File size: 23,982 Bytes
a4edbe8 |
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 |
2023-10-17 10:25:39,666 ----------------------------------------------------------------------------------------------------
2023-10-17 10:25:39,667 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=13, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-17 10:25:39,668 ----------------------------------------------------------------------------------------------------
2023-10-17 10:25:39,668 MultiCorpus: 6183 train + 680 dev + 2113 test sentences
- NER_HIPE_2022 Corpus: 6183 train + 680 dev + 2113 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/topres19th/en/with_doc_seperator
2023-10-17 10:25:39,668 ----------------------------------------------------------------------------------------------------
2023-10-17 10:25:39,668 Train: 6183 sentences
2023-10-17 10:25:39,668 (train_with_dev=False, train_with_test=False)
2023-10-17 10:25:39,668 ----------------------------------------------------------------------------------------------------
2023-10-17 10:25:39,668 Training Params:
2023-10-17 10:25:39,668 - learning_rate: "5e-05"
2023-10-17 10:25:39,668 - mini_batch_size: "4"
2023-10-17 10:25:39,668 - max_epochs: "10"
2023-10-17 10:25:39,668 - shuffle: "True"
2023-10-17 10:25:39,668 ----------------------------------------------------------------------------------------------------
2023-10-17 10:25:39,668 Plugins:
2023-10-17 10:25:39,669 - TensorboardLogger
2023-10-17 10:25:39,669 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 10:25:39,669 ----------------------------------------------------------------------------------------------------
2023-10-17 10:25:39,669 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 10:25:39,669 - metric: "('micro avg', 'f1-score')"
2023-10-17 10:25:39,669 ----------------------------------------------------------------------------------------------------
2023-10-17 10:25:39,669 Computation:
2023-10-17 10:25:39,669 - compute on device: cuda:0
2023-10-17 10:25:39,669 - embedding storage: none
2023-10-17 10:25:39,669 ----------------------------------------------------------------------------------------------------
2023-10-17 10:25:39,669 Model training base path: "hmbench-topres19th/en-hmteams/teams-base-historic-multilingual-discriminator-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-2"
2023-10-17 10:25:39,669 ----------------------------------------------------------------------------------------------------
2023-10-17 10:25:39,669 ----------------------------------------------------------------------------------------------------
2023-10-17 10:25:39,669 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 10:25:51,632 epoch 1 - iter 154/1546 - loss 2.02180981 - time (sec): 11.96 - samples/sec: 944.73 - lr: 0.000005 - momentum: 0.000000
2023-10-17 10:26:04,276 epoch 1 - iter 308/1546 - loss 1.05133169 - time (sec): 24.60 - samples/sec: 1008.11 - lr: 0.000010 - momentum: 0.000000
2023-10-17 10:26:16,281 epoch 1 - iter 462/1546 - loss 0.75054260 - time (sec): 36.61 - samples/sec: 1031.68 - lr: 0.000015 - momentum: 0.000000
2023-10-17 10:26:28,650 epoch 1 - iter 616/1546 - loss 0.59271900 - time (sec): 48.98 - samples/sec: 1032.05 - lr: 0.000020 - momentum: 0.000000
2023-10-17 10:26:41,225 epoch 1 - iter 770/1546 - loss 0.49705880 - time (sec): 61.55 - samples/sec: 1028.74 - lr: 0.000025 - momentum: 0.000000
2023-10-17 10:26:53,393 epoch 1 - iter 924/1546 - loss 0.43170337 - time (sec): 73.72 - samples/sec: 1033.06 - lr: 0.000030 - momentum: 0.000000
2023-10-17 10:27:05,348 epoch 1 - iter 1078/1546 - loss 0.39066886 - time (sec): 85.68 - samples/sec: 1021.48 - lr: 0.000035 - momentum: 0.000000
2023-10-17 10:27:17,415 epoch 1 - iter 1232/1546 - loss 0.35716844 - time (sec): 97.74 - samples/sec: 1020.75 - lr: 0.000040 - momentum: 0.000000
2023-10-17 10:27:29,146 epoch 1 - iter 1386/1546 - loss 0.33123533 - time (sec): 109.47 - samples/sec: 1026.36 - lr: 0.000045 - momentum: 0.000000
2023-10-17 10:27:40,598 epoch 1 - iter 1540/1546 - loss 0.31206315 - time (sec): 120.93 - samples/sec: 1024.74 - lr: 0.000050 - momentum: 0.000000
2023-10-17 10:27:41,031 ----------------------------------------------------------------------------------------------------
2023-10-17 10:27:41,031 EPOCH 1 done: loss 0.3115 - lr: 0.000050
2023-10-17 10:27:43,666 DEV : loss 0.07279833406209946 - f1-score (micro avg) 0.7049
2023-10-17 10:27:43,693 saving best model
2023-10-17 10:27:44,227 ----------------------------------------------------------------------------------------------------
2023-10-17 10:27:55,853 epoch 2 - iter 154/1546 - loss 0.17648886 - time (sec): 11.62 - samples/sec: 1047.19 - lr: 0.000049 - momentum: 0.000000
2023-10-17 10:28:07,775 epoch 2 - iter 308/1546 - loss 0.15188272 - time (sec): 23.55 - samples/sec: 1070.06 - lr: 0.000049 - momentum: 0.000000
2023-10-17 10:28:19,828 epoch 2 - iter 462/1546 - loss 0.13981404 - time (sec): 35.60 - samples/sec: 1037.47 - lr: 0.000048 - momentum: 0.000000
2023-10-17 10:28:32,332 epoch 2 - iter 616/1546 - loss 0.12814078 - time (sec): 48.10 - samples/sec: 1024.94 - lr: 0.000048 - momentum: 0.000000
2023-10-17 10:28:44,172 epoch 2 - iter 770/1546 - loss 0.12573561 - time (sec): 59.94 - samples/sec: 1022.59 - lr: 0.000047 - momentum: 0.000000
2023-10-17 10:28:56,254 epoch 2 - iter 924/1546 - loss 0.12073221 - time (sec): 72.03 - samples/sec: 1017.97 - lr: 0.000047 - momentum: 0.000000
2023-10-17 10:29:09,100 epoch 2 - iter 1078/1546 - loss 0.11547940 - time (sec): 84.87 - samples/sec: 1013.64 - lr: 0.000046 - momentum: 0.000000
2023-10-17 10:29:20,955 epoch 2 - iter 1232/1546 - loss 0.11335352 - time (sec): 96.73 - samples/sec: 1017.70 - lr: 0.000046 - momentum: 0.000000
2023-10-17 10:29:32,654 epoch 2 - iter 1386/1546 - loss 0.11157211 - time (sec): 108.42 - samples/sec: 1024.63 - lr: 0.000045 - momentum: 0.000000
2023-10-17 10:29:44,656 epoch 2 - iter 1540/1546 - loss 0.10827885 - time (sec): 120.43 - samples/sec: 1029.36 - lr: 0.000044 - momentum: 0.000000
2023-10-17 10:29:45,108 ----------------------------------------------------------------------------------------------------
2023-10-17 10:29:45,108 EPOCH 2 done: loss 0.1081 - lr: 0.000044
2023-10-17 10:29:48,007 DEV : loss 0.05769108608365059 - f1-score (micro avg) 0.7859
2023-10-17 10:29:48,040 saving best model
2023-10-17 10:29:49,482 ----------------------------------------------------------------------------------------------------
2023-10-17 10:30:01,407 epoch 3 - iter 154/1546 - loss 0.09035284 - time (sec): 11.92 - samples/sec: 985.24 - lr: 0.000044 - momentum: 0.000000
2023-10-17 10:30:12,820 epoch 3 - iter 308/1546 - loss 0.07445859 - time (sec): 23.33 - samples/sec: 1097.98 - lr: 0.000043 - momentum: 0.000000
2023-10-17 10:30:24,703 epoch 3 - iter 462/1546 - loss 0.07415434 - time (sec): 35.22 - samples/sec: 1100.85 - lr: 0.000043 - momentum: 0.000000
2023-10-17 10:30:37,609 epoch 3 - iter 616/1546 - loss 0.07540870 - time (sec): 48.12 - samples/sec: 1045.62 - lr: 0.000042 - momentum: 0.000000
2023-10-17 10:30:50,460 epoch 3 - iter 770/1546 - loss 0.07301929 - time (sec): 60.97 - samples/sec: 1021.49 - lr: 0.000042 - momentum: 0.000000
2023-10-17 10:31:04,196 epoch 3 - iter 924/1546 - loss 0.07303987 - time (sec): 74.71 - samples/sec: 1004.25 - lr: 0.000041 - momentum: 0.000000
2023-10-17 10:31:15,649 epoch 3 - iter 1078/1546 - loss 0.07020216 - time (sec): 86.16 - samples/sec: 1010.69 - lr: 0.000041 - momentum: 0.000000
2023-10-17 10:31:27,261 epoch 3 - iter 1232/1546 - loss 0.07010983 - time (sec): 97.78 - samples/sec: 1016.90 - lr: 0.000040 - momentum: 0.000000
2023-10-17 10:31:39,005 epoch 3 - iter 1386/1546 - loss 0.06794363 - time (sec): 109.52 - samples/sec: 1025.10 - lr: 0.000039 - momentum: 0.000000
2023-10-17 10:31:50,746 epoch 3 - iter 1540/1546 - loss 0.06707353 - time (sec): 121.26 - samples/sec: 1021.84 - lr: 0.000039 - momentum: 0.000000
2023-10-17 10:31:51,242 ----------------------------------------------------------------------------------------------------
2023-10-17 10:31:51,242 EPOCH 3 done: loss 0.0670 - lr: 0.000039
2023-10-17 10:31:54,247 DEV : loss 0.07943776994943619 - f1-score (micro avg) 0.8114
2023-10-17 10:31:54,285 saving best model
2023-10-17 10:31:55,701 ----------------------------------------------------------------------------------------------------
2023-10-17 10:32:07,901 epoch 4 - iter 154/1546 - loss 0.04828653 - time (sec): 12.20 - samples/sec: 1063.56 - lr: 0.000038 - momentum: 0.000000
2023-10-17 10:32:20,064 epoch 4 - iter 308/1546 - loss 0.05173450 - time (sec): 24.36 - samples/sec: 1039.44 - lr: 0.000038 - momentum: 0.000000
2023-10-17 10:32:32,324 epoch 4 - iter 462/1546 - loss 0.04954644 - time (sec): 36.62 - samples/sec: 1016.86 - lr: 0.000037 - momentum: 0.000000
2023-10-17 10:32:44,801 epoch 4 - iter 616/1546 - loss 0.05004385 - time (sec): 49.10 - samples/sec: 1003.51 - lr: 0.000037 - momentum: 0.000000
2023-10-17 10:32:56,865 epoch 4 - iter 770/1546 - loss 0.04750404 - time (sec): 61.16 - samples/sec: 1016.02 - lr: 0.000036 - momentum: 0.000000
2023-10-17 10:33:09,083 epoch 4 - iter 924/1546 - loss 0.04870217 - time (sec): 73.38 - samples/sec: 1014.04 - lr: 0.000036 - momentum: 0.000000
2023-10-17 10:33:20,791 epoch 4 - iter 1078/1546 - loss 0.04965165 - time (sec): 85.09 - samples/sec: 1018.08 - lr: 0.000035 - momentum: 0.000000
2023-10-17 10:33:32,692 epoch 4 - iter 1232/1546 - loss 0.04736677 - time (sec): 96.99 - samples/sec: 1026.67 - lr: 0.000034 - momentum: 0.000000
2023-10-17 10:33:44,844 epoch 4 - iter 1386/1546 - loss 0.04811731 - time (sec): 109.14 - samples/sec: 1026.04 - lr: 0.000034 - momentum: 0.000000
2023-10-17 10:33:57,129 epoch 4 - iter 1540/1546 - loss 0.04803105 - time (sec): 121.42 - samples/sec: 1019.68 - lr: 0.000033 - momentum: 0.000000
2023-10-17 10:33:57,607 ----------------------------------------------------------------------------------------------------
2023-10-17 10:33:57,608 EPOCH 4 done: loss 0.0481 - lr: 0.000033
2023-10-17 10:34:00,432 DEV : loss 0.0833667516708374 - f1-score (micro avg) 0.7711
2023-10-17 10:34:00,460 ----------------------------------------------------------------------------------------------------
2023-10-17 10:34:12,396 epoch 5 - iter 154/1546 - loss 0.02385740 - time (sec): 11.93 - samples/sec: 1024.45 - lr: 0.000033 - momentum: 0.000000
2023-10-17 10:34:24,347 epoch 5 - iter 308/1546 - loss 0.02748508 - time (sec): 23.89 - samples/sec: 1001.54 - lr: 0.000032 - momentum: 0.000000
2023-10-17 10:34:36,363 epoch 5 - iter 462/1546 - loss 0.02582250 - time (sec): 35.90 - samples/sec: 995.55 - lr: 0.000032 - momentum: 0.000000
2023-10-17 10:34:48,343 epoch 5 - iter 616/1546 - loss 0.02616104 - time (sec): 47.88 - samples/sec: 1005.06 - lr: 0.000031 - momentum: 0.000000
2023-10-17 10:35:00,744 epoch 5 - iter 770/1546 - loss 0.03219186 - time (sec): 60.28 - samples/sec: 1003.30 - lr: 0.000031 - momentum: 0.000000
2023-10-17 10:35:12,809 epoch 5 - iter 924/1546 - loss 0.03235715 - time (sec): 72.35 - samples/sec: 1007.34 - lr: 0.000030 - momentum: 0.000000
2023-10-17 10:35:24,651 epoch 5 - iter 1078/1546 - loss 0.03225049 - time (sec): 84.19 - samples/sec: 1029.36 - lr: 0.000029 - momentum: 0.000000
2023-10-17 10:35:36,841 epoch 5 - iter 1232/1546 - loss 0.03312944 - time (sec): 96.38 - samples/sec: 1028.66 - lr: 0.000029 - momentum: 0.000000
2023-10-17 10:35:49,293 epoch 5 - iter 1386/1546 - loss 0.03394222 - time (sec): 108.83 - samples/sec: 1020.22 - lr: 0.000028 - momentum: 0.000000
2023-10-17 10:36:01,342 epoch 5 - iter 1540/1546 - loss 0.03495607 - time (sec): 120.88 - samples/sec: 1024.69 - lr: 0.000028 - momentum: 0.000000
2023-10-17 10:36:01,814 ----------------------------------------------------------------------------------------------------
2023-10-17 10:36:01,814 EPOCH 5 done: loss 0.0350 - lr: 0.000028
2023-10-17 10:36:05,180 DEV : loss 0.110050730407238 - f1-score (micro avg) 0.7099
2023-10-17 10:36:05,213 ----------------------------------------------------------------------------------------------------
2023-10-17 10:36:17,215 epoch 6 - iter 154/1546 - loss 0.03050029 - time (sec): 12.00 - samples/sec: 1034.82 - lr: 0.000027 - momentum: 0.000000
2023-10-17 10:36:29,436 epoch 6 - iter 308/1546 - loss 0.02599671 - time (sec): 24.22 - samples/sec: 1030.57 - lr: 0.000027 - momentum: 0.000000
2023-10-17 10:36:41,414 epoch 6 - iter 462/1546 - loss 0.02421473 - time (sec): 36.20 - samples/sec: 1038.52 - lr: 0.000026 - momentum: 0.000000
2023-10-17 10:36:53,413 epoch 6 - iter 616/1546 - loss 0.02336062 - time (sec): 48.20 - samples/sec: 1040.54 - lr: 0.000026 - momentum: 0.000000
2023-10-17 10:37:05,424 epoch 6 - iter 770/1546 - loss 0.02547084 - time (sec): 60.21 - samples/sec: 1032.66 - lr: 0.000025 - momentum: 0.000000
2023-10-17 10:37:17,945 epoch 6 - iter 924/1546 - loss 0.02397867 - time (sec): 72.73 - samples/sec: 1024.75 - lr: 0.000024 - momentum: 0.000000
2023-10-17 10:37:29,949 epoch 6 - iter 1078/1546 - loss 0.02277072 - time (sec): 84.73 - samples/sec: 1036.78 - lr: 0.000024 - momentum: 0.000000
2023-10-17 10:37:41,864 epoch 6 - iter 1232/1546 - loss 0.02426498 - time (sec): 96.65 - samples/sec: 1029.50 - lr: 0.000023 - momentum: 0.000000
2023-10-17 10:37:53,712 epoch 6 - iter 1386/1546 - loss 0.02385236 - time (sec): 108.50 - samples/sec: 1028.24 - lr: 0.000023 - momentum: 0.000000
2023-10-17 10:38:05,568 epoch 6 - iter 1540/1546 - loss 0.02402382 - time (sec): 120.35 - samples/sec: 1028.83 - lr: 0.000022 - momentum: 0.000000
2023-10-17 10:38:06,019 ----------------------------------------------------------------------------------------------------
2023-10-17 10:38:06,019 EPOCH 6 done: loss 0.0241 - lr: 0.000022
2023-10-17 10:38:08,835 DEV : loss 0.09071236848831177 - f1-score (micro avg) 0.7718
2023-10-17 10:38:08,863 ----------------------------------------------------------------------------------------------------
2023-10-17 10:38:20,993 epoch 7 - iter 154/1546 - loss 0.01858194 - time (sec): 12.13 - samples/sec: 1035.21 - lr: 0.000022 - momentum: 0.000000
2023-10-17 10:38:33,142 epoch 7 - iter 308/1546 - loss 0.02029265 - time (sec): 24.28 - samples/sec: 1076.84 - lr: 0.000021 - momentum: 0.000000
2023-10-17 10:38:44,918 epoch 7 - iter 462/1546 - loss 0.01874414 - time (sec): 36.05 - samples/sec: 1064.32 - lr: 0.000021 - momentum: 0.000000
2023-10-17 10:38:56,851 epoch 7 - iter 616/1546 - loss 0.02012904 - time (sec): 47.99 - samples/sec: 1052.19 - lr: 0.000020 - momentum: 0.000000
2023-10-17 10:39:08,577 epoch 7 - iter 770/1546 - loss 0.02056826 - time (sec): 59.71 - samples/sec: 1049.28 - lr: 0.000019 - momentum: 0.000000
2023-10-17 10:39:20,396 epoch 7 - iter 924/1546 - loss 0.01990787 - time (sec): 71.53 - samples/sec: 1037.05 - lr: 0.000019 - momentum: 0.000000
2023-10-17 10:39:32,595 epoch 7 - iter 1078/1546 - loss 0.02052998 - time (sec): 83.73 - samples/sec: 1028.28 - lr: 0.000018 - momentum: 0.000000
2023-10-17 10:39:44,797 epoch 7 - iter 1232/1546 - loss 0.01888219 - time (sec): 95.93 - samples/sec: 1030.99 - lr: 0.000018 - momentum: 0.000000
2023-10-17 10:39:56,820 epoch 7 - iter 1386/1546 - loss 0.01791790 - time (sec): 107.95 - samples/sec: 1035.40 - lr: 0.000017 - momentum: 0.000000
2023-10-17 10:40:09,502 epoch 7 - iter 1540/1546 - loss 0.01840040 - time (sec): 120.64 - samples/sec: 1027.33 - lr: 0.000017 - momentum: 0.000000
2023-10-17 10:40:10,068 ----------------------------------------------------------------------------------------------------
2023-10-17 10:40:10,068 EPOCH 7 done: loss 0.0185 - lr: 0.000017
2023-10-17 10:40:13,083 DEV : loss 0.12031986564397812 - f1-score (micro avg) 0.7705
2023-10-17 10:40:13,114 ----------------------------------------------------------------------------------------------------
2023-10-17 10:40:25,169 epoch 8 - iter 154/1546 - loss 0.01247555 - time (sec): 12.05 - samples/sec: 1025.95 - lr: 0.000016 - momentum: 0.000000
2023-10-17 10:40:37,477 epoch 8 - iter 308/1546 - loss 0.00887144 - time (sec): 24.36 - samples/sec: 1038.46 - lr: 0.000016 - momentum: 0.000000
2023-10-17 10:40:49,680 epoch 8 - iter 462/1546 - loss 0.00915473 - time (sec): 36.56 - samples/sec: 1015.25 - lr: 0.000015 - momentum: 0.000000
2023-10-17 10:41:02,185 epoch 8 - iter 616/1546 - loss 0.01189933 - time (sec): 49.07 - samples/sec: 1008.89 - lr: 0.000014 - momentum: 0.000000
2023-10-17 10:41:14,290 epoch 8 - iter 770/1546 - loss 0.01107024 - time (sec): 61.17 - samples/sec: 1018.87 - lr: 0.000014 - momentum: 0.000000
2023-10-17 10:41:26,129 epoch 8 - iter 924/1546 - loss 0.01086037 - time (sec): 73.01 - samples/sec: 1027.72 - lr: 0.000013 - momentum: 0.000000
2023-10-17 10:41:38,145 epoch 8 - iter 1078/1546 - loss 0.01157159 - time (sec): 85.03 - samples/sec: 1024.92 - lr: 0.000013 - momentum: 0.000000
2023-10-17 10:41:49,914 epoch 8 - iter 1232/1546 - loss 0.01193759 - time (sec): 96.80 - samples/sec: 1024.57 - lr: 0.000012 - momentum: 0.000000
2023-10-17 10:42:02,221 epoch 8 - iter 1386/1546 - loss 0.01219084 - time (sec): 109.10 - samples/sec: 1023.58 - lr: 0.000012 - momentum: 0.000000
2023-10-17 10:42:13,995 epoch 8 - iter 1540/1546 - loss 0.01168510 - time (sec): 120.88 - samples/sec: 1023.56 - lr: 0.000011 - momentum: 0.000000
2023-10-17 10:42:14,461 ----------------------------------------------------------------------------------------------------
2023-10-17 10:42:14,461 EPOCH 8 done: loss 0.0117 - lr: 0.000011
2023-10-17 10:42:17,363 DEV : loss 0.11697618663311005 - f1-score (micro avg) 0.8008
2023-10-17 10:42:17,397 ----------------------------------------------------------------------------------------------------
2023-10-17 10:42:29,121 epoch 9 - iter 154/1546 - loss 0.01140863 - time (sec): 11.72 - samples/sec: 1045.07 - lr: 0.000011 - momentum: 0.000000
2023-10-17 10:42:40,816 epoch 9 - iter 308/1546 - loss 0.01020266 - time (sec): 23.42 - samples/sec: 1032.47 - lr: 0.000010 - momentum: 0.000000
2023-10-17 10:42:52,458 epoch 9 - iter 462/1546 - loss 0.00903518 - time (sec): 35.06 - samples/sec: 1077.02 - lr: 0.000009 - momentum: 0.000000
2023-10-17 10:43:04,002 epoch 9 - iter 616/1546 - loss 0.00887521 - time (sec): 46.60 - samples/sec: 1062.22 - lr: 0.000009 - momentum: 0.000000
2023-10-17 10:43:15,609 epoch 9 - iter 770/1546 - loss 0.00783009 - time (sec): 58.21 - samples/sec: 1056.47 - lr: 0.000008 - momentum: 0.000000
2023-10-17 10:43:28,043 epoch 9 - iter 924/1546 - loss 0.00793805 - time (sec): 70.64 - samples/sec: 1039.01 - lr: 0.000008 - momentum: 0.000000
2023-10-17 10:43:39,954 epoch 9 - iter 1078/1546 - loss 0.00743572 - time (sec): 82.55 - samples/sec: 1035.87 - lr: 0.000007 - momentum: 0.000000
2023-10-17 10:43:51,915 epoch 9 - iter 1232/1546 - loss 0.00757961 - time (sec): 94.52 - samples/sec: 1050.24 - lr: 0.000007 - momentum: 0.000000
2023-10-17 10:44:04,164 epoch 9 - iter 1386/1546 - loss 0.00743361 - time (sec): 106.76 - samples/sec: 1040.78 - lr: 0.000006 - momentum: 0.000000
2023-10-17 10:44:16,128 epoch 9 - iter 1540/1546 - loss 0.00756154 - time (sec): 118.73 - samples/sec: 1044.29 - lr: 0.000006 - momentum: 0.000000
2023-10-17 10:44:16,573 ----------------------------------------------------------------------------------------------------
2023-10-17 10:44:16,574 EPOCH 9 done: loss 0.0076 - lr: 0.000006
2023-10-17 10:44:19,531 DEV : loss 0.12403418123722076 - f1-score (micro avg) 0.7884
2023-10-17 10:44:19,562 ----------------------------------------------------------------------------------------------------
2023-10-17 10:44:31,581 epoch 10 - iter 154/1546 - loss 0.00254844 - time (sec): 12.02 - samples/sec: 1087.04 - lr: 0.000005 - momentum: 0.000000
2023-10-17 10:44:43,680 epoch 10 - iter 308/1546 - loss 0.00244110 - time (sec): 24.12 - samples/sec: 1018.42 - lr: 0.000004 - momentum: 0.000000
2023-10-17 10:44:55,700 epoch 10 - iter 462/1546 - loss 0.00305097 - time (sec): 36.14 - samples/sec: 1005.43 - lr: 0.000004 - momentum: 0.000000
2023-10-17 10:45:07,998 epoch 10 - iter 616/1546 - loss 0.00292169 - time (sec): 48.43 - samples/sec: 1012.33 - lr: 0.000003 - momentum: 0.000000
2023-10-17 10:45:20,453 epoch 10 - iter 770/1546 - loss 0.00318972 - time (sec): 60.89 - samples/sec: 1004.67 - lr: 0.000003 - momentum: 0.000000
2023-10-17 10:45:32,649 epoch 10 - iter 924/1546 - loss 0.00372386 - time (sec): 73.08 - samples/sec: 1001.91 - lr: 0.000002 - momentum: 0.000000
2023-10-17 10:45:44,939 epoch 10 - iter 1078/1546 - loss 0.00392698 - time (sec): 85.37 - samples/sec: 1008.88 - lr: 0.000002 - momentum: 0.000000
2023-10-17 10:45:57,060 epoch 10 - iter 1232/1546 - loss 0.00386062 - time (sec): 97.50 - samples/sec: 1028.14 - lr: 0.000001 - momentum: 0.000000
2023-10-17 10:46:09,103 epoch 10 - iter 1386/1546 - loss 0.00402196 - time (sec): 109.54 - samples/sec: 1020.51 - lr: 0.000001 - momentum: 0.000000
2023-10-17 10:46:20,872 epoch 10 - iter 1540/1546 - loss 0.00428314 - time (sec): 121.31 - samples/sec: 1021.41 - lr: 0.000000 - momentum: 0.000000
2023-10-17 10:46:21,321 ----------------------------------------------------------------------------------------------------
2023-10-17 10:46:21,322 EPOCH 10 done: loss 0.0043 - lr: 0.000000
2023-10-17 10:46:24,218 DEV : loss 0.1269896924495697 - f1-score (micro avg) 0.7866
2023-10-17 10:46:24,806 ----------------------------------------------------------------------------------------------------
2023-10-17 10:46:24,808 Loading model from best epoch ...
2023-10-17 10:46:27,155 SequenceTagger predicts: Dictionary with 13 tags: O, S-LOC, B-LOC, E-LOC, I-LOC, S-BUILDING, B-BUILDING, E-BUILDING, I-BUILDING, S-STREET, B-STREET, E-STREET, I-STREET
2023-10-17 10:46:36,273
Results:
- F-score (micro) 0.7831
- F-score (macro) 0.6698
- Accuracy 0.6639
By class:
precision recall f1-score support
LOC 0.7952 0.8700 0.8309 946
BUILDING 0.6107 0.4919 0.5449 185
STREET 0.7111 0.5714 0.6337 56
micro avg 0.7697 0.7970 0.7831 1187
macro avg 0.7057 0.6444 0.6698 1187
weighted avg 0.7625 0.7970 0.7770 1187
2023-10-17 10:46:36,274 ----------------------------------------------------------------------------------------------------
|