
2023-10-25 21:03:21,366 DEV : loss 0.11980650573968887 - f1-score (micro avg) 0.8296
2023-10-25 21:03:21,372 saving best model
2023-10-25 21:03:22,161 ----------------------------------------------------------------------------------------------------
2023-10-25 21:03:23,687 epoch 5 - iter 27/272 - loss 0.03676313 - time (sec): 1.52 - samples/sec: 2900.24 - lr: 0.000020 - momentum: 0.000000
2023-10-25 21:03:25,283 epoch 5 - iter 54/272 - loss 0.03268754 - time (sec): 3.12 - samples/sec: 3331.32 - lr: 0.000019 - momentum: 0.000000
2023-10-25 21:03:26,777 epoch 5 - iter 81/272 - loss 0.04226472 - time (sec): 4.61 - samples/sec: 3131.07 - lr: 0.000019 - momentum: 0.000000
2023-10-25 21:03:28,306 epoch 5 - iter 108/272 - loss 0.03782648 - time (sec): 6.14 - samples/sec: 3192.44 - lr: 0.000019 - momentum: 0.000000
2023-10-25 21:03:29,883 epoch 5 - iter 135/272 - loss 0.03598627 - time (sec): 7.72 - samples/sec: 3111.21 - lr: 0.000018 - momentum: 0.000000
2023-10-25 21:03:31,497 epoch 5 - iter 162/272 - loss 0.03756257 - time (sec): 9.33 - samples/sec: 3159.82 - lr: 0.000018 - momentum: 0.000000
2023-10-25 21:03:33,034 epoch 5 - iter 189/272 - loss 0.03380943 - time (sec): 10.87 - samples/sec: 3210.33 - lr: 0.000018 - momentum: 0.000000
2023-10-25 21:03:34,673 epoch 5 - iter 216/272 - loss 0.03483693 - time (sec): 12.51 - samples/sec: 3217.67 - lr: 0.000017 - momentum: 0.000000
2023-10-25 21:03:36,212 epoch 5 - iter 243/272 - loss 0.03499034 - time (sec): 14.05 - samples/sec: 3227.06 - lr: 0.000017 - momentum: 0.000000
2023-10-25 21:03:37,816 epoch 5 - iter 270/272 - loss 0.03406093 - time (sec): 15.65 - samples/sec: 3302.50 - lr: 0.000017 - momentum: 0.000000
2023-10-25 21:03:37,918 ----------------------------------------------------------------------------------------------------
2023-10-25 21:03:37,918 EPOCH 5 done: loss 0.0339 - lr: 0.000017
2023-10-25 21:03:39,053 DEV : loss 0.14052635431289673 - f1-score (micro avg) 0.7978
2023-10-25 21:03:39,061 ----------------------------------------------------------------------------------------------------
2023-10-25 21:03:40,614 epoch 6 - iter 27/272 - loss 0.02057250 - time (sec): 1.55 - samples/sec: 3540.09 - lr: 0.000016 - momentum: 0.000000
2023-10-25 21:03:42,091 epoch 6 - iter 54/272 - loss 0.02284458 - time (sec): 3.03 - samples/sec: 3522.62 - lr: 0.000016 - momentum: 0.000000
2023-10-25 21:03:43,623 epoch 6 - iter 81/272 - loss 0.02523857 - time (sec): 4.56 - samples/sec: 3529.28 - lr: 0.000016 - momentum: 0.000000
2023-10-25 21:03:45,164 epoch 6 - iter 108/272 - loss 0.02438575 - time (sec): 6.10 - samples/sec: 3472.15 - lr: 0.000015 - momentum: 0.000000
2023-10-25 21:03:46,765 epoch 6 - iter 135/272 - loss 0.02417577 - time (sec): 7.70 - samples/sec: 3480.96 - lr: 0.000015 - momentum: 0.000000
2023-10-25 21:03:48,374 epoch 6 - iter 162/272 - loss 0.02230052 - time (sec): 9.31 - samples/sec: 3423.41 - lr: 0.000015 - momentum: 0.000000
2023-10-25 21:03:49,941 epoch 6 - iter 189/272 - loss 0.02171114 - time (sec): 10.88 - samples/sec: 3429.88 - lr: 0.000014 - momentum: 0.000000
2023-10-25 21:03:51,501 epoch 6 - iter 216/272 - loss 0.02180756 - time (sec): 12.44 - samples/sec: 3401.89 - lr: 0.000014 - momentum: 0.000000
2023-10-25 21:03:53,056 epoch 6 - iter 243/272 - loss 0.02467375 - time (sec): 13.99 - samples/sec: 3394.02 - lr: 0.000014 - momentum: 0.000000
2023-10-25 21:03:54,577 epoch 6 - iter 270/272 - loss 0.02404132 - time (sec): 15.51 - samples/sec: 3337.83 - lr: 0.000013 - momentum: 0.000000
2023-10-25 21:03:54,683 ----------------------------------------------------------------------------------------------------
2023-10-25 21:03:54,683 EPOCH 6 done: loss 0.0241 - lr: 0.000013
2023-10-25 21:03:55,907 DEV : loss 0.18079355359077454 - f1-score (micro avg) 0.7837
2023-10-25 21:03:55,916 ----------------------------------------------------------------------------------------------------
2023-10-25 21:03:57,423 epoch 7 - iter 27/272 - loss 0.02323336 - time (sec): 1.51 - samples/sec: 3398.22 - lr: 0.000013 - momentum: 0.000000
2023-10-25 21:03:58,985 epoch 7 - iter 54/272 - loss 0.02182723 - time (sec): 3.07 - samples/sec: 3232.35 - lr: 0.000013 - momentum: 0.000000
2023-10-25 21:04:00,477 epoch 7 - iter 81/272 - loss 0.01847419 - time (sec): 4.56 - samples/sec: 3471.43 - lr: 0.000012 - momentum: 0.000000
2023-10-25 21:04:02,065 epoch 7 - iter 108/272 - loss 0.01720781 - time (sec): 6.15 - samples/sec: 3507.88 - lr: 0.000012 - momentum: 0.000000
2023-10-25 21:04:03,678 epoch 7 - iter 135/272 - loss 0.01776059 - time (sec): 7.76 - samples/sec: 3528.75 - lr: 0.000012 - momentum: 0.000000
2023-10-25 21:04:05,202 epoch 7 - iter 162/272 - loss 0.02107932 - time (sec): 9.28 - samples/sec: 3526.34 - lr: 0.000011 - momentum: 0.000000
2023-10-25 21:04:06,709 epoch 7 - iter 189/272 - loss 0.01920136 - time (sec): 10.79 - samples/sec: 3489.78 - lr: 0.000011 - momentum: 0.000000
2023-10-25 21:04:08,341 epoch 7 - iter 216/272 - loss 0.01899765 - time (sec): 12.42 - samples/sec: 3496.34 - lr: 0.000011 - momentum: 0.000000
2023-10-25 21:04:09,831 epoch 7 - iter 243/272 - loss 0.01879286 - time (sec): 13.91 - samples/sec: 3393.48 - lr: 0.000010 - momentum: 0.000000
2023-10-25 21:04:11,305 epoch 7 - iter 270/272 - loss 0.01843357 - time (sec): 15.39 - samples/sec: 3356.66 - lr: 0.000010 - momentum: 0.000000
2023-10-25 21:04:11,415 ----------------------------------------------------------------------------------------------------
2023-10-25 21:04:11,416 EPOCH 7 done: loss 0.0188 - lr: 0.000010
2023-10-25 21:04:12,682 DEV : loss 0.15969781577587128 - f1-score (micro avg) 0.8305
2023-10-25 21:04:12,689 saving best model
2023-10-25 21:04:13,973 ----------------------------------------------------------------------------------------------------
2023-10-25 21:04:15,512 epoch 8 - iter 27/272 - loss 0.01464099 - time (sec): 1.54 - samples/sec: 3892.95 - lr: 0.000010 - momentum: 0.000000
2023-10-25 21:04:17,057 epoch 8 - iter 54/272 - loss 0.01115447 - time (sec): 3.08 - samples/sec: 3804.70 - lr: 0.000009 - momentum: 0.000000
2023-10-25 21:04:18,556 epoch 8 - iter 81/272 - loss 0.01060002 - time (sec): 4.58 - samples/sec: 3539.69 - lr: 0.000009 - momentum: 0.000000
2023-10-25 21:04:20,061 epoch 8 - iter 108/272 - loss 0.01043073 - time (sec): 6.09 - samples/sec: 3533.46 - lr: 0.000009 - momentum: 0.000000
2023-10-25 21:04:21,636 epoch 8 - iter 135/272 - loss 0.01027025 - time (sec): 7.66 - samples/sec: 3482.52 - lr: 0.000008 - momentum: 0.000000
2023-10-25 21:04:23,132 epoch 8 - iter 162/272 - loss 0.01146835 - time (sec): 9.16 - samples/sec: 3434.44 - lr: 0.000008 - momentum: 0.000000
2023-10-25 21:04:24,628 epoch 8 - iter 189/272 - loss 0.01133446 - time (sec): 10.65 - samples/sec: 3404.02 - lr: 0.000008 - momentum: 0.000000
2023-10-25 21:04:26,193 epoch 8 - iter 216/272 - loss 0.01203709 - time (sec): 12.22 - samples/sec: 3413.26 - lr: 0.000007 - momentum: 0.000000
2023-10-25 21:04:27,820 epoch 8 - iter 243/272 - loss 0.01153724 - time (sec): 13.84 - samples/sec: 3410.96 - lr: 0.000007 - momentum: 0.000000
2023-10-25 21:04:29,331 epoch 8 - iter 270/272 - loss 0.01213084 - time (sec): 15.35 - samples/sec: 3368.92 - lr: 0.000007 - momentum: 0.000000
2023-10-25 21:04:29,438 ----------------------------------------------------------------------------------------------------
2023-10-25 21:04:29,439 EPOCH 8 done: loss 0.0121 - lr: 0.000007
2023-10-25 21:04:30,581 DEV : loss 0.1854364275932312 - f1-score (micro avg) 0.8066
2023-10-25 21:04:30,588 ----------------------------------------------------------------------------------------------------
2023-10-25 21:04:32,228 epoch 9 - iter 27/272 - loss 0.00265158 - time (sec): 1.64 - samples/sec: 3595.47 - lr: 0.000006 - momentum: 0.000000
2023-10-25 21:04:33,785 epoch 9 - iter 54/272 - loss 0.00311819 - time (sec): 3.20 - samples/sec: 3216.61 - lr: 0.000006 - momentum: 0.000000
2023-10-25 21:04:35,405 epoch 9 - iter 81/272 - loss 0.00570425 - time (sec): 4.82 - samples/sec: 3341.78 - lr: 0.000006 - momentum: 0.000000
2023-10-25 21:04:37,015 epoch 9 - iter 108/272 - loss 0.00786022 - time (sec): 6.43 - samples/sec: 3331.93 - lr: 0.000005 - momentum: 0.000000
2023-10-25 21:04:38,648 epoch 9 - iter 135/272 - loss 0.00774158 - time (sec): 8.06 - samples/sec: 3278.98 - lr: 0.000005 - momentum: 0.000000
2023-10-25 21:04:40,250 epoch 9 - iter 162/272 - loss 0.00653485 - time (sec): 9.66 - samples/sec: 3266.43 - lr: 0.000005 - momentum: 0.000000
2023-10-25 21:04:41,736 epoch 9 - iter 189/272 - loss 0.00646242 - time (sec): 11.15 - samples/sec: 3233.93 - lr: 0.000004 - momentum: 0.000000
2023-10-25 21:04:43,262 epoch 9 - iter 216/272 - loss 0.00777496 - time (sec): 12.67 - samples/sec: 3213.16 - lr: 0.000004 - momentum: 0.000000
2023-10-25 21:04:44,834 epoch 9 - iter 243/272 - loss 0.00741950 - time (sec): 14.25 - samples/sec: 3268.88 - lr: 0.000004 - momentum: 0.000000
2023-10-25 21:04:46,387 epoch 9 - iter 270/272 - loss 0.00739841 - time (sec): 15.80 - samples/sec: 3264.98 - lr: 0.000003 - momentum: 0.000000
2023-10-25 21:04:46,502 ----------------------------------------------------------------------------------------------------
2023-10-25 21:04:46,502 EPOCH 9 done: loss 0.0073 - lr: 0.000003
2023-10-25 21:04:47,711 DEV : loss 0.1734946370124817 - f1-score (micro avg) 0.8287
2023-10-25 21:04:47,718 ----------------------------------------------------------------------------------------------------
2023-10-25 21:04:49,223 epoch 10 - iter 27/272 - loss 0.01618621 - time (sec): 1.50 - samples/sec: 2918.63 - lr: 0.000003 - momentum: 0.000000
2023-10-25 21:04:50,710 epoch 10 - iter 54/272 - loss 0.00960812 - time (sec): 2.99 - samples/sec: 3052.34 - lr: 0.000003 - momentum: 0.000000
2023-10-25 21:04:52,202 epoch 10 - iter 81/272 - loss 0.00757130 - time (sec): 4.48 - samples/sec: 3127.09 - lr: 0.000002 - momentum: 0.000000
2023-10-25 21:04:53,765 epoch 10 - iter 108/272 - loss 0.00746314 - time (sec): 6.04 - samples/sec: 3239.68 - lr: 0.000002 - momentum: 0.000000
2023-10-25 21:04:55,313 epoch 10 - iter 135/272 - loss 0.00634969 - time (sec): 7.59 - samples/sec: 3340.14 - lr: 0.000002 - momentum: 0.000000
2023-10-25 21:04:56,852 epoch 10 - iter 162/272 - loss 0.00612372 - time (sec): 9.13 - samples/sec: 3324.67 - lr: 0.000001 - momentum: 0.000000
2023-10-25 21:04:58,389 epoch 10 - iter 189/272 - loss 0.00539305 - time (sec): 10.67 - samples/sec: 3303.91 - lr: 0.000001 - momentum: 0.000000
2023-10-25 21:04:59,958 epoch 10 - iter 216/272 - loss 0.00535593 - time (sec): 12.24 - samples/sec: 3355.96 - lr: 0.000001 - momentum: 0.000000
2023-10-25 21:05:01,476 epoch 10 - iter 243/272 - loss 0.00627154 - time (sec): 13.76 - samples/sec: 3355.88 - lr: 0.000000 - momentum: 0.000000
2023-10-25 21:05:03,068 epoch 10 - iter 270/272 - loss 0.00597542 - time (sec): 15.35 - samples/sec: 3376.78 - lr: 0.000000 - momentum: 0.000000
2023-10-25 21:05:03,174 ----------------------------------------------------------------------------------------------------
2023-10-25 21:05:03,175 EPOCH 10 done: loss 0.0062 - lr: 0.000000
2023-10-25 21:05:04,336 DEV : loss 0.17761798202991486 - f1-score (micro avg) 0.8287
2023-10-25 21:05:04,881 ----------------------------------------------------------------------------------------------------
2023-10-25 21:05:04,882 Loading model from best epoch ...
2023-10-25 21:05:06,915 SequenceTagger predicts: Dictionary with 17 tags: O, S-LOC, B-LOC, E-LOC, I-LOC, S-PER, B-PER, E-PER, I-PER, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd, S-ORG, B-ORG, E-ORG, I-ORG
2023-10-25 21:05:09,305
Results:
- F-score (micro) 0.7896
- F-score (macro) 0.7538
- Accuracy 0.6658
By class:
precision recall f1-score support
LOC 0.8435 0.8462 0.8448 312
PER 0.7102 0.8365 0.7682 208
ORG 0.5273 0.5273 0.5273 55
HumanProd 0.8077 0.9545 0.8750 22
micro avg 0.7637 0.8174 0.7896 597
macro avg 0.7222 0.7911 0.7538 597
weighted avg 0.7666 0.8174 0.7900 597
2023-10-25 21:05:09,305 ----------------------------------------------------------------------------------------------------
|