File size: 23,865 Bytes
aab20cd |
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 |
2023-10-25 20:57:45,050 ----------------------------------------------------------------------------------------------------
2023-10-25 20:57:45,051 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 20:57:45,051 ----------------------------------------------------------------------------------------------------
2023-10-25 20:57:45,051 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
2023-10-25 20:57:45,051 ----------------------------------------------------------------------------------------------------
2023-10-25 20:57:45,051 Train: 1166 sentences
2023-10-25 20:57:45,051 (train_with_dev=False, train_with_test=False)
2023-10-25 20:57:45,051 ----------------------------------------------------------------------------------------------------
2023-10-25 20:57:45,051 Training Params:
2023-10-25 20:57:45,051 - learning_rate: "5e-05"
2023-10-25 20:57:45,051 - mini_batch_size: "8"
2023-10-25 20:57:45,051 - max_epochs: "10"
2023-10-25 20:57:45,051 - shuffle: "True"
2023-10-25 20:57:45,051 ----------------------------------------------------------------------------------------------------
2023-10-25 20:57:45,051 Plugins:
2023-10-25 20:57:45,051 - TensorboardLogger
2023-10-25 20:57:45,051 - LinearScheduler | warmup_fraction: '0.1'
2023-10-25 20:57:45,051 ----------------------------------------------------------------------------------------------------
2023-10-25 20:57:45,051 Final evaluation on model from best epoch (best-model.pt)
2023-10-25 20:57:45,051 - metric: "('micro avg', 'f1-score')"
2023-10-25 20:57:45,051 ----------------------------------------------------------------------------------------------------
2023-10-25 20:57:45,052 Computation:
2023-10-25 20:57:45,052 - compute on device: cuda:0
2023-10-25 20:57:45,052 - embedding storage: none
2023-10-25 20:57:45,052 ----------------------------------------------------------------------------------------------------
2023-10-25 20:57:45,052 Model training base path: "hmbench-newseye/fi-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-2"
2023-10-25 20:57:45,052 ----------------------------------------------------------------------------------------------------
2023-10-25 20:57:45,052 ----------------------------------------------------------------------------------------------------
2023-10-25 20:57:45,052 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-25 20:57:45,830 epoch 1 - iter 14/146 - loss 3.01622847 - time (sec): 0.78 - samples/sec: 4751.19 - lr: 0.000004 - momentum: 0.000000
2023-10-25 20:57:46,820 epoch 1 - iter 28/146 - loss 2.22477731 - time (sec): 1.77 - samples/sec: 4730.67 - lr: 0.000009 - momentum: 0.000000
2023-10-25 20:57:47,986 epoch 1 - iter 42/146 - loss 1.61094234 - time (sec): 2.93 - samples/sec: 4669.33 - lr: 0.000014 - momentum: 0.000000
2023-10-25 20:57:48,920 epoch 1 - iter 56/146 - loss 1.36453122 - time (sec): 3.87 - samples/sec: 4681.34 - lr: 0.000019 - momentum: 0.000000
2023-10-25 20:57:49,784 epoch 1 - iter 70/146 - loss 1.18628181 - time (sec): 4.73 - samples/sec: 4724.75 - lr: 0.000024 - momentum: 0.000000
2023-10-25 20:57:50,749 epoch 1 - iter 84/146 - loss 1.04111618 - time (sec): 5.70 - samples/sec: 4734.07 - lr: 0.000028 - momentum: 0.000000
2023-10-25 20:57:51,569 epoch 1 - iter 98/146 - loss 0.95230138 - time (sec): 6.52 - samples/sec: 4705.80 - lr: 0.000033 - momentum: 0.000000
2023-10-25 20:57:52,336 epoch 1 - iter 112/146 - loss 0.88580683 - time (sec): 7.28 - samples/sec: 4691.89 - lr: 0.000038 - momentum: 0.000000
2023-10-25 20:57:53,173 epoch 1 - iter 126/146 - loss 0.82747503 - time (sec): 8.12 - samples/sec: 4669.42 - lr: 0.000043 - momentum: 0.000000
2023-10-25 20:57:54,064 epoch 1 - iter 140/146 - loss 0.77136082 - time (sec): 9.01 - samples/sec: 4629.27 - lr: 0.000048 - momentum: 0.000000
2023-10-25 20:57:54,587 ----------------------------------------------------------------------------------------------------
2023-10-25 20:57:54,588 EPOCH 1 done: loss 0.7478 - lr: 0.000048
2023-10-25 20:57:55,254 DEV : loss 0.1681840568780899 - f1-score (micro avg) 0.5124
2023-10-25 20:57:55,258 saving best model
2023-10-25 20:57:55,724 ----------------------------------------------------------------------------------------------------
2023-10-25 20:57:56,621 epoch 2 - iter 14/146 - loss 0.20128551 - time (sec): 0.90 - samples/sec: 4563.17 - lr: 0.000050 - momentum: 0.000000
2023-10-25 20:57:57,445 epoch 2 - iter 28/146 - loss 0.20146634 - time (sec): 1.72 - samples/sec: 4541.90 - lr: 0.000049 - momentum: 0.000000
2023-10-25 20:57:58,360 epoch 2 - iter 42/146 - loss 0.17940294 - time (sec): 2.63 - samples/sec: 4542.16 - lr: 0.000048 - momentum: 0.000000
2023-10-25 20:57:59,324 epoch 2 - iter 56/146 - loss 0.17263735 - time (sec): 3.60 - samples/sec: 4578.63 - lr: 0.000048 - momentum: 0.000000
2023-10-25 20:58:00,282 epoch 2 - iter 70/146 - loss 0.16408791 - time (sec): 4.56 - samples/sec: 4539.43 - lr: 0.000047 - momentum: 0.000000
2023-10-25 20:58:01,100 epoch 2 - iter 84/146 - loss 0.16825407 - time (sec): 5.37 - samples/sec: 4527.82 - lr: 0.000047 - momentum: 0.000000
2023-10-25 20:58:01,982 epoch 2 - iter 98/146 - loss 0.17462443 - time (sec): 6.26 - samples/sec: 4519.57 - lr: 0.000046 - momentum: 0.000000
2023-10-25 20:58:02,872 epoch 2 - iter 112/146 - loss 0.17533292 - time (sec): 7.15 - samples/sec: 4507.14 - lr: 0.000046 - momentum: 0.000000
2023-10-25 20:58:03,920 epoch 2 - iter 126/146 - loss 0.17354615 - time (sec): 8.19 - samples/sec: 4578.78 - lr: 0.000045 - momentum: 0.000000
2023-10-25 20:58:04,812 epoch 2 - iter 140/146 - loss 0.16374339 - time (sec): 9.09 - samples/sec: 4679.53 - lr: 0.000045 - momentum: 0.000000
2023-10-25 20:58:05,196 ----------------------------------------------------------------------------------------------------
2023-10-25 20:58:05,196 EPOCH 2 done: loss 0.1616 - lr: 0.000045
2023-10-25 20:58:06,105 DEV : loss 0.11024896055459976 - f1-score (micro avg) 0.682
2023-10-25 20:58:06,110 saving best model
2023-10-25 20:58:06,722 ----------------------------------------------------------------------------------------------------
2023-10-25 20:58:07,698 epoch 3 - iter 14/146 - loss 0.63961016 - time (sec): 0.97 - samples/sec: 4852.67 - lr: 0.000044 - momentum: 0.000000
2023-10-25 20:58:08,656 epoch 3 - iter 28/146 - loss 0.35982451 - time (sec): 1.93 - samples/sec: 4830.06 - lr: 0.000043 - momentum: 0.000000
2023-10-25 20:58:09,621 epoch 3 - iter 42/146 - loss 0.26511020 - time (sec): 2.90 - samples/sec: 4682.42 - lr: 0.000043 - momentum: 0.000000
2023-10-25 20:58:10,686 epoch 3 - iter 56/146 - loss 0.22289918 - time (sec): 3.96 - samples/sec: 4397.11 - lr: 0.000042 - momentum: 0.000000
2023-10-25 20:58:11,506 epoch 3 - iter 70/146 - loss 0.19749367 - time (sec): 4.78 - samples/sec: 4428.40 - lr: 0.000042 - momentum: 0.000000
2023-10-25 20:58:12,273 epoch 3 - iter 84/146 - loss 0.17871944 - time (sec): 5.55 - samples/sec: 4475.12 - lr: 0.000041 - momentum: 0.000000
2023-10-25 20:58:13,143 epoch 3 - iter 98/146 - loss 0.16277166 - time (sec): 6.42 - samples/sec: 4557.86 - lr: 0.000041 - momentum: 0.000000
2023-10-25 20:58:14,048 epoch 3 - iter 112/146 - loss 0.15065108 - time (sec): 7.32 - samples/sec: 4587.59 - lr: 0.000040 - momentum: 0.000000
2023-10-25 20:58:14,988 epoch 3 - iter 126/146 - loss 0.14336333 - time (sec): 8.26 - samples/sec: 4661.00 - lr: 0.000040 - momentum: 0.000000
2023-10-25 20:58:15,926 epoch 3 - iter 140/146 - loss 0.13551970 - time (sec): 9.20 - samples/sec: 4615.07 - lr: 0.000039 - momentum: 0.000000
2023-10-25 20:58:16,350 ----------------------------------------------------------------------------------------------------
2023-10-25 20:58:16,350 EPOCH 3 done: loss 0.1345 - lr: 0.000039
2023-10-25 20:58:17,262 DEV : loss 0.11921205371618271 - f1-score (micro avg) 0.7597
2023-10-25 20:58:17,266 saving best model
2023-10-25 20:58:17,746 ----------------------------------------------------------------------------------------------------
2023-10-25 20:58:18,649 epoch 4 - iter 14/146 - loss 0.10516627 - time (sec): 0.90 - samples/sec: 4576.84 - lr: 0.000038 - momentum: 0.000000
2023-10-25 20:58:19,517 epoch 4 - iter 28/146 - loss 0.07107744 - time (sec): 1.77 - samples/sec: 4923.37 - lr: 0.000038 - momentum: 0.000000
2023-10-25 20:58:20,388 epoch 4 - iter 42/146 - loss 0.06235583 - time (sec): 2.64 - samples/sec: 4755.79 - lr: 0.000037 - momentum: 0.000000
2023-10-25 20:58:21,254 epoch 4 - iter 56/146 - loss 0.05881476 - time (sec): 3.51 - samples/sec: 4755.40 - lr: 0.000037 - momentum: 0.000000
2023-10-25 20:58:22,257 epoch 4 - iter 70/146 - loss 0.05872437 - time (sec): 4.51 - samples/sec: 4726.52 - lr: 0.000036 - momentum: 0.000000
2023-10-25 20:58:23,161 epoch 4 - iter 84/146 - loss 0.05638929 - time (sec): 5.41 - samples/sec: 4665.22 - lr: 0.000036 - momentum: 0.000000
2023-10-25 20:58:23,967 epoch 4 - iter 98/146 - loss 0.05876076 - time (sec): 6.22 - samples/sec: 4726.33 - lr: 0.000035 - momentum: 0.000000
2023-10-25 20:58:24,995 epoch 4 - iter 112/146 - loss 0.05938172 - time (sec): 7.25 - samples/sec: 4730.39 - lr: 0.000035 - momentum: 0.000000
2023-10-25 20:58:25,817 epoch 4 - iter 126/146 - loss 0.05658355 - time (sec): 8.07 - samples/sec: 4768.04 - lr: 0.000034 - momentum: 0.000000
2023-10-25 20:58:26,711 epoch 4 - iter 140/146 - loss 0.05586036 - time (sec): 8.96 - samples/sec: 4756.48 - lr: 0.000034 - momentum: 0.000000
2023-10-25 20:58:27,068 ----------------------------------------------------------------------------------------------------
2023-10-25 20:58:27,068 EPOCH 4 done: loss 0.0544 - lr: 0.000034
2023-10-25 20:58:27,984 DEV : loss 0.12713664770126343 - f1-score (micro avg) 0.7342
2023-10-25 20:58:27,988 ----------------------------------------------------------------------------------------------------
2023-10-25 20:58:28,752 epoch 5 - iter 14/146 - loss 0.02468832 - time (sec): 0.76 - samples/sec: 4653.36 - lr: 0.000033 - momentum: 0.000000
2023-10-25 20:58:29,736 epoch 5 - iter 28/146 - loss 0.02918576 - time (sec): 1.75 - samples/sec: 4582.50 - lr: 0.000032 - momentum: 0.000000
2023-10-25 20:58:30,628 epoch 5 - iter 42/146 - loss 0.02943205 - time (sec): 2.64 - samples/sec: 4764.19 - lr: 0.000032 - momentum: 0.000000
2023-10-25 20:58:31,539 epoch 5 - iter 56/146 - loss 0.03396994 - time (sec): 3.55 - samples/sec: 4835.17 - lr: 0.000031 - momentum: 0.000000
2023-10-25 20:58:32,304 epoch 5 - iter 70/146 - loss 0.03515634 - time (sec): 4.31 - samples/sec: 4808.79 - lr: 0.000031 - momentum: 0.000000
2023-10-25 20:58:33,274 epoch 5 - iter 84/146 - loss 0.03735341 - time (sec): 5.28 - samples/sec: 4747.26 - lr: 0.000030 - momentum: 0.000000
2023-10-25 20:58:34,265 epoch 5 - iter 98/146 - loss 0.03861920 - time (sec): 6.28 - samples/sec: 4763.08 - lr: 0.000030 - momentum: 0.000000
2023-10-25 20:58:35,213 epoch 5 - iter 112/146 - loss 0.03553686 - time (sec): 7.22 - samples/sec: 4752.47 - lr: 0.000029 - momentum: 0.000000
2023-10-25 20:58:36,085 epoch 5 - iter 126/146 - loss 0.03369232 - time (sec): 8.10 - samples/sec: 4762.98 - lr: 0.000029 - momentum: 0.000000
2023-10-25 20:58:36,989 epoch 5 - iter 140/146 - loss 0.03255683 - time (sec): 9.00 - samples/sec: 4784.99 - lr: 0.000028 - momentum: 0.000000
2023-10-25 20:58:37,319 ----------------------------------------------------------------------------------------------------
2023-10-25 20:58:37,320 EPOCH 5 done: loss 0.0329 - lr: 0.000028
2023-10-25 20:58:38,235 DEV : loss 0.13207438588142395 - f1-score (micro avg) 0.7431
2023-10-25 20:58:38,239 ----------------------------------------------------------------------------------------------------
2023-10-25 20:58:39,294 epoch 6 - iter 14/146 - loss 0.01695521 - time (sec): 1.05 - samples/sec: 4951.89 - lr: 0.000027 - momentum: 0.000000
2023-10-25 20:58:40,188 epoch 6 - iter 28/146 - loss 0.02311012 - time (sec): 1.95 - samples/sec: 4724.81 - lr: 0.000027 - momentum: 0.000000
2023-10-25 20:58:41,100 epoch 6 - iter 42/146 - loss 0.02372267 - time (sec): 2.86 - samples/sec: 4704.55 - lr: 0.000026 - momentum: 0.000000
2023-10-25 20:58:42,262 epoch 6 - iter 56/146 - loss 0.02395132 - time (sec): 4.02 - samples/sec: 4535.45 - lr: 0.000026 - momentum: 0.000000
2023-10-25 20:58:43,220 epoch 6 - iter 70/146 - loss 0.02718854 - time (sec): 4.98 - samples/sec: 4455.30 - lr: 0.000025 - momentum: 0.000000
2023-10-25 20:58:44,250 epoch 6 - iter 84/146 - loss 0.02461396 - time (sec): 6.01 - samples/sec: 4453.46 - lr: 0.000025 - momentum: 0.000000
2023-10-25 20:58:45,088 epoch 6 - iter 98/146 - loss 0.02420481 - time (sec): 6.85 - samples/sec: 4489.02 - lr: 0.000024 - momentum: 0.000000
2023-10-25 20:58:45,977 epoch 6 - iter 112/146 - loss 0.02314700 - time (sec): 7.74 - samples/sec: 4520.31 - lr: 0.000024 - momentum: 0.000000
2023-10-25 20:58:46,885 epoch 6 - iter 126/146 - loss 0.02490428 - time (sec): 8.64 - samples/sec: 4577.35 - lr: 0.000023 - momentum: 0.000000
2023-10-25 20:58:47,670 epoch 6 - iter 140/146 - loss 0.02435065 - time (sec): 9.43 - samples/sec: 4547.67 - lr: 0.000023 - momentum: 0.000000
2023-10-25 20:58:48,021 ----------------------------------------------------------------------------------------------------
2023-10-25 20:58:48,021 EPOCH 6 done: loss 0.0237 - lr: 0.000023
2023-10-25 20:58:48,942 DEV : loss 0.144234299659729 - f1-score (micro avg) 0.7811
2023-10-25 20:58:48,946 saving best model
2023-10-25 20:58:49,550 ----------------------------------------------------------------------------------------------------
2023-10-25 20:58:50,465 epoch 7 - iter 14/146 - loss 0.02913716 - time (sec): 0.91 - samples/sec: 4483.73 - lr: 0.000022 - momentum: 0.000000
2023-10-25 20:58:51,312 epoch 7 - iter 28/146 - loss 0.01908361 - time (sec): 1.76 - samples/sec: 4764.35 - lr: 0.000021 - momentum: 0.000000
2023-10-25 20:58:52,484 epoch 7 - iter 42/146 - loss 0.01671570 - time (sec): 2.93 - samples/sec: 4581.70 - lr: 0.000021 - momentum: 0.000000
2023-10-25 20:58:53,411 epoch 7 - iter 56/146 - loss 0.01520409 - time (sec): 3.86 - samples/sec: 4584.57 - lr: 0.000020 - momentum: 0.000000
2023-10-25 20:58:54,229 epoch 7 - iter 70/146 - loss 0.01433610 - time (sec): 4.67 - samples/sec: 4661.75 - lr: 0.000020 - momentum: 0.000000
2023-10-25 20:58:55,224 epoch 7 - iter 84/146 - loss 0.01388687 - time (sec): 5.67 - samples/sec: 4643.06 - lr: 0.000019 - momentum: 0.000000
2023-10-25 20:58:56,081 epoch 7 - iter 98/146 - loss 0.01544967 - time (sec): 6.53 - samples/sec: 4644.60 - lr: 0.000019 - momentum: 0.000000
2023-10-25 20:58:56,956 epoch 7 - iter 112/146 - loss 0.01505549 - time (sec): 7.40 - samples/sec: 4623.87 - lr: 0.000018 - momentum: 0.000000
2023-10-25 20:58:57,784 epoch 7 - iter 126/146 - loss 0.01601917 - time (sec): 8.23 - samples/sec: 4633.05 - lr: 0.000018 - momentum: 0.000000
2023-10-25 20:58:58,651 epoch 7 - iter 140/146 - loss 0.01602666 - time (sec): 9.10 - samples/sec: 4709.87 - lr: 0.000017 - momentum: 0.000000
2023-10-25 20:58:59,008 ----------------------------------------------------------------------------------------------------
2023-10-25 20:58:59,009 EPOCH 7 done: loss 0.0161 - lr: 0.000017
2023-10-25 20:58:59,927 DEV : loss 0.1582878828048706 - f1-score (micro avg) 0.7458
2023-10-25 20:58:59,932 ----------------------------------------------------------------------------------------------------
2023-10-25 20:59:00,830 epoch 8 - iter 14/146 - loss 0.00430692 - time (sec): 0.90 - samples/sec: 4845.41 - lr: 0.000016 - momentum: 0.000000
2023-10-25 20:59:01,630 epoch 8 - iter 28/146 - loss 0.00918028 - time (sec): 1.70 - samples/sec: 4999.65 - lr: 0.000016 - momentum: 0.000000
2023-10-25 20:59:02,471 epoch 8 - iter 42/146 - loss 0.01094742 - time (sec): 2.54 - samples/sec: 4764.20 - lr: 0.000015 - momentum: 0.000000
2023-10-25 20:59:03,356 epoch 8 - iter 56/146 - loss 0.01000401 - time (sec): 3.42 - samples/sec: 4687.03 - lr: 0.000015 - momentum: 0.000000
2023-10-25 20:59:04,189 epoch 8 - iter 70/146 - loss 0.01022771 - time (sec): 4.26 - samples/sec: 4830.38 - lr: 0.000014 - momentum: 0.000000
2023-10-25 20:59:05,109 epoch 8 - iter 84/146 - loss 0.01016162 - time (sec): 5.18 - samples/sec: 4856.82 - lr: 0.000014 - momentum: 0.000000
2023-10-25 20:59:05,996 epoch 8 - iter 98/146 - loss 0.01079623 - time (sec): 6.06 - samples/sec: 4861.19 - lr: 0.000013 - momentum: 0.000000
2023-10-25 20:59:06,993 epoch 8 - iter 112/146 - loss 0.00990041 - time (sec): 7.06 - samples/sec: 4791.22 - lr: 0.000013 - momentum: 0.000000
2023-10-25 20:59:08,025 epoch 8 - iter 126/146 - loss 0.00935641 - time (sec): 8.09 - samples/sec: 4816.07 - lr: 0.000012 - momentum: 0.000000
2023-10-25 20:59:08,886 epoch 8 - iter 140/146 - loss 0.00928960 - time (sec): 8.95 - samples/sec: 4776.76 - lr: 0.000012 - momentum: 0.000000
2023-10-25 20:59:09,321 ----------------------------------------------------------------------------------------------------
2023-10-25 20:59:09,321 EPOCH 8 done: loss 0.0107 - lr: 0.000012
2023-10-25 20:59:10,237 DEV : loss 0.17408005893230438 - f1-score (micro avg) 0.7442
2023-10-25 20:59:10,241 ----------------------------------------------------------------------------------------------------
2023-10-25 20:59:11,090 epoch 9 - iter 14/146 - loss 0.00897464 - time (sec): 0.85 - samples/sec: 4664.55 - lr: 0.000011 - momentum: 0.000000
2023-10-25 20:59:12,014 epoch 9 - iter 28/146 - loss 0.01043103 - time (sec): 1.77 - samples/sec: 4651.90 - lr: 0.000010 - momentum: 0.000000
2023-10-25 20:59:12,849 epoch 9 - iter 42/146 - loss 0.01054132 - time (sec): 2.61 - samples/sec: 4743.05 - lr: 0.000010 - momentum: 0.000000
2023-10-25 20:59:13,741 epoch 9 - iter 56/146 - loss 0.00869160 - time (sec): 3.50 - samples/sec: 4620.65 - lr: 0.000009 - momentum: 0.000000
2023-10-25 20:59:14,641 epoch 9 - iter 70/146 - loss 0.00705254 - time (sec): 4.40 - samples/sec: 4619.49 - lr: 0.000009 - momentum: 0.000000
2023-10-25 20:59:15,607 epoch 9 - iter 84/146 - loss 0.00677544 - time (sec): 5.36 - samples/sec: 4692.36 - lr: 0.000008 - momentum: 0.000000
2023-10-25 20:59:16,540 epoch 9 - iter 98/146 - loss 0.00713812 - time (sec): 6.30 - samples/sec: 4716.11 - lr: 0.000008 - momentum: 0.000000
2023-10-25 20:59:17,422 epoch 9 - iter 112/146 - loss 0.01059142 - time (sec): 7.18 - samples/sec: 4714.65 - lr: 0.000007 - momentum: 0.000000
2023-10-25 20:59:18,563 epoch 9 - iter 126/146 - loss 0.00958536 - time (sec): 8.32 - samples/sec: 4608.11 - lr: 0.000007 - momentum: 0.000000
2023-10-25 20:59:19,488 epoch 9 - iter 140/146 - loss 0.01016471 - time (sec): 9.25 - samples/sec: 4617.10 - lr: 0.000006 - momentum: 0.000000
2023-10-25 20:59:19,835 ----------------------------------------------------------------------------------------------------
2023-10-25 20:59:19,835 EPOCH 9 done: loss 0.0098 - lr: 0.000006
2023-10-25 20:59:20,751 DEV : loss 0.1856423020362854 - f1-score (micro avg) 0.7553
2023-10-25 20:59:20,755 ----------------------------------------------------------------------------------------------------
2023-10-25 20:59:21,541 epoch 10 - iter 14/146 - loss 0.00410973 - time (sec): 0.78 - samples/sec: 4832.34 - lr: 0.000005 - momentum: 0.000000
2023-10-25 20:59:22,346 epoch 10 - iter 28/146 - loss 0.00363689 - time (sec): 1.59 - samples/sec: 4706.27 - lr: 0.000005 - momentum: 0.000000
2023-10-25 20:59:23,198 epoch 10 - iter 42/146 - loss 0.00418360 - time (sec): 2.44 - samples/sec: 4664.27 - lr: 0.000004 - momentum: 0.000000
2023-10-25 20:59:24,273 epoch 10 - iter 56/146 - loss 0.00693018 - time (sec): 3.52 - samples/sec: 4809.94 - lr: 0.000004 - momentum: 0.000000
2023-10-25 20:59:25,255 epoch 10 - iter 70/146 - loss 0.00605639 - time (sec): 4.50 - samples/sec: 4743.21 - lr: 0.000003 - momentum: 0.000000
2023-10-25 20:59:26,236 epoch 10 - iter 84/146 - loss 0.00595401 - time (sec): 5.48 - samples/sec: 4846.88 - lr: 0.000003 - momentum: 0.000000
2023-10-25 20:59:27,182 epoch 10 - iter 98/146 - loss 0.00546547 - time (sec): 6.43 - samples/sec: 4870.19 - lr: 0.000002 - momentum: 0.000000
2023-10-25 20:59:27,970 epoch 10 - iter 112/146 - loss 0.00649974 - time (sec): 7.21 - samples/sec: 4852.31 - lr: 0.000002 - momentum: 0.000000
2023-10-25 20:59:28,775 epoch 10 - iter 126/146 - loss 0.00591971 - time (sec): 8.02 - samples/sec: 4824.36 - lr: 0.000001 - momentum: 0.000000
2023-10-25 20:59:29,578 epoch 10 - iter 140/146 - loss 0.00621200 - time (sec): 8.82 - samples/sec: 4858.39 - lr: 0.000000 - momentum: 0.000000
2023-10-25 20:59:29,909 ----------------------------------------------------------------------------------------------------
2023-10-25 20:59:29,910 EPOCH 10 done: loss 0.0063 - lr: 0.000000
2023-10-25 20:59:30,826 DEV : loss 0.18521195650100708 - f1-score (micro avg) 0.7611
2023-10-25 20:59:31,305 ----------------------------------------------------------------------------------------------------
2023-10-25 20:59:31,306 Loading model from best epoch ...
2023-10-25 20:59:32,867 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
2023-10-25 20:59:34,407
Results:
- F-score (micro) 0.7399
- F-score (macro) 0.6807
- Accuracy 0.6134
By class:
precision recall f1-score support
PER 0.7945 0.8333 0.8135 348
LOC 0.6228 0.8161 0.7065 261
ORG 0.4203 0.5577 0.4793 52
HumanProd 0.6800 0.7727 0.7234 22
micro avg 0.6854 0.8038 0.7399 683
macro avg 0.6294 0.7450 0.6807 683
weighted avg 0.6967 0.8038 0.7442 683
2023-10-25 20:59:34,407 ----------------------------------------------------------------------------------------------------
|