File size: 24,009 Bytes
81c9ad9 |
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 |
2023-10-17 09:46:43,097 ----------------------------------------------------------------------------------------------------
2023-10-17 09:46:43,098 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=25, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-17 09:46:43,098 ----------------------------------------------------------------------------------------------------
2023-10-17 09:46:43,098 MultiCorpus: 1214 train + 266 dev + 251 test sentences
- NER_HIPE_2022 Corpus: 1214 train + 266 dev + 251 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/ajmc/en/with_doc_seperator
2023-10-17 09:46:43,099 ----------------------------------------------------------------------------------------------------
2023-10-17 09:46:43,099 Train: 1214 sentences
2023-10-17 09:46:43,099 (train_with_dev=False, train_with_test=False)
2023-10-17 09:46:43,099 ----------------------------------------------------------------------------------------------------
2023-10-17 09:46:43,099 Training Params:
2023-10-17 09:46:43,099 - learning_rate: "3e-05"
2023-10-17 09:46:43,099 - mini_batch_size: "8"
2023-10-17 09:46:43,099 - max_epochs: "10"
2023-10-17 09:46:43,099 - shuffle: "True"
2023-10-17 09:46:43,099 ----------------------------------------------------------------------------------------------------
2023-10-17 09:46:43,099 Plugins:
2023-10-17 09:46:43,099 - TensorboardLogger
2023-10-17 09:46:43,099 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 09:46:43,099 ----------------------------------------------------------------------------------------------------
2023-10-17 09:46:43,099 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 09:46:43,099 - metric: "('micro avg', 'f1-score')"
2023-10-17 09:46:43,099 ----------------------------------------------------------------------------------------------------
2023-10-17 09:46:43,100 Computation:
2023-10-17 09:46:43,100 - compute on device: cuda:0
2023-10-17 09:46:43,100 - embedding storage: none
2023-10-17 09:46:43,100 ----------------------------------------------------------------------------------------------------
2023-10-17 09:46:43,100 Model training base path: "hmbench-ajmc/en-hmteams/teams-base-historic-multilingual-discriminator-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-2"
2023-10-17 09:46:43,100 ----------------------------------------------------------------------------------------------------
2023-10-17 09:46:43,100 ----------------------------------------------------------------------------------------------------
2023-10-17 09:46:43,100 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 09:46:43,959 epoch 1 - iter 15/152 - loss 4.08709480 - time (sec): 0.86 - samples/sec: 3453.45 - lr: 0.000003 - momentum: 0.000000
2023-10-17 09:46:44,872 epoch 1 - iter 30/152 - loss 3.67349628 - time (sec): 1.77 - samples/sec: 3469.52 - lr: 0.000006 - momentum: 0.000000
2023-10-17 09:46:45,776 epoch 1 - iter 45/152 - loss 2.98257228 - time (sec): 2.67 - samples/sec: 3487.86 - lr: 0.000009 - momentum: 0.000000
2023-10-17 09:46:46,625 epoch 1 - iter 60/152 - loss 2.43261488 - time (sec): 3.52 - samples/sec: 3470.07 - lr: 0.000012 - momentum: 0.000000
2023-10-17 09:46:47,505 epoch 1 - iter 75/152 - loss 2.07662538 - time (sec): 4.40 - samples/sec: 3528.10 - lr: 0.000015 - momentum: 0.000000
2023-10-17 09:46:48,321 epoch 1 - iter 90/152 - loss 1.85236036 - time (sec): 5.22 - samples/sec: 3504.06 - lr: 0.000018 - momentum: 0.000000
2023-10-17 09:46:49,206 epoch 1 - iter 105/152 - loss 1.66941765 - time (sec): 6.10 - samples/sec: 3488.69 - lr: 0.000021 - momentum: 0.000000
2023-10-17 09:46:50,046 epoch 1 - iter 120/152 - loss 1.51562782 - time (sec): 6.94 - samples/sec: 3505.92 - lr: 0.000023 - momentum: 0.000000
2023-10-17 09:46:50,908 epoch 1 - iter 135/152 - loss 1.39014384 - time (sec): 7.81 - samples/sec: 3523.69 - lr: 0.000026 - momentum: 0.000000
2023-10-17 09:46:51,799 epoch 1 - iter 150/152 - loss 1.27593710 - time (sec): 8.70 - samples/sec: 3530.79 - lr: 0.000029 - momentum: 0.000000
2023-10-17 09:46:51,913 ----------------------------------------------------------------------------------------------------
2023-10-17 09:46:51,913 EPOCH 1 done: loss 1.2662 - lr: 0.000029
2023-10-17 09:46:52,828 DEV : loss 0.23923853039741516 - f1-score (micro avg) 0.5092
2023-10-17 09:46:52,835 saving best model
2023-10-17 09:46:53,170 ----------------------------------------------------------------------------------------------------
2023-10-17 09:46:54,038 epoch 2 - iter 15/152 - loss 0.25968914 - time (sec): 0.87 - samples/sec: 3534.88 - lr: 0.000030 - momentum: 0.000000
2023-10-17 09:46:54,955 epoch 2 - iter 30/152 - loss 0.22279190 - time (sec): 1.78 - samples/sec: 3502.97 - lr: 0.000029 - momentum: 0.000000
2023-10-17 09:46:55,883 epoch 2 - iter 45/152 - loss 0.21763067 - time (sec): 2.71 - samples/sec: 3365.72 - lr: 0.000029 - momentum: 0.000000
2023-10-17 09:46:56,778 epoch 2 - iter 60/152 - loss 0.21582908 - time (sec): 3.61 - samples/sec: 3366.66 - lr: 0.000029 - momentum: 0.000000
2023-10-17 09:46:57,651 epoch 2 - iter 75/152 - loss 0.20207098 - time (sec): 4.48 - samples/sec: 3414.93 - lr: 0.000028 - momentum: 0.000000
2023-10-17 09:46:58,492 epoch 2 - iter 90/152 - loss 0.19077238 - time (sec): 5.32 - samples/sec: 3434.00 - lr: 0.000028 - momentum: 0.000000
2023-10-17 09:46:59,339 epoch 2 - iter 105/152 - loss 0.17992756 - time (sec): 6.17 - samples/sec: 3449.70 - lr: 0.000028 - momentum: 0.000000
2023-10-17 09:47:00,207 epoch 2 - iter 120/152 - loss 0.17275352 - time (sec): 7.04 - samples/sec: 3467.53 - lr: 0.000027 - momentum: 0.000000
2023-10-17 09:47:01,059 epoch 2 - iter 135/152 - loss 0.17265650 - time (sec): 7.89 - samples/sec: 3503.04 - lr: 0.000027 - momentum: 0.000000
2023-10-17 09:47:01,892 epoch 2 - iter 150/152 - loss 0.17032603 - time (sec): 8.72 - samples/sec: 3520.36 - lr: 0.000027 - momentum: 0.000000
2023-10-17 09:47:01,988 ----------------------------------------------------------------------------------------------------
2023-10-17 09:47:01,988 EPOCH 2 done: loss 0.1692 - lr: 0.000027
2023-10-17 09:47:02,937 DEV : loss 0.12957707047462463 - f1-score (micro avg) 0.7696
2023-10-17 09:47:02,944 saving best model
2023-10-17 09:47:03,404 ----------------------------------------------------------------------------------------------------
2023-10-17 09:47:04,236 epoch 3 - iter 15/152 - loss 0.11222872 - time (sec): 0.83 - samples/sec: 3805.52 - lr: 0.000026 - momentum: 0.000000
2023-10-17 09:47:05,077 epoch 3 - iter 30/152 - loss 0.10389302 - time (sec): 1.67 - samples/sec: 3593.03 - lr: 0.000026 - momentum: 0.000000
2023-10-17 09:47:05,894 epoch 3 - iter 45/152 - loss 0.10041671 - time (sec): 2.49 - samples/sec: 3604.24 - lr: 0.000026 - momentum: 0.000000
2023-10-17 09:47:06,752 epoch 3 - iter 60/152 - loss 0.09597802 - time (sec): 3.35 - samples/sec: 3565.52 - lr: 0.000025 - momentum: 0.000000
2023-10-17 09:47:07,632 epoch 3 - iter 75/152 - loss 0.10191719 - time (sec): 4.23 - samples/sec: 3539.25 - lr: 0.000025 - momentum: 0.000000
2023-10-17 09:47:08,493 epoch 3 - iter 90/152 - loss 0.10150171 - time (sec): 5.09 - samples/sec: 3546.28 - lr: 0.000025 - momentum: 0.000000
2023-10-17 09:47:09,349 epoch 3 - iter 105/152 - loss 0.09625097 - time (sec): 5.94 - samples/sec: 3565.65 - lr: 0.000024 - momentum: 0.000000
2023-10-17 09:47:10,243 epoch 3 - iter 120/152 - loss 0.09520497 - time (sec): 6.84 - samples/sec: 3600.78 - lr: 0.000024 - momentum: 0.000000
2023-10-17 09:47:11,064 epoch 3 - iter 135/152 - loss 0.09136958 - time (sec): 7.66 - samples/sec: 3623.79 - lr: 0.000024 - momentum: 0.000000
2023-10-17 09:47:11,924 epoch 3 - iter 150/152 - loss 0.09165308 - time (sec): 8.52 - samples/sec: 3604.06 - lr: 0.000023 - momentum: 0.000000
2023-10-17 09:47:12,031 ----------------------------------------------------------------------------------------------------
2023-10-17 09:47:12,031 EPOCH 3 done: loss 0.0928 - lr: 0.000023
2023-10-17 09:47:12,981 DEV : loss 0.1297578513622284 - f1-score (micro avg) 0.8223
2023-10-17 09:47:12,988 saving best model
2023-10-17 09:47:13,500 ----------------------------------------------------------------------------------------------------
2023-10-17 09:47:14,297 epoch 4 - iter 15/152 - loss 0.04986111 - time (sec): 0.79 - samples/sec: 3513.29 - lr: 0.000023 - momentum: 0.000000
2023-10-17 09:47:15,108 epoch 4 - iter 30/152 - loss 0.06192583 - time (sec): 1.61 - samples/sec: 3453.16 - lr: 0.000023 - momentum: 0.000000
2023-10-17 09:47:15,949 epoch 4 - iter 45/152 - loss 0.06310085 - time (sec): 2.45 - samples/sec: 3442.15 - lr: 0.000022 - momentum: 0.000000
2023-10-17 09:47:16,807 epoch 4 - iter 60/152 - loss 0.05987327 - time (sec): 3.31 - samples/sec: 3434.64 - lr: 0.000022 - momentum: 0.000000
2023-10-17 09:47:17,655 epoch 4 - iter 75/152 - loss 0.06234632 - time (sec): 4.15 - samples/sec: 3493.64 - lr: 0.000022 - momentum: 0.000000
2023-10-17 09:47:18,530 epoch 4 - iter 90/152 - loss 0.06373864 - time (sec): 5.03 - samples/sec: 3538.26 - lr: 0.000021 - momentum: 0.000000
2023-10-17 09:47:19,363 epoch 4 - iter 105/152 - loss 0.06595611 - time (sec): 5.86 - samples/sec: 3564.53 - lr: 0.000021 - momentum: 0.000000
2023-10-17 09:47:20,231 epoch 4 - iter 120/152 - loss 0.06558915 - time (sec): 6.73 - samples/sec: 3574.10 - lr: 0.000021 - momentum: 0.000000
2023-10-17 09:47:21,136 epoch 4 - iter 135/152 - loss 0.06663958 - time (sec): 7.63 - samples/sec: 3575.29 - lr: 0.000020 - momentum: 0.000000
2023-10-17 09:47:22,058 epoch 4 - iter 150/152 - loss 0.06251869 - time (sec): 8.56 - samples/sec: 3576.16 - lr: 0.000020 - momentum: 0.000000
2023-10-17 09:47:22,163 ----------------------------------------------------------------------------------------------------
2023-10-17 09:47:22,164 EPOCH 4 done: loss 0.0631 - lr: 0.000020
2023-10-17 09:47:23,133 DEV : loss 0.1361662745475769 - f1-score (micro avg) 0.8489
2023-10-17 09:47:23,140 saving best model
2023-10-17 09:47:23,624 ----------------------------------------------------------------------------------------------------
2023-10-17 09:47:24,515 epoch 5 - iter 15/152 - loss 0.02755188 - time (sec): 0.89 - samples/sec: 3563.62 - lr: 0.000020 - momentum: 0.000000
2023-10-17 09:47:25,340 epoch 5 - iter 30/152 - loss 0.05347785 - time (sec): 1.71 - samples/sec: 3458.92 - lr: 0.000019 - momentum: 0.000000
2023-10-17 09:47:26,187 epoch 5 - iter 45/152 - loss 0.05330478 - time (sec): 2.56 - samples/sec: 3507.33 - lr: 0.000019 - momentum: 0.000000
2023-10-17 09:47:27,011 epoch 5 - iter 60/152 - loss 0.05452557 - time (sec): 3.39 - samples/sec: 3502.84 - lr: 0.000019 - momentum: 0.000000
2023-10-17 09:47:27,829 epoch 5 - iter 75/152 - loss 0.05130820 - time (sec): 4.20 - samples/sec: 3498.76 - lr: 0.000018 - momentum: 0.000000
2023-10-17 09:47:28,697 epoch 5 - iter 90/152 - loss 0.05010256 - time (sec): 5.07 - samples/sec: 3590.48 - lr: 0.000018 - momentum: 0.000000
2023-10-17 09:47:29,623 epoch 5 - iter 105/152 - loss 0.04542425 - time (sec): 6.00 - samples/sec: 3590.37 - lr: 0.000018 - momentum: 0.000000
2023-10-17 09:47:30,536 epoch 5 - iter 120/152 - loss 0.04294284 - time (sec): 6.91 - samples/sec: 3549.85 - lr: 0.000017 - momentum: 0.000000
2023-10-17 09:47:31,399 epoch 5 - iter 135/152 - loss 0.04426878 - time (sec): 7.77 - samples/sec: 3532.31 - lr: 0.000017 - momentum: 0.000000
2023-10-17 09:47:32,276 epoch 5 - iter 150/152 - loss 0.04471386 - time (sec): 8.65 - samples/sec: 3538.31 - lr: 0.000017 - momentum: 0.000000
2023-10-17 09:47:32,391 ----------------------------------------------------------------------------------------------------
2023-10-17 09:47:32,391 EPOCH 5 done: loss 0.0453 - lr: 0.000017
2023-10-17 09:47:33,354 DEV : loss 0.1634262502193451 - f1-score (micro avg) 0.8416
2023-10-17 09:47:33,361 ----------------------------------------------------------------------------------------------------
2023-10-17 09:47:34,309 epoch 6 - iter 15/152 - loss 0.02365755 - time (sec): 0.95 - samples/sec: 3485.90 - lr: 0.000016 - momentum: 0.000000
2023-10-17 09:47:35,243 epoch 6 - iter 30/152 - loss 0.03444613 - time (sec): 1.88 - samples/sec: 3533.94 - lr: 0.000016 - momentum: 0.000000
2023-10-17 09:47:36,111 epoch 6 - iter 45/152 - loss 0.03248390 - time (sec): 2.75 - samples/sec: 3539.09 - lr: 0.000016 - momentum: 0.000000
2023-10-17 09:47:36,966 epoch 6 - iter 60/152 - loss 0.02712965 - time (sec): 3.60 - samples/sec: 3551.06 - lr: 0.000015 - momentum: 0.000000
2023-10-17 09:47:37,834 epoch 6 - iter 75/152 - loss 0.02826301 - time (sec): 4.47 - samples/sec: 3539.72 - lr: 0.000015 - momentum: 0.000000
2023-10-17 09:47:38,671 epoch 6 - iter 90/152 - loss 0.03537780 - time (sec): 5.31 - samples/sec: 3493.33 - lr: 0.000015 - momentum: 0.000000
2023-10-17 09:47:39,488 epoch 6 - iter 105/152 - loss 0.03787211 - time (sec): 6.13 - samples/sec: 3487.84 - lr: 0.000014 - momentum: 0.000000
2023-10-17 09:47:40,346 epoch 6 - iter 120/152 - loss 0.03866626 - time (sec): 6.98 - samples/sec: 3490.52 - lr: 0.000014 - momentum: 0.000000
2023-10-17 09:47:41,241 epoch 6 - iter 135/152 - loss 0.03845673 - time (sec): 7.88 - samples/sec: 3480.34 - lr: 0.000014 - momentum: 0.000000
2023-10-17 09:47:42,099 epoch 6 - iter 150/152 - loss 0.03701677 - time (sec): 8.74 - samples/sec: 3507.30 - lr: 0.000013 - momentum: 0.000000
2023-10-17 09:47:42,196 ----------------------------------------------------------------------------------------------------
2023-10-17 09:47:42,196 EPOCH 6 done: loss 0.0367 - lr: 0.000013
2023-10-17 09:47:43,228 DEV : loss 0.1642225831747055 - f1-score (micro avg) 0.8565
2023-10-17 09:47:43,235 saving best model
2023-10-17 09:47:43,723 ----------------------------------------------------------------------------------------------------
2023-10-17 09:47:44,589 epoch 7 - iter 15/152 - loss 0.04256990 - time (sec): 0.86 - samples/sec: 3311.66 - lr: 0.000013 - momentum: 0.000000
2023-10-17 09:47:45,408 epoch 7 - iter 30/152 - loss 0.03497680 - time (sec): 1.68 - samples/sec: 3455.63 - lr: 0.000013 - momentum: 0.000000
2023-10-17 09:47:46,308 epoch 7 - iter 45/152 - loss 0.03729344 - time (sec): 2.58 - samples/sec: 3502.72 - lr: 0.000012 - momentum: 0.000000
2023-10-17 09:47:47,172 epoch 7 - iter 60/152 - loss 0.03052896 - time (sec): 3.45 - samples/sec: 3484.81 - lr: 0.000012 - momentum: 0.000000
2023-10-17 09:47:48,031 epoch 7 - iter 75/152 - loss 0.02753474 - time (sec): 4.31 - samples/sec: 3500.61 - lr: 0.000012 - momentum: 0.000000
2023-10-17 09:47:48,900 epoch 7 - iter 90/152 - loss 0.02898334 - time (sec): 5.18 - samples/sec: 3449.95 - lr: 0.000011 - momentum: 0.000000
2023-10-17 09:47:49,794 epoch 7 - iter 105/152 - loss 0.02789337 - time (sec): 6.07 - samples/sec: 3447.00 - lr: 0.000011 - momentum: 0.000000
2023-10-17 09:47:50,693 epoch 7 - iter 120/152 - loss 0.02796994 - time (sec): 6.97 - samples/sec: 3455.20 - lr: 0.000011 - momentum: 0.000000
2023-10-17 09:47:51,613 epoch 7 - iter 135/152 - loss 0.02876490 - time (sec): 7.89 - samples/sec: 3463.24 - lr: 0.000010 - momentum: 0.000000
2023-10-17 09:47:52,484 epoch 7 - iter 150/152 - loss 0.02832453 - time (sec): 8.76 - samples/sec: 3499.54 - lr: 0.000010 - momentum: 0.000000
2023-10-17 09:47:52,593 ----------------------------------------------------------------------------------------------------
2023-10-17 09:47:52,593 EPOCH 7 done: loss 0.0280 - lr: 0.000010
2023-10-17 09:47:53,732 DEV : loss 0.18156136572360992 - f1-score (micro avg) 0.8605
2023-10-17 09:47:53,739 saving best model
2023-10-17 09:47:54,201 ----------------------------------------------------------------------------------------------------
2023-10-17 09:47:55,053 epoch 8 - iter 15/152 - loss 0.06044905 - time (sec): 0.85 - samples/sec: 3469.12 - lr: 0.000010 - momentum: 0.000000
2023-10-17 09:47:55,916 epoch 8 - iter 30/152 - loss 0.03014943 - time (sec): 1.71 - samples/sec: 3474.85 - lr: 0.000009 - momentum: 0.000000
2023-10-17 09:47:56,804 epoch 8 - iter 45/152 - loss 0.02352535 - time (sec): 2.60 - samples/sec: 3544.30 - lr: 0.000009 - momentum: 0.000000
2023-10-17 09:47:57,630 epoch 8 - iter 60/152 - loss 0.02298262 - time (sec): 3.43 - samples/sec: 3537.91 - lr: 0.000009 - momentum: 0.000000
2023-10-17 09:47:58,467 epoch 8 - iter 75/152 - loss 0.02611197 - time (sec): 4.26 - samples/sec: 3542.55 - lr: 0.000008 - momentum: 0.000000
2023-10-17 09:47:59,376 epoch 8 - iter 90/152 - loss 0.02299230 - time (sec): 5.17 - samples/sec: 3547.10 - lr: 0.000008 - momentum: 0.000000
2023-10-17 09:48:00,198 epoch 8 - iter 105/152 - loss 0.02172982 - time (sec): 6.00 - samples/sec: 3544.12 - lr: 0.000008 - momentum: 0.000000
2023-10-17 09:48:01,086 epoch 8 - iter 120/152 - loss 0.02326433 - time (sec): 6.88 - samples/sec: 3575.70 - lr: 0.000007 - momentum: 0.000000
2023-10-17 09:48:01,916 epoch 8 - iter 135/152 - loss 0.02337614 - time (sec): 7.71 - samples/sec: 3581.30 - lr: 0.000007 - momentum: 0.000000
2023-10-17 09:48:02,790 epoch 8 - iter 150/152 - loss 0.02314189 - time (sec): 8.59 - samples/sec: 3568.10 - lr: 0.000007 - momentum: 0.000000
2023-10-17 09:48:02,890 ----------------------------------------------------------------------------------------------------
2023-10-17 09:48:02,890 EPOCH 8 done: loss 0.0236 - lr: 0.000007
2023-10-17 09:48:03,876 DEV : loss 0.1801178753376007 - f1-score (micro avg) 0.8501
2023-10-17 09:48:03,883 ----------------------------------------------------------------------------------------------------
2023-10-17 09:48:04,750 epoch 9 - iter 15/152 - loss 0.01430670 - time (sec): 0.87 - samples/sec: 3286.70 - lr: 0.000006 - momentum: 0.000000
2023-10-17 09:48:05,628 epoch 9 - iter 30/152 - loss 0.01715728 - time (sec): 1.74 - samples/sec: 3406.95 - lr: 0.000006 - momentum: 0.000000
2023-10-17 09:48:06,509 epoch 9 - iter 45/152 - loss 0.02458273 - time (sec): 2.62 - samples/sec: 3420.35 - lr: 0.000006 - momentum: 0.000000
2023-10-17 09:48:07,393 epoch 9 - iter 60/152 - loss 0.02244915 - time (sec): 3.51 - samples/sec: 3491.25 - lr: 0.000005 - momentum: 0.000000
2023-10-17 09:48:08,282 epoch 9 - iter 75/152 - loss 0.01954133 - time (sec): 4.40 - samples/sec: 3471.12 - lr: 0.000005 - momentum: 0.000000
2023-10-17 09:48:09,152 epoch 9 - iter 90/152 - loss 0.02153946 - time (sec): 5.27 - samples/sec: 3459.86 - lr: 0.000005 - momentum: 0.000000
2023-10-17 09:48:10,000 epoch 9 - iter 105/152 - loss 0.01973313 - time (sec): 6.12 - samples/sec: 3442.79 - lr: 0.000004 - momentum: 0.000000
2023-10-17 09:48:10,883 epoch 9 - iter 120/152 - loss 0.01775200 - time (sec): 7.00 - samples/sec: 3425.38 - lr: 0.000004 - momentum: 0.000000
2023-10-17 09:48:11,818 epoch 9 - iter 135/152 - loss 0.01978596 - time (sec): 7.93 - samples/sec: 3458.04 - lr: 0.000004 - momentum: 0.000000
2023-10-17 09:48:12,655 epoch 9 - iter 150/152 - loss 0.01852494 - time (sec): 8.77 - samples/sec: 3478.16 - lr: 0.000004 - momentum: 0.000000
2023-10-17 09:48:12,774 ----------------------------------------------------------------------------------------------------
2023-10-17 09:48:12,774 EPOCH 9 done: loss 0.0190 - lr: 0.000004
2023-10-17 09:48:13,727 DEV : loss 0.18825717270374298 - f1-score (micro avg) 0.8698
2023-10-17 09:48:13,733 saving best model
2023-10-17 09:48:14,211 ----------------------------------------------------------------------------------------------------
2023-10-17 09:48:15,017 epoch 10 - iter 15/152 - loss 0.01308198 - time (sec): 0.80 - samples/sec: 3750.13 - lr: 0.000003 - momentum: 0.000000
2023-10-17 09:48:15,889 epoch 10 - iter 30/152 - loss 0.00736048 - time (sec): 1.68 - samples/sec: 3675.34 - lr: 0.000003 - momentum: 0.000000
2023-10-17 09:48:16,733 epoch 10 - iter 45/152 - loss 0.01711075 - time (sec): 2.52 - samples/sec: 3714.84 - lr: 0.000002 - momentum: 0.000000
2023-10-17 09:48:17,592 epoch 10 - iter 60/152 - loss 0.02056631 - time (sec): 3.38 - samples/sec: 3611.98 - lr: 0.000002 - momentum: 0.000000
2023-10-17 09:48:18,466 epoch 10 - iter 75/152 - loss 0.01835657 - time (sec): 4.25 - samples/sec: 3611.82 - lr: 0.000002 - momentum: 0.000000
2023-10-17 09:48:19,342 epoch 10 - iter 90/152 - loss 0.01965530 - time (sec): 5.13 - samples/sec: 3555.02 - lr: 0.000002 - momentum: 0.000000
2023-10-17 09:48:20,242 epoch 10 - iter 105/152 - loss 0.01854851 - time (sec): 6.03 - samples/sec: 3536.19 - lr: 0.000001 - momentum: 0.000000
2023-10-17 09:48:21,094 epoch 10 - iter 120/152 - loss 0.01829723 - time (sec): 6.88 - samples/sec: 3564.57 - lr: 0.000001 - momentum: 0.000000
2023-10-17 09:48:21,946 epoch 10 - iter 135/152 - loss 0.01684338 - time (sec): 7.73 - samples/sec: 3534.19 - lr: 0.000001 - momentum: 0.000000
2023-10-17 09:48:22,843 epoch 10 - iter 150/152 - loss 0.01616166 - time (sec): 8.63 - samples/sec: 3544.29 - lr: 0.000000 - momentum: 0.000000
2023-10-17 09:48:22,961 ----------------------------------------------------------------------------------------------------
2023-10-17 09:48:22,961 EPOCH 10 done: loss 0.0160 - lr: 0.000000
2023-10-17 09:48:23,944 DEV : loss 0.18469107151031494 - f1-score (micro avg) 0.8629
2023-10-17 09:48:24,317 ----------------------------------------------------------------------------------------------------
2023-10-17 09:48:24,320 Loading model from best epoch ...
2023-10-17 09:48:25,886 SequenceTagger predicts: Dictionary with 25 tags: O, S-scope, B-scope, E-scope, I-scope, S-pers, B-pers, E-pers, I-pers, S-work, B-work, E-work, I-work, S-loc, B-loc, E-loc, I-loc, S-date, B-date, E-date, I-date, S-object, B-object, E-object, I-object
2023-10-17 09:48:26,924
Results:
- F-score (micro) 0.8292
- F-score (macro) 0.642
- Accuracy 0.7133
By class:
precision recall f1-score support
scope 0.7730 0.8344 0.8025 151
work 0.7685 0.8737 0.8177 95
pers 0.9091 0.9375 0.9231 96
date 0.0000 0.0000 0.0000 3
loc 0.6667 0.6667 0.6667 3
micro avg 0.7963 0.8649 0.8292 348
macro avg 0.6235 0.6625 0.6420 348
weighted avg 0.8017 0.8649 0.8319 348
2023-10-17 09:48:26,924 ----------------------------------------------------------------------------------------------------
|