File size: 24,031 Bytes
2ef9e95 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 |
2023-10-18 16:36:23,556 ----------------------------------------------------------------------------------------------------
2023-10-18 16:36:23,556 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): BertModel(
(embeddings): BertEmbeddings(
(word_embeddings): Embedding(32001, 128)
(position_embeddings): Embedding(512, 128)
(token_type_embeddings): Embedding(2, 128)
(LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(encoder): BertEncoder(
(layer): ModuleList(
(0-1): 2 x BertLayer(
(attention): BertAttention(
(self): BertSelfAttention(
(query): Linear(in_features=128, out_features=128, bias=True)
(key): Linear(in_features=128, out_features=128, bias=True)
(value): Linear(in_features=128, out_features=128, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(output): BertSelfOutput(
(dense): Linear(in_features=128, out_features=128, bias=True)
(LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(intermediate): BertIntermediate(
(dense): Linear(in_features=128, out_features=512, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): BertOutput(
(dense): Linear(in_features=512, out_features=128, bias=True)
(LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
(pooler): BertPooler(
(dense): Linear(in_features=128, out_features=128, bias=True)
(activation): Tanh()
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=128, out_features=25, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-18 16:36:23,556 ----------------------------------------------------------------------------------------------------
2023-10-18 16:36:23,556 MultiCorpus: 966 train + 219 dev + 204 test sentences
- NER_HIPE_2022 Corpus: 966 train + 219 dev + 204 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/ajmc/fr/with_doc_seperator
2023-10-18 16:36:23,556 ----------------------------------------------------------------------------------------------------
2023-10-18 16:36:23,556 Train: 966 sentences
2023-10-18 16:36:23,556 (train_with_dev=False, train_with_test=False)
2023-10-18 16:36:23,556 ----------------------------------------------------------------------------------------------------
2023-10-18 16:36:23,556 Training Params:
2023-10-18 16:36:23,557 - learning_rate: "5e-05"
2023-10-18 16:36:23,557 - mini_batch_size: "4"
2023-10-18 16:36:23,557 - max_epochs: "10"
2023-10-18 16:36:23,557 - shuffle: "True"
2023-10-18 16:36:23,557 ----------------------------------------------------------------------------------------------------
2023-10-18 16:36:23,557 Plugins:
2023-10-18 16:36:23,557 - TensorboardLogger
2023-10-18 16:36:23,557 - LinearScheduler | warmup_fraction: '0.1'
2023-10-18 16:36:23,557 ----------------------------------------------------------------------------------------------------
2023-10-18 16:36:23,557 Final evaluation on model from best epoch (best-model.pt)
2023-10-18 16:36:23,557 - metric: "('micro avg', 'f1-score')"
2023-10-18 16:36:23,557 ----------------------------------------------------------------------------------------------------
2023-10-18 16:36:23,557 Computation:
2023-10-18 16:36:23,557 - compute on device: cuda:0
2023-10-18 16:36:23,557 - embedding storage: none
2023-10-18 16:36:23,557 ----------------------------------------------------------------------------------------------------
2023-10-18 16:36:23,557 Model training base path: "hmbench-ajmc/fr-dbmdz/bert-tiny-historic-multilingual-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-1"
2023-10-18 16:36:23,557 ----------------------------------------------------------------------------------------------------
2023-10-18 16:36:23,557 ----------------------------------------------------------------------------------------------------
2023-10-18 16:36:23,557 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-18 16:36:23,903 epoch 1 - iter 24/242 - loss 4.06894328 - time (sec): 0.35 - samples/sec: 7303.90 - lr: 0.000005 - momentum: 0.000000
2023-10-18 16:36:24,299 epoch 1 - iter 48/242 - loss 4.02616879 - time (sec): 0.74 - samples/sec: 6639.93 - lr: 0.000010 - momentum: 0.000000
2023-10-18 16:36:24,687 epoch 1 - iter 72/242 - loss 3.82600153 - time (sec): 1.13 - samples/sec: 6689.58 - lr: 0.000015 - momentum: 0.000000
2023-10-18 16:36:25,057 epoch 1 - iter 96/242 - loss 3.59033518 - time (sec): 1.50 - samples/sec: 6922.07 - lr: 0.000020 - momentum: 0.000000
2023-10-18 16:36:25,433 epoch 1 - iter 120/242 - loss 3.38518381 - time (sec): 1.88 - samples/sec: 6651.40 - lr: 0.000025 - momentum: 0.000000
2023-10-18 16:36:25,802 epoch 1 - iter 144/242 - loss 3.11809328 - time (sec): 2.24 - samples/sec: 6616.40 - lr: 0.000030 - momentum: 0.000000
2023-10-18 16:36:26,173 epoch 1 - iter 168/242 - loss 2.85941045 - time (sec): 2.62 - samples/sec: 6562.65 - lr: 0.000035 - momentum: 0.000000
2023-10-18 16:36:26,542 epoch 1 - iter 192/242 - loss 2.60859374 - time (sec): 2.98 - samples/sec: 6557.08 - lr: 0.000039 - momentum: 0.000000
2023-10-18 16:36:26,921 epoch 1 - iter 216/242 - loss 2.38057612 - time (sec): 3.36 - samples/sec: 6626.07 - lr: 0.000044 - momentum: 0.000000
2023-10-18 16:36:27,289 epoch 1 - iter 240/242 - loss 2.22565024 - time (sec): 3.73 - samples/sec: 6578.79 - lr: 0.000049 - momentum: 0.000000
2023-10-18 16:36:27,318 ----------------------------------------------------------------------------------------------------
2023-10-18 16:36:27,318 EPOCH 1 done: loss 2.2152 - lr: 0.000049
2023-10-18 16:36:27,805 DEV : loss 0.6376588940620422 - f1-score (micro avg) 0.0
2023-10-18 16:36:27,810 ----------------------------------------------------------------------------------------------------
2023-10-18 16:36:28,173 epoch 2 - iter 24/242 - loss 0.71028364 - time (sec): 0.36 - samples/sec: 6819.79 - lr: 0.000049 - momentum: 0.000000
2023-10-18 16:36:28,539 epoch 2 - iter 48/242 - loss 0.71336343 - time (sec): 0.73 - samples/sec: 7166.82 - lr: 0.000049 - momentum: 0.000000
2023-10-18 16:36:28,907 epoch 2 - iter 72/242 - loss 0.67520329 - time (sec): 1.10 - samples/sec: 7266.90 - lr: 0.000048 - momentum: 0.000000
2023-10-18 16:36:29,273 epoch 2 - iter 96/242 - loss 0.66293218 - time (sec): 1.46 - samples/sec: 7276.61 - lr: 0.000048 - momentum: 0.000000
2023-10-18 16:36:29,622 epoch 2 - iter 120/242 - loss 0.65478994 - time (sec): 1.81 - samples/sec: 6927.43 - lr: 0.000047 - momentum: 0.000000
2023-10-18 16:36:29,980 epoch 2 - iter 144/242 - loss 0.64752733 - time (sec): 2.17 - samples/sec: 6783.23 - lr: 0.000047 - momentum: 0.000000
2023-10-18 16:36:30,346 epoch 2 - iter 168/242 - loss 0.64034132 - time (sec): 2.54 - samples/sec: 6729.42 - lr: 0.000046 - momentum: 0.000000
2023-10-18 16:36:30,713 epoch 2 - iter 192/242 - loss 0.63776300 - time (sec): 2.90 - samples/sec: 6775.14 - lr: 0.000046 - momentum: 0.000000
2023-10-18 16:36:31,074 epoch 2 - iter 216/242 - loss 0.62689871 - time (sec): 3.26 - samples/sec: 6784.92 - lr: 0.000045 - momentum: 0.000000
2023-10-18 16:36:31,458 epoch 2 - iter 240/242 - loss 0.60546341 - time (sec): 3.65 - samples/sec: 6743.08 - lr: 0.000045 - momentum: 0.000000
2023-10-18 16:36:31,484 ----------------------------------------------------------------------------------------------------
2023-10-18 16:36:31,484 EPOCH 2 done: loss 0.6054 - lr: 0.000045
2023-10-18 16:36:31,911 DEV : loss 0.44410139322280884 - f1-score (micro avg) 0.1726
2023-10-18 16:36:31,915 saving best model
2023-10-18 16:36:31,941 ----------------------------------------------------------------------------------------------------
2023-10-18 16:36:32,302 epoch 3 - iter 24/242 - loss 0.47995476 - time (sec): 0.36 - samples/sec: 6864.00 - lr: 0.000044 - momentum: 0.000000
2023-10-18 16:36:32,645 epoch 3 - iter 48/242 - loss 0.51537977 - time (sec): 0.70 - samples/sec: 6600.20 - lr: 0.000043 - momentum: 0.000000
2023-10-18 16:36:33,009 epoch 3 - iter 72/242 - loss 0.52591455 - time (sec): 1.07 - samples/sec: 6524.08 - lr: 0.000043 - momentum: 0.000000
2023-10-18 16:36:33,351 epoch 3 - iter 96/242 - loss 0.50834626 - time (sec): 1.41 - samples/sec: 6739.80 - lr: 0.000042 - momentum: 0.000000
2023-10-18 16:36:33,688 epoch 3 - iter 120/242 - loss 0.49923645 - time (sec): 1.75 - samples/sec: 6818.81 - lr: 0.000042 - momentum: 0.000000
2023-10-18 16:36:34,040 epoch 3 - iter 144/242 - loss 0.47877640 - time (sec): 2.10 - samples/sec: 6982.35 - lr: 0.000041 - momentum: 0.000000
2023-10-18 16:36:34,400 epoch 3 - iter 168/242 - loss 0.47198182 - time (sec): 2.46 - samples/sec: 6952.98 - lr: 0.000041 - momentum: 0.000000
2023-10-18 16:36:34,757 epoch 3 - iter 192/242 - loss 0.46892658 - time (sec): 2.82 - samples/sec: 6916.36 - lr: 0.000040 - momentum: 0.000000
2023-10-18 16:36:35,142 epoch 3 - iter 216/242 - loss 0.45945229 - time (sec): 3.20 - samples/sec: 6897.19 - lr: 0.000040 - momentum: 0.000000
2023-10-18 16:36:35,519 epoch 3 - iter 240/242 - loss 0.45603902 - time (sec): 3.58 - samples/sec: 6859.49 - lr: 0.000039 - momentum: 0.000000
2023-10-18 16:36:35,546 ----------------------------------------------------------------------------------------------------
2023-10-18 16:36:35,547 EPOCH 3 done: loss 0.4571 - lr: 0.000039
2023-10-18 16:36:35,977 DEV : loss 0.3233063220977783 - f1-score (micro avg) 0.4841
2023-10-18 16:36:35,981 saving best model
2023-10-18 16:36:36,017 ----------------------------------------------------------------------------------------------------
2023-10-18 16:36:36,391 epoch 4 - iter 24/242 - loss 0.45516957 - time (sec): 0.37 - samples/sec: 7271.44 - lr: 0.000038 - momentum: 0.000000
2023-10-18 16:36:36,760 epoch 4 - iter 48/242 - loss 0.40167627 - time (sec): 0.74 - samples/sec: 6707.20 - lr: 0.000038 - momentum: 0.000000
2023-10-18 16:36:37,131 epoch 4 - iter 72/242 - loss 0.40130941 - time (sec): 1.11 - samples/sec: 6552.25 - lr: 0.000037 - momentum: 0.000000
2023-10-18 16:36:37,514 epoch 4 - iter 96/242 - loss 0.39325755 - time (sec): 1.50 - samples/sec: 6506.63 - lr: 0.000037 - momentum: 0.000000
2023-10-18 16:36:37,903 epoch 4 - iter 120/242 - loss 0.38875783 - time (sec): 1.88 - samples/sec: 6623.69 - lr: 0.000036 - momentum: 0.000000
2023-10-18 16:36:38,271 epoch 4 - iter 144/242 - loss 0.38181180 - time (sec): 2.25 - samples/sec: 6590.83 - lr: 0.000036 - momentum: 0.000000
2023-10-18 16:36:38,647 epoch 4 - iter 168/242 - loss 0.37363013 - time (sec): 2.63 - samples/sec: 6529.96 - lr: 0.000035 - momentum: 0.000000
2023-10-18 16:36:39,013 epoch 4 - iter 192/242 - loss 0.36751058 - time (sec): 3.00 - samples/sec: 6612.55 - lr: 0.000035 - momentum: 0.000000
2023-10-18 16:36:39,388 epoch 4 - iter 216/242 - loss 0.37614560 - time (sec): 3.37 - samples/sec: 6580.07 - lr: 0.000034 - momentum: 0.000000
2023-10-18 16:36:39,751 epoch 4 - iter 240/242 - loss 0.37512641 - time (sec): 3.73 - samples/sec: 6591.02 - lr: 0.000033 - momentum: 0.000000
2023-10-18 16:36:39,779 ----------------------------------------------------------------------------------------------------
2023-10-18 16:36:39,779 EPOCH 4 done: loss 0.3752 - lr: 0.000033
2023-10-18 16:36:40,207 DEV : loss 0.2869265079498291 - f1-score (micro avg) 0.4813
2023-10-18 16:36:40,211 ----------------------------------------------------------------------------------------------------
2023-10-18 16:36:40,588 epoch 5 - iter 24/242 - loss 0.34496146 - time (sec): 0.38 - samples/sec: 6220.26 - lr: 0.000033 - momentum: 0.000000
2023-10-18 16:36:40,973 epoch 5 - iter 48/242 - loss 0.34981845 - time (sec): 0.76 - samples/sec: 6280.07 - lr: 0.000032 - momentum: 0.000000
2023-10-18 16:36:41,332 epoch 5 - iter 72/242 - loss 0.34066324 - time (sec): 1.12 - samples/sec: 6387.39 - lr: 0.000032 - momentum: 0.000000
2023-10-18 16:36:41,722 epoch 5 - iter 96/242 - loss 0.34406469 - time (sec): 1.51 - samples/sec: 6597.13 - lr: 0.000031 - momentum: 0.000000
2023-10-18 16:36:42,097 epoch 5 - iter 120/242 - loss 0.34773483 - time (sec): 1.89 - samples/sec: 6664.73 - lr: 0.000031 - momentum: 0.000000
2023-10-18 16:36:42,484 epoch 5 - iter 144/242 - loss 0.34325992 - time (sec): 2.27 - samples/sec: 6613.90 - lr: 0.000030 - momentum: 0.000000
2023-10-18 16:36:42,869 epoch 5 - iter 168/242 - loss 0.34035695 - time (sec): 2.66 - samples/sec: 6510.16 - lr: 0.000030 - momentum: 0.000000
2023-10-18 16:36:43,256 epoch 5 - iter 192/242 - loss 0.33886687 - time (sec): 3.04 - samples/sec: 6527.83 - lr: 0.000029 - momentum: 0.000000
2023-10-18 16:36:43,626 epoch 5 - iter 216/242 - loss 0.34406810 - time (sec): 3.41 - samples/sec: 6519.10 - lr: 0.000028 - momentum: 0.000000
2023-10-18 16:36:43,992 epoch 5 - iter 240/242 - loss 0.33921636 - time (sec): 3.78 - samples/sec: 6507.41 - lr: 0.000028 - momentum: 0.000000
2023-10-18 16:36:44,020 ----------------------------------------------------------------------------------------------------
2023-10-18 16:36:44,020 EPOCH 5 done: loss 0.3406 - lr: 0.000028
2023-10-18 16:36:44,444 DEV : loss 0.2742924690246582 - f1-score (micro avg) 0.4738
2023-10-18 16:36:44,449 ----------------------------------------------------------------------------------------------------
2023-10-18 16:36:44,817 epoch 6 - iter 24/242 - loss 0.31470992 - time (sec): 0.37 - samples/sec: 6960.30 - lr: 0.000027 - momentum: 0.000000
2023-10-18 16:36:45,180 epoch 6 - iter 48/242 - loss 0.33456371 - time (sec): 0.73 - samples/sec: 7029.92 - lr: 0.000027 - momentum: 0.000000
2023-10-18 16:36:45,545 epoch 6 - iter 72/242 - loss 0.30996261 - time (sec): 1.10 - samples/sec: 6973.68 - lr: 0.000026 - momentum: 0.000000
2023-10-18 16:36:45,924 epoch 6 - iter 96/242 - loss 0.31807708 - time (sec): 1.48 - samples/sec: 6892.55 - lr: 0.000026 - momentum: 0.000000
2023-10-18 16:36:46,313 epoch 6 - iter 120/242 - loss 0.31619253 - time (sec): 1.86 - samples/sec: 6800.34 - lr: 0.000025 - momentum: 0.000000
2023-10-18 16:36:46,706 epoch 6 - iter 144/242 - loss 0.30833559 - time (sec): 2.26 - samples/sec: 6714.31 - lr: 0.000025 - momentum: 0.000000
2023-10-18 16:36:47,071 epoch 6 - iter 168/242 - loss 0.30530212 - time (sec): 2.62 - samples/sec: 6621.78 - lr: 0.000024 - momentum: 0.000000
2023-10-18 16:36:47,459 epoch 6 - iter 192/242 - loss 0.31101338 - time (sec): 3.01 - samples/sec: 6580.76 - lr: 0.000023 - momentum: 0.000000
2023-10-18 16:36:47,826 epoch 6 - iter 216/242 - loss 0.31555222 - time (sec): 3.38 - samples/sec: 6603.40 - lr: 0.000023 - momentum: 0.000000
2023-10-18 16:36:48,218 epoch 6 - iter 240/242 - loss 0.31269087 - time (sec): 3.77 - samples/sec: 6528.91 - lr: 0.000022 - momentum: 0.000000
2023-10-18 16:36:48,243 ----------------------------------------------------------------------------------------------------
2023-10-18 16:36:48,243 EPOCH 6 done: loss 0.3112 - lr: 0.000022
2023-10-18 16:36:48,681 DEV : loss 0.25722795724868774 - f1-score (micro avg) 0.4952
2023-10-18 16:36:48,685 saving best model
2023-10-18 16:36:48,721 ----------------------------------------------------------------------------------------------------
2023-10-18 16:36:49,070 epoch 7 - iter 24/242 - loss 0.28473485 - time (sec): 0.35 - samples/sec: 8026.85 - lr: 0.000022 - momentum: 0.000000
2023-10-18 16:36:49,410 epoch 7 - iter 48/242 - loss 0.28032042 - time (sec): 0.69 - samples/sec: 7682.13 - lr: 0.000021 - momentum: 0.000000
2023-10-18 16:36:49,749 epoch 7 - iter 72/242 - loss 0.28315683 - time (sec): 1.03 - samples/sec: 7471.26 - lr: 0.000021 - momentum: 0.000000
2023-10-18 16:36:50,084 epoch 7 - iter 96/242 - loss 0.29487916 - time (sec): 1.36 - samples/sec: 7312.20 - lr: 0.000020 - momentum: 0.000000
2023-10-18 16:36:50,414 epoch 7 - iter 120/242 - loss 0.29815540 - time (sec): 1.69 - samples/sec: 7239.42 - lr: 0.000020 - momentum: 0.000000
2023-10-18 16:36:50,787 epoch 7 - iter 144/242 - loss 0.29268716 - time (sec): 2.06 - samples/sec: 7062.20 - lr: 0.000019 - momentum: 0.000000
2023-10-18 16:36:51,176 epoch 7 - iter 168/242 - loss 0.29539971 - time (sec): 2.45 - samples/sec: 6963.28 - lr: 0.000018 - momentum: 0.000000
2023-10-18 16:36:51,556 epoch 7 - iter 192/242 - loss 0.29791900 - time (sec): 2.83 - samples/sec: 6917.39 - lr: 0.000018 - momentum: 0.000000
2023-10-18 16:36:51,946 epoch 7 - iter 216/242 - loss 0.29986562 - time (sec): 3.22 - samples/sec: 6849.24 - lr: 0.000017 - momentum: 0.000000
2023-10-18 16:36:52,324 epoch 7 - iter 240/242 - loss 0.29782140 - time (sec): 3.60 - samples/sec: 6830.49 - lr: 0.000017 - momentum: 0.000000
2023-10-18 16:36:52,351 ----------------------------------------------------------------------------------------------------
2023-10-18 16:36:52,352 EPOCH 7 done: loss 0.2979 - lr: 0.000017
2023-10-18 16:36:52,783 DEV : loss 0.2522064745426178 - f1-score (micro avg) 0.5171
2023-10-18 16:36:52,787 saving best model
2023-10-18 16:36:52,823 ----------------------------------------------------------------------------------------------------
2023-10-18 16:36:53,199 epoch 8 - iter 24/242 - loss 0.41692853 - time (sec): 0.37 - samples/sec: 7290.35 - lr: 0.000016 - momentum: 0.000000
2023-10-18 16:36:53,552 epoch 8 - iter 48/242 - loss 0.35604078 - time (sec): 0.73 - samples/sec: 6821.61 - lr: 0.000016 - momentum: 0.000000
2023-10-18 16:36:53,915 epoch 8 - iter 72/242 - loss 0.32862463 - time (sec): 1.09 - samples/sec: 6801.12 - lr: 0.000015 - momentum: 0.000000
2023-10-18 16:36:54,296 epoch 8 - iter 96/242 - loss 0.30818150 - time (sec): 1.47 - samples/sec: 6767.13 - lr: 0.000015 - momentum: 0.000000
2023-10-18 16:36:54,674 epoch 8 - iter 120/242 - loss 0.29987867 - time (sec): 1.85 - samples/sec: 6785.04 - lr: 0.000014 - momentum: 0.000000
2023-10-18 16:36:55,043 epoch 8 - iter 144/242 - loss 0.29155435 - time (sec): 2.22 - samples/sec: 6669.29 - lr: 0.000013 - momentum: 0.000000
2023-10-18 16:36:55,419 epoch 8 - iter 168/242 - loss 0.28828688 - time (sec): 2.60 - samples/sec: 6654.60 - lr: 0.000013 - momentum: 0.000000
2023-10-18 16:36:55,807 epoch 8 - iter 192/242 - loss 0.28787647 - time (sec): 2.98 - samples/sec: 6646.64 - lr: 0.000012 - momentum: 0.000000
2023-10-18 16:36:56,187 epoch 8 - iter 216/242 - loss 0.29399445 - time (sec): 3.36 - samples/sec: 6649.16 - lr: 0.000012 - momentum: 0.000000
2023-10-18 16:36:56,576 epoch 8 - iter 240/242 - loss 0.28922123 - time (sec): 3.75 - samples/sec: 6566.51 - lr: 0.000011 - momentum: 0.000000
2023-10-18 16:36:56,606 ----------------------------------------------------------------------------------------------------
2023-10-18 16:36:56,607 EPOCH 8 done: loss 0.2887 - lr: 0.000011
2023-10-18 16:36:57,041 DEV : loss 0.2508045434951782 - f1-score (micro avg) 0.5126
2023-10-18 16:36:57,045 ----------------------------------------------------------------------------------------------------
2023-10-18 16:36:57,407 epoch 9 - iter 24/242 - loss 0.31471641 - time (sec): 0.36 - samples/sec: 6546.29 - lr: 0.000011 - momentum: 0.000000
2023-10-18 16:36:57,779 epoch 9 - iter 48/242 - loss 0.30325743 - time (sec): 0.73 - samples/sec: 6389.28 - lr: 0.000010 - momentum: 0.000000
2023-10-18 16:36:58,145 epoch 9 - iter 72/242 - loss 0.28445914 - time (sec): 1.10 - samples/sec: 6611.58 - lr: 0.000010 - momentum: 0.000000
2023-10-18 16:36:58,499 epoch 9 - iter 96/242 - loss 0.27733067 - time (sec): 1.45 - samples/sec: 6587.85 - lr: 0.000009 - momentum: 0.000000
2023-10-18 16:36:58,857 epoch 9 - iter 120/242 - loss 0.28394335 - time (sec): 1.81 - samples/sec: 6657.66 - lr: 0.000008 - momentum: 0.000000
2023-10-18 16:36:59,226 epoch 9 - iter 144/242 - loss 0.28730005 - time (sec): 2.18 - samples/sec: 6708.11 - lr: 0.000008 - momentum: 0.000000
2023-10-18 16:36:59,592 epoch 9 - iter 168/242 - loss 0.29711454 - time (sec): 2.55 - samples/sec: 6718.46 - lr: 0.000007 - momentum: 0.000000
2023-10-18 16:36:59,962 epoch 9 - iter 192/242 - loss 0.29329494 - time (sec): 2.92 - samples/sec: 6707.05 - lr: 0.000007 - momentum: 0.000000
2023-10-18 16:37:00,335 epoch 9 - iter 216/242 - loss 0.28875404 - time (sec): 3.29 - samples/sec: 6731.96 - lr: 0.000006 - momentum: 0.000000
2023-10-18 16:37:00,702 epoch 9 - iter 240/242 - loss 0.28175821 - time (sec): 3.66 - samples/sec: 6731.70 - lr: 0.000006 - momentum: 0.000000
2023-10-18 16:37:00,729 ----------------------------------------------------------------------------------------------------
2023-10-18 16:37:00,729 EPOCH 9 done: loss 0.2822 - lr: 0.000006
2023-10-18 16:37:01,161 DEV : loss 0.24359287321567535 - f1-score (micro avg) 0.5346
2023-10-18 16:37:01,166 saving best model
2023-10-18 16:37:01,200 ----------------------------------------------------------------------------------------------------
2023-10-18 16:37:01,568 epoch 10 - iter 24/242 - loss 0.26971381 - time (sec): 0.37 - samples/sec: 5710.03 - lr: 0.000005 - momentum: 0.000000
2023-10-18 16:37:01,933 epoch 10 - iter 48/242 - loss 0.27723315 - time (sec): 0.73 - samples/sec: 6180.39 - lr: 0.000005 - momentum: 0.000000
2023-10-18 16:37:02,292 epoch 10 - iter 72/242 - loss 0.27734124 - time (sec): 1.09 - samples/sec: 6415.68 - lr: 0.000004 - momentum: 0.000000
2023-10-18 16:37:02,665 epoch 10 - iter 96/242 - loss 0.28231197 - time (sec): 1.46 - samples/sec: 6552.71 - lr: 0.000003 - momentum: 0.000000
2023-10-18 16:37:03,035 epoch 10 - iter 120/242 - loss 0.27821230 - time (sec): 1.83 - samples/sec: 6595.37 - lr: 0.000003 - momentum: 0.000000
2023-10-18 16:37:03,427 epoch 10 - iter 144/242 - loss 0.27186754 - time (sec): 2.23 - samples/sec: 6586.10 - lr: 0.000002 - momentum: 0.000000
2023-10-18 16:37:03,804 epoch 10 - iter 168/242 - loss 0.27995109 - time (sec): 2.60 - samples/sec: 6562.68 - lr: 0.000002 - momentum: 0.000000
2023-10-18 16:37:04,180 epoch 10 - iter 192/242 - loss 0.28424878 - time (sec): 2.98 - samples/sec: 6557.62 - lr: 0.000001 - momentum: 0.000000
2023-10-18 16:37:04,562 epoch 10 - iter 216/242 - loss 0.27775247 - time (sec): 3.36 - samples/sec: 6536.76 - lr: 0.000001 - momentum: 0.000000
2023-10-18 16:37:04,904 epoch 10 - iter 240/242 - loss 0.27507515 - time (sec): 3.70 - samples/sec: 6662.50 - lr: 0.000000 - momentum: 0.000000
2023-10-18 16:37:04,927 ----------------------------------------------------------------------------------------------------
2023-10-18 16:37:04,927 EPOCH 10 done: loss 0.2750 - lr: 0.000000
2023-10-18 16:37:05,368 DEV : loss 0.2437872439622879 - f1-score (micro avg) 0.5329
2023-10-18 16:37:05,400 ----------------------------------------------------------------------------------------------------
2023-10-18 16:37:05,401 Loading model from best epoch ...
2023-10-18 16:37:05,478 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-object, B-object, E-object, I-object, S-date, B-date, E-date, I-date
2023-10-18 16:37:05,894
Results:
- F-score (micro) 0.4903
- F-score (macro) 0.2649
- Accuracy 0.3486
By class:
precision recall f1-score support
scope 0.3547 0.5581 0.4337 129
pers 0.5707 0.7554 0.6502 139
work 0.4643 0.1625 0.2407 80
loc 0.0000 0.0000 0.0000 9
date 0.0000 0.0000 0.0000 3
micro avg 0.4578 0.5278 0.4903 360
macro avg 0.2779 0.2952 0.2649 360
weighted avg 0.4506 0.5278 0.4600 360
2023-10-18 16:37:05,895 ----------------------------------------------------------------------------------------------------
|