File size: 23,925 Bytes
6b3c4d2 |
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 |
2023-10-17 19:51:52,385 ----------------------------------------------------------------------------------------------------
2023-10-17 19:51:52,386 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): ElectraModel(
(embeddings): ElectraEmbeddings(
(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): ElectraEncoder(
(layer): ModuleList(
(0-11): 12 x ElectraLayer(
(attention): ElectraAttention(
(self): ElectraSelfAttention(
(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): ElectraSelfOutput(
(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): ElectraIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): ElectraOutput(
(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)
)
)
)
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=768, out_features=17, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-17 19:51:52,386 ----------------------------------------------------------------------------------------------------
2023-10-17 19:51:52,386 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
2023-10-17 19:51:52,386 ----------------------------------------------------------------------------------------------------
2023-10-17 19:51:52,386 Train: 1085 sentences
2023-10-17 19:51:52,386 (train_with_dev=False, train_with_test=False)
2023-10-17 19:51:52,386 ----------------------------------------------------------------------------------------------------
2023-10-17 19:51:52,386 Training Params:
2023-10-17 19:51:52,386 - learning_rate: "3e-05"
2023-10-17 19:51:52,386 - mini_batch_size: "4"
2023-10-17 19:51:52,386 - max_epochs: "10"
2023-10-17 19:51:52,386 - shuffle: "True"
2023-10-17 19:51:52,386 ----------------------------------------------------------------------------------------------------
2023-10-17 19:51:52,386 Plugins:
2023-10-17 19:51:52,386 - TensorboardLogger
2023-10-17 19:51:52,386 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 19:51:52,386 ----------------------------------------------------------------------------------------------------
2023-10-17 19:51:52,386 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 19:51:52,386 - metric: "('micro avg', 'f1-score')"
2023-10-17 19:51:52,386 ----------------------------------------------------------------------------------------------------
2023-10-17 19:51:52,386 Computation:
2023-10-17 19:51:52,386 - compute on device: cuda:0
2023-10-17 19:51:52,386 - embedding storage: none
2023-10-17 19:51:52,387 ----------------------------------------------------------------------------------------------------
2023-10-17 19:51:52,387 Model training base path: "hmbench-newseye/sv-hmteams/teams-base-historic-multilingual-discriminator-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-2"
2023-10-17 19:51:52,387 ----------------------------------------------------------------------------------------------------
2023-10-17 19:51:52,387 ----------------------------------------------------------------------------------------------------
2023-10-17 19:51:52,387 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 19:51:53,991 epoch 1 - iter 27/272 - loss 3.68614455 - time (sec): 1.60 - samples/sec: 3013.94 - lr: 0.000003 - momentum: 0.000000
2023-10-17 19:51:55,492 epoch 1 - iter 54/272 - loss 3.27911665 - time (sec): 3.10 - samples/sec: 2936.58 - lr: 0.000006 - momentum: 0.000000
2023-10-17 19:51:57,102 epoch 1 - iter 81/272 - loss 2.42782165 - time (sec): 4.71 - samples/sec: 3172.39 - lr: 0.000009 - momentum: 0.000000
2023-10-17 19:51:58,619 epoch 1 - iter 108/272 - loss 1.93547446 - time (sec): 6.23 - samples/sec: 3222.39 - lr: 0.000012 - momentum: 0.000000
2023-10-17 19:52:00,174 epoch 1 - iter 135/272 - loss 1.66887966 - time (sec): 7.79 - samples/sec: 3167.82 - lr: 0.000015 - momentum: 0.000000
2023-10-17 19:52:01,759 epoch 1 - iter 162/272 - loss 1.43150079 - time (sec): 9.37 - samples/sec: 3235.06 - lr: 0.000018 - momentum: 0.000000
2023-10-17 19:52:03,295 epoch 1 - iter 189/272 - loss 1.27708928 - time (sec): 10.91 - samples/sec: 3248.58 - lr: 0.000021 - momentum: 0.000000
2023-10-17 19:52:05,064 epoch 1 - iter 216/272 - loss 1.11895645 - time (sec): 12.68 - samples/sec: 3302.85 - lr: 0.000024 - momentum: 0.000000
2023-10-17 19:52:06,572 epoch 1 - iter 243/272 - loss 1.03839737 - time (sec): 14.18 - samples/sec: 3295.24 - lr: 0.000027 - momentum: 0.000000
2023-10-17 19:52:08,173 epoch 1 - iter 270/272 - loss 0.95699653 - time (sec): 15.78 - samples/sec: 3280.80 - lr: 0.000030 - momentum: 0.000000
2023-10-17 19:52:08,284 ----------------------------------------------------------------------------------------------------
2023-10-17 19:52:08,285 EPOCH 1 done: loss 0.9543 - lr: 0.000030
2023-10-17 19:52:09,325 DEV : loss 0.17717474699020386 - f1-score (micro avg) 0.5914
2023-10-17 19:52:09,329 saving best model
2023-10-17 19:52:09,689 ----------------------------------------------------------------------------------------------------
2023-10-17 19:52:11,253 epoch 2 - iter 27/272 - loss 0.16203503 - time (sec): 1.56 - samples/sec: 3304.40 - lr: 0.000030 - momentum: 0.000000
2023-10-17 19:52:12,809 epoch 2 - iter 54/272 - loss 0.17472666 - time (sec): 3.12 - samples/sec: 3375.93 - lr: 0.000029 - momentum: 0.000000
2023-10-17 19:52:14,492 epoch 2 - iter 81/272 - loss 0.18110744 - time (sec): 4.80 - samples/sec: 3347.54 - lr: 0.000029 - momentum: 0.000000
2023-10-17 19:52:16,126 epoch 2 - iter 108/272 - loss 0.17792809 - time (sec): 6.44 - samples/sec: 3312.91 - lr: 0.000029 - momentum: 0.000000
2023-10-17 19:52:17,559 epoch 2 - iter 135/272 - loss 0.17288065 - time (sec): 7.87 - samples/sec: 3279.42 - lr: 0.000028 - momentum: 0.000000
2023-10-17 19:52:19,203 epoch 2 - iter 162/272 - loss 0.18156776 - time (sec): 9.51 - samples/sec: 3278.49 - lr: 0.000028 - momentum: 0.000000
2023-10-17 19:52:20,692 epoch 2 - iter 189/272 - loss 0.17360395 - time (sec): 11.00 - samples/sec: 3239.67 - lr: 0.000028 - momentum: 0.000000
2023-10-17 19:52:22,312 epoch 2 - iter 216/272 - loss 0.16404397 - time (sec): 12.62 - samples/sec: 3275.19 - lr: 0.000027 - momentum: 0.000000
2023-10-17 19:52:23,951 epoch 2 - iter 243/272 - loss 0.15732808 - time (sec): 14.26 - samples/sec: 3265.49 - lr: 0.000027 - momentum: 0.000000
2023-10-17 19:52:25,422 epoch 2 - iter 270/272 - loss 0.15548570 - time (sec): 15.73 - samples/sec: 3287.26 - lr: 0.000027 - momentum: 0.000000
2023-10-17 19:52:25,512 ----------------------------------------------------------------------------------------------------
2023-10-17 19:52:25,513 EPOCH 2 done: loss 0.1553 - lr: 0.000027
2023-10-17 19:52:26,935 DEV : loss 0.1165740042924881 - f1-score (micro avg) 0.7687
2023-10-17 19:52:26,939 saving best model
2023-10-17 19:52:27,426 ----------------------------------------------------------------------------------------------------
2023-10-17 19:52:29,141 epoch 3 - iter 27/272 - loss 0.07582860 - time (sec): 1.71 - samples/sec: 3187.78 - lr: 0.000026 - momentum: 0.000000
2023-10-17 19:52:30,790 epoch 3 - iter 54/272 - loss 0.08158520 - time (sec): 3.36 - samples/sec: 3334.75 - lr: 0.000026 - momentum: 0.000000
2023-10-17 19:52:32,345 epoch 3 - iter 81/272 - loss 0.08034283 - time (sec): 4.92 - samples/sec: 3356.99 - lr: 0.000026 - momentum: 0.000000
2023-10-17 19:52:33,932 epoch 3 - iter 108/272 - loss 0.08966396 - time (sec): 6.50 - samples/sec: 3372.91 - lr: 0.000025 - momentum: 0.000000
2023-10-17 19:52:35,546 epoch 3 - iter 135/272 - loss 0.08284123 - time (sec): 8.12 - samples/sec: 3357.93 - lr: 0.000025 - momentum: 0.000000
2023-10-17 19:52:37,046 epoch 3 - iter 162/272 - loss 0.08763320 - time (sec): 9.62 - samples/sec: 3337.11 - lr: 0.000025 - momentum: 0.000000
2023-10-17 19:52:38,616 epoch 3 - iter 189/272 - loss 0.08674857 - time (sec): 11.19 - samples/sec: 3323.91 - lr: 0.000024 - momentum: 0.000000
2023-10-17 19:52:40,057 epoch 3 - iter 216/272 - loss 0.08622983 - time (sec): 12.63 - samples/sec: 3291.27 - lr: 0.000024 - momentum: 0.000000
2023-10-17 19:52:41,676 epoch 3 - iter 243/272 - loss 0.08383335 - time (sec): 14.25 - samples/sec: 3304.56 - lr: 0.000024 - momentum: 0.000000
2023-10-17 19:52:43,140 epoch 3 - iter 270/272 - loss 0.08380410 - time (sec): 15.71 - samples/sec: 3298.30 - lr: 0.000023 - momentum: 0.000000
2023-10-17 19:52:43,229 ----------------------------------------------------------------------------------------------------
2023-10-17 19:52:43,229 EPOCH 3 done: loss 0.0836 - lr: 0.000023
2023-10-17 19:52:44,666 DEV : loss 0.11597966402769089 - f1-score (micro avg) 0.7549
2023-10-17 19:52:44,671 ----------------------------------------------------------------------------------------------------
2023-10-17 19:52:46,238 epoch 4 - iter 27/272 - loss 0.03817143 - time (sec): 1.57 - samples/sec: 3114.22 - lr: 0.000023 - momentum: 0.000000
2023-10-17 19:52:47,801 epoch 4 - iter 54/272 - loss 0.03346733 - time (sec): 3.13 - samples/sec: 3175.76 - lr: 0.000023 - momentum: 0.000000
2023-10-17 19:52:49,482 epoch 4 - iter 81/272 - loss 0.04517407 - time (sec): 4.81 - samples/sec: 3294.02 - lr: 0.000022 - momentum: 0.000000
2023-10-17 19:52:50,892 epoch 4 - iter 108/272 - loss 0.04319358 - time (sec): 6.22 - samples/sec: 3243.83 - lr: 0.000022 - momentum: 0.000000
2023-10-17 19:52:52,431 epoch 4 - iter 135/272 - loss 0.04534533 - time (sec): 7.76 - samples/sec: 3249.82 - lr: 0.000022 - momentum: 0.000000
2023-10-17 19:52:54,112 epoch 4 - iter 162/272 - loss 0.04889301 - time (sec): 9.44 - samples/sec: 3285.66 - lr: 0.000021 - momentum: 0.000000
2023-10-17 19:52:55,709 epoch 4 - iter 189/272 - loss 0.05064160 - time (sec): 11.04 - samples/sec: 3265.73 - lr: 0.000021 - momentum: 0.000000
2023-10-17 19:52:57,341 epoch 4 - iter 216/272 - loss 0.05215477 - time (sec): 12.67 - samples/sec: 3256.61 - lr: 0.000021 - momentum: 0.000000
2023-10-17 19:52:58,816 epoch 4 - iter 243/272 - loss 0.05121743 - time (sec): 14.14 - samples/sec: 3280.18 - lr: 0.000020 - momentum: 0.000000
2023-10-17 19:53:00,346 epoch 4 - iter 270/272 - loss 0.05390532 - time (sec): 15.67 - samples/sec: 3301.63 - lr: 0.000020 - momentum: 0.000000
2023-10-17 19:53:00,433 ----------------------------------------------------------------------------------------------------
2023-10-17 19:53:00,434 EPOCH 4 done: loss 0.0540 - lr: 0.000020
2023-10-17 19:53:01,866 DEV : loss 0.1187073215842247 - f1-score (micro avg) 0.7993
2023-10-17 19:53:01,870 saving best model
2023-10-17 19:53:02,345 ----------------------------------------------------------------------------------------------------
2023-10-17 19:53:03,909 epoch 5 - iter 27/272 - loss 0.04055481 - time (sec): 1.56 - samples/sec: 3484.10 - lr: 0.000020 - momentum: 0.000000
2023-10-17 19:53:05,449 epoch 5 - iter 54/272 - loss 0.04184600 - time (sec): 3.10 - samples/sec: 3498.71 - lr: 0.000019 - momentum: 0.000000
2023-10-17 19:53:06,981 epoch 5 - iter 81/272 - loss 0.03728394 - time (sec): 4.63 - samples/sec: 3467.31 - lr: 0.000019 - momentum: 0.000000
2023-10-17 19:53:08,584 epoch 5 - iter 108/272 - loss 0.03770208 - time (sec): 6.23 - samples/sec: 3376.47 - lr: 0.000019 - momentum: 0.000000
2023-10-17 19:53:10,292 epoch 5 - iter 135/272 - loss 0.03610795 - time (sec): 7.94 - samples/sec: 3291.87 - lr: 0.000018 - momentum: 0.000000
2023-10-17 19:53:11,877 epoch 5 - iter 162/272 - loss 0.03311822 - time (sec): 9.53 - samples/sec: 3282.79 - lr: 0.000018 - momentum: 0.000000
2023-10-17 19:53:13,423 epoch 5 - iter 189/272 - loss 0.03480952 - time (sec): 11.07 - samples/sec: 3272.28 - lr: 0.000018 - momentum: 0.000000
2023-10-17 19:53:15,042 epoch 5 - iter 216/272 - loss 0.03330190 - time (sec): 12.69 - samples/sec: 3281.77 - lr: 0.000017 - momentum: 0.000000
2023-10-17 19:53:16,639 epoch 5 - iter 243/272 - loss 0.03351532 - time (sec): 14.29 - samples/sec: 3239.63 - lr: 0.000017 - momentum: 0.000000
2023-10-17 19:53:18,245 epoch 5 - iter 270/272 - loss 0.03210995 - time (sec): 15.89 - samples/sec: 3262.48 - lr: 0.000017 - momentum: 0.000000
2023-10-17 19:53:18,326 ----------------------------------------------------------------------------------------------------
2023-10-17 19:53:18,326 EPOCH 5 done: loss 0.0321 - lr: 0.000017
2023-10-17 19:53:19,962 DEV : loss 0.1374482363462448 - f1-score (micro avg) 0.8007
2023-10-17 19:53:19,967 saving best model
2023-10-17 19:53:20,434 ----------------------------------------------------------------------------------------------------
2023-10-17 19:53:21,874 epoch 6 - iter 27/272 - loss 0.01735343 - time (sec): 1.44 - samples/sec: 3253.81 - lr: 0.000016 - momentum: 0.000000
2023-10-17 19:53:23,513 epoch 6 - iter 54/272 - loss 0.03248622 - time (sec): 3.08 - samples/sec: 3203.61 - lr: 0.000016 - momentum: 0.000000
2023-10-17 19:53:25,105 epoch 6 - iter 81/272 - loss 0.02783744 - time (sec): 4.67 - samples/sec: 3253.09 - lr: 0.000016 - momentum: 0.000000
2023-10-17 19:53:26,637 epoch 6 - iter 108/272 - loss 0.02635058 - time (sec): 6.20 - samples/sec: 3270.39 - lr: 0.000015 - momentum: 0.000000
2023-10-17 19:53:28,205 epoch 6 - iter 135/272 - loss 0.02425374 - time (sec): 7.77 - samples/sec: 3335.38 - lr: 0.000015 - momentum: 0.000000
2023-10-17 19:53:29,847 epoch 6 - iter 162/272 - loss 0.02336843 - time (sec): 9.41 - samples/sec: 3393.10 - lr: 0.000015 - momentum: 0.000000
2023-10-17 19:53:31,441 epoch 6 - iter 189/272 - loss 0.02378891 - time (sec): 11.00 - samples/sec: 3362.54 - lr: 0.000014 - momentum: 0.000000
2023-10-17 19:53:32,990 epoch 6 - iter 216/272 - loss 0.02378025 - time (sec): 12.55 - samples/sec: 3323.37 - lr: 0.000014 - momentum: 0.000000
2023-10-17 19:53:34,525 epoch 6 - iter 243/272 - loss 0.02280147 - time (sec): 14.09 - samples/sec: 3307.37 - lr: 0.000014 - momentum: 0.000000
2023-10-17 19:53:36,095 epoch 6 - iter 270/272 - loss 0.02523429 - time (sec): 15.66 - samples/sec: 3308.99 - lr: 0.000013 - momentum: 0.000000
2023-10-17 19:53:36,194 ----------------------------------------------------------------------------------------------------
2023-10-17 19:53:36,194 EPOCH 6 done: loss 0.0252 - lr: 0.000013
2023-10-17 19:53:37,620 DEV : loss 0.15510904788970947 - f1-score (micro avg) 0.8015
2023-10-17 19:53:37,625 saving best model
2023-10-17 19:53:38,100 ----------------------------------------------------------------------------------------------------
2023-10-17 19:53:39,945 epoch 7 - iter 27/272 - loss 0.02080006 - time (sec): 1.84 - samples/sec: 3386.14 - lr: 0.000013 - momentum: 0.000000
2023-10-17 19:53:41,440 epoch 7 - iter 54/272 - loss 0.02050320 - time (sec): 3.34 - samples/sec: 3404.08 - lr: 0.000013 - momentum: 0.000000
2023-10-17 19:53:42,852 epoch 7 - iter 81/272 - loss 0.01895013 - time (sec): 4.75 - samples/sec: 3325.27 - lr: 0.000012 - momentum: 0.000000
2023-10-17 19:53:44,336 epoch 7 - iter 108/272 - loss 0.01855506 - time (sec): 6.23 - samples/sec: 3236.28 - lr: 0.000012 - momentum: 0.000000
2023-10-17 19:53:45,911 epoch 7 - iter 135/272 - loss 0.01638277 - time (sec): 7.81 - samples/sec: 3220.02 - lr: 0.000012 - momentum: 0.000000
2023-10-17 19:53:47,494 epoch 7 - iter 162/272 - loss 0.01646283 - time (sec): 9.39 - samples/sec: 3229.98 - lr: 0.000011 - momentum: 0.000000
2023-10-17 19:53:49,033 epoch 7 - iter 189/272 - loss 0.01493295 - time (sec): 10.93 - samples/sec: 3265.83 - lr: 0.000011 - momentum: 0.000000
2023-10-17 19:53:50,573 epoch 7 - iter 216/272 - loss 0.01531514 - time (sec): 12.47 - samples/sec: 3293.68 - lr: 0.000011 - momentum: 0.000000
2023-10-17 19:53:52,264 epoch 7 - iter 243/272 - loss 0.01629045 - time (sec): 14.16 - samples/sec: 3275.62 - lr: 0.000010 - momentum: 0.000000
2023-10-17 19:53:53,909 epoch 7 - iter 270/272 - loss 0.01759728 - time (sec): 15.81 - samples/sec: 3268.76 - lr: 0.000010 - momentum: 0.000000
2023-10-17 19:53:54,017 ----------------------------------------------------------------------------------------------------
2023-10-17 19:53:54,017 EPOCH 7 done: loss 0.0175 - lr: 0.000010
2023-10-17 19:53:55,472 DEV : loss 0.1710209846496582 - f1-score (micro avg) 0.8118
2023-10-17 19:53:55,477 saving best model
2023-10-17 19:53:55,954 ----------------------------------------------------------------------------------------------------
2023-10-17 19:53:57,507 epoch 8 - iter 27/272 - loss 0.01604160 - time (sec): 1.55 - samples/sec: 3207.60 - lr: 0.000010 - momentum: 0.000000
2023-10-17 19:53:59,234 epoch 8 - iter 54/272 - loss 0.01064724 - time (sec): 3.28 - samples/sec: 3353.42 - lr: 0.000009 - momentum: 0.000000
2023-10-17 19:54:00,786 epoch 8 - iter 81/272 - loss 0.00922370 - time (sec): 4.83 - samples/sec: 3384.65 - lr: 0.000009 - momentum: 0.000000
2023-10-17 19:54:02,279 epoch 8 - iter 108/272 - loss 0.01069396 - time (sec): 6.32 - samples/sec: 3298.73 - lr: 0.000009 - momentum: 0.000000
2023-10-17 19:54:03,890 epoch 8 - iter 135/272 - loss 0.01198809 - time (sec): 7.93 - samples/sec: 3323.78 - lr: 0.000008 - momentum: 0.000000
2023-10-17 19:54:05,494 epoch 8 - iter 162/272 - loss 0.01143605 - time (sec): 9.54 - samples/sec: 3313.47 - lr: 0.000008 - momentum: 0.000000
2023-10-17 19:54:07,237 epoch 8 - iter 189/272 - loss 0.01236243 - time (sec): 11.28 - samples/sec: 3354.86 - lr: 0.000008 - momentum: 0.000000
2023-10-17 19:54:08,640 epoch 8 - iter 216/272 - loss 0.01336567 - time (sec): 12.68 - samples/sec: 3310.13 - lr: 0.000007 - momentum: 0.000000
2023-10-17 19:54:10,128 epoch 8 - iter 243/272 - loss 0.01295151 - time (sec): 14.17 - samples/sec: 3263.95 - lr: 0.000007 - momentum: 0.000000
2023-10-17 19:54:11,769 epoch 8 - iter 270/272 - loss 0.01256788 - time (sec): 15.81 - samples/sec: 3273.79 - lr: 0.000007 - momentum: 0.000000
2023-10-17 19:54:11,859 ----------------------------------------------------------------------------------------------------
2023-10-17 19:54:11,859 EPOCH 8 done: loss 0.0126 - lr: 0.000007
2023-10-17 19:54:13,299 DEV : loss 0.17594939470291138 - f1-score (micro avg) 0.8118
2023-10-17 19:54:13,305 ----------------------------------------------------------------------------------------------------
2023-10-17 19:54:14,823 epoch 9 - iter 27/272 - loss 0.00194437 - time (sec): 1.52 - samples/sec: 3162.49 - lr: 0.000006 - momentum: 0.000000
2023-10-17 19:54:16,484 epoch 9 - iter 54/272 - loss 0.00302095 - time (sec): 3.18 - samples/sec: 3141.62 - lr: 0.000006 - momentum: 0.000000
2023-10-17 19:54:17,957 epoch 9 - iter 81/272 - loss 0.00280715 - time (sec): 4.65 - samples/sec: 3030.41 - lr: 0.000006 - momentum: 0.000000
2023-10-17 19:54:19,699 epoch 9 - iter 108/272 - loss 0.00750767 - time (sec): 6.39 - samples/sec: 3148.64 - lr: 0.000005 - momentum: 0.000000
2023-10-17 19:54:21,168 epoch 9 - iter 135/272 - loss 0.01060348 - time (sec): 7.86 - samples/sec: 3153.49 - lr: 0.000005 - momentum: 0.000000
2023-10-17 19:54:22,717 epoch 9 - iter 162/272 - loss 0.01005783 - time (sec): 9.41 - samples/sec: 3157.07 - lr: 0.000005 - momentum: 0.000000
2023-10-17 19:54:24,425 epoch 9 - iter 189/272 - loss 0.00964176 - time (sec): 11.12 - samples/sec: 3278.39 - lr: 0.000004 - momentum: 0.000000
2023-10-17 19:54:26,194 epoch 9 - iter 216/272 - loss 0.01037476 - time (sec): 12.89 - samples/sec: 3239.70 - lr: 0.000004 - momentum: 0.000000
2023-10-17 19:54:27,722 epoch 9 - iter 243/272 - loss 0.00974212 - time (sec): 14.42 - samples/sec: 3207.43 - lr: 0.000004 - momentum: 0.000000
2023-10-17 19:54:29,301 epoch 9 - iter 270/272 - loss 0.00880051 - time (sec): 15.99 - samples/sec: 3238.36 - lr: 0.000003 - momentum: 0.000000
2023-10-17 19:54:29,386 ----------------------------------------------------------------------------------------------------
2023-10-17 19:54:29,386 EPOCH 9 done: loss 0.0089 - lr: 0.000003
2023-10-17 19:54:30,831 DEV : loss 0.18299776315689087 - f1-score (micro avg) 0.8059
2023-10-17 19:54:30,836 ----------------------------------------------------------------------------------------------------
2023-10-17 19:54:32,339 epoch 10 - iter 27/272 - loss 0.00732078 - time (sec): 1.50 - samples/sec: 3372.08 - lr: 0.000003 - momentum: 0.000000
2023-10-17 19:54:33,860 epoch 10 - iter 54/272 - loss 0.00549127 - time (sec): 3.02 - samples/sec: 3236.59 - lr: 0.000003 - momentum: 0.000000
2023-10-17 19:54:35,365 epoch 10 - iter 81/272 - loss 0.00525623 - time (sec): 4.53 - samples/sec: 3208.88 - lr: 0.000002 - momentum: 0.000000
2023-10-17 19:54:36,901 epoch 10 - iter 108/272 - loss 0.00566812 - time (sec): 6.06 - samples/sec: 3298.58 - lr: 0.000002 - momentum: 0.000000
2023-10-17 19:54:38,576 epoch 10 - iter 135/272 - loss 0.00528647 - time (sec): 7.74 - samples/sec: 3324.83 - lr: 0.000002 - momentum: 0.000000
2023-10-17 19:54:40,325 epoch 10 - iter 162/272 - loss 0.00586342 - time (sec): 9.49 - samples/sec: 3329.11 - lr: 0.000001 - momentum: 0.000000
2023-10-17 19:54:41,862 epoch 10 - iter 189/272 - loss 0.00551525 - time (sec): 11.03 - samples/sec: 3281.65 - lr: 0.000001 - momentum: 0.000000
2023-10-17 19:54:43,484 epoch 10 - iter 216/272 - loss 0.00658959 - time (sec): 12.65 - samples/sec: 3263.82 - lr: 0.000001 - momentum: 0.000000
2023-10-17 19:54:45,242 epoch 10 - iter 243/272 - loss 0.00839720 - time (sec): 14.40 - samples/sec: 3257.94 - lr: 0.000000 - momentum: 0.000000
2023-10-17 19:54:46,813 epoch 10 - iter 270/272 - loss 0.00763962 - time (sec): 15.98 - samples/sec: 3242.16 - lr: 0.000000 - momentum: 0.000000
2023-10-17 19:54:46,910 ----------------------------------------------------------------------------------------------------
2023-10-17 19:54:46,911 EPOCH 10 done: loss 0.0076 - lr: 0.000000
2023-10-17 19:54:48,333 DEV : loss 0.1869087666273117 - f1-score (micro avg) 0.8067
2023-10-17 19:54:48,709 ----------------------------------------------------------------------------------------------------
2023-10-17 19:54:48,710 Loading model from best epoch ...
2023-10-17 19:54:50,049 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
2023-10-17 19:54:52,035
Results:
- F-score (micro) 0.781
- F-score (macro) 0.7137
- Accuracy 0.6595
By class:
precision recall f1-score support
LOC 0.7988 0.8526 0.8248 312
PER 0.7143 0.8654 0.7826 208
ORG 0.5778 0.4727 0.5200 55
HumanProd 0.6061 0.9091 0.7273 22
micro avg 0.7421 0.8241 0.7810 597
macro avg 0.6742 0.7749 0.7137 597
weighted avg 0.7419 0.8241 0.7784 597
2023-10-17 19:54:52,036 ----------------------------------------------------------------------------------------------------
|