File size: 24,205 Bytes
74c6741 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 |
2023-10-18 18:52:52,741 ----------------------------------------------------------------------------------------------------
2023-10-18 18:52:52,741 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=21, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-18 18:52:52,741 ----------------------------------------------------------------------------------------------------
2023-10-18 18:52:52,742 MultiCorpus: 5901 train + 1287 dev + 1505 test sentences
- NER_HIPE_2022 Corpus: 5901 train + 1287 dev + 1505 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/hipe2020/fr/with_doc_seperator
2023-10-18 18:52:52,742 ----------------------------------------------------------------------------------------------------
2023-10-18 18:52:52,742 Train: 5901 sentences
2023-10-18 18:52:52,742 (train_with_dev=False, train_with_test=False)
2023-10-18 18:52:52,742 ----------------------------------------------------------------------------------------------------
2023-10-18 18:52:52,742 Training Params:
2023-10-18 18:52:52,742 - learning_rate: "3e-05"
2023-10-18 18:52:52,742 - mini_batch_size: "8"
2023-10-18 18:52:52,742 - max_epochs: "10"
2023-10-18 18:52:52,742 - shuffle: "True"
2023-10-18 18:52:52,742 ----------------------------------------------------------------------------------------------------
2023-10-18 18:52:52,742 Plugins:
2023-10-18 18:52:52,742 - TensorboardLogger
2023-10-18 18:52:52,742 - LinearScheduler | warmup_fraction: '0.1'
2023-10-18 18:52:52,742 ----------------------------------------------------------------------------------------------------
2023-10-18 18:52:52,742 Final evaluation on model from best epoch (best-model.pt)
2023-10-18 18:52:52,742 - metric: "('micro avg', 'f1-score')"
2023-10-18 18:52:52,742 ----------------------------------------------------------------------------------------------------
2023-10-18 18:52:52,742 Computation:
2023-10-18 18:52:52,742 - compute on device: cuda:0
2023-10-18 18:52:52,742 - embedding storage: none
2023-10-18 18:52:52,742 ----------------------------------------------------------------------------------------------------
2023-10-18 18:52:52,742 Model training base path: "hmbench-hipe2020/fr-dbmdz/bert-tiny-historic-multilingual-cased-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-1"
2023-10-18 18:52:52,742 ----------------------------------------------------------------------------------------------------
2023-10-18 18:52:52,742 ----------------------------------------------------------------------------------------------------
2023-10-18 18:52:52,742 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-18 18:52:54,502 epoch 1 - iter 73/738 - loss 3.81874353 - time (sec): 1.76 - samples/sec: 9383.08 - lr: 0.000003 - momentum: 0.000000
2023-10-18 18:52:56,284 epoch 1 - iter 146/738 - loss 3.68234957 - time (sec): 3.54 - samples/sec: 8933.75 - lr: 0.000006 - momentum: 0.000000
2023-10-18 18:52:58,194 epoch 1 - iter 219/738 - loss 3.37630765 - time (sec): 5.45 - samples/sec: 9239.17 - lr: 0.000009 - momentum: 0.000000
2023-10-18 18:52:59,863 epoch 1 - iter 292/738 - loss 3.02254508 - time (sec): 7.12 - samples/sec: 9456.64 - lr: 0.000012 - momentum: 0.000000
2023-10-18 18:53:01,582 epoch 1 - iter 365/738 - loss 2.66170049 - time (sec): 8.84 - samples/sec: 9361.30 - lr: 0.000015 - momentum: 0.000000
2023-10-18 18:53:03,272 epoch 1 - iter 438/738 - loss 2.36532082 - time (sec): 10.53 - samples/sec: 9328.38 - lr: 0.000018 - momentum: 0.000000
2023-10-18 18:53:05,062 epoch 1 - iter 511/738 - loss 2.11033803 - time (sec): 12.32 - samples/sec: 9374.21 - lr: 0.000021 - momentum: 0.000000
2023-10-18 18:53:06,754 epoch 1 - iter 584/738 - loss 1.93341360 - time (sec): 14.01 - samples/sec: 9374.65 - lr: 0.000024 - momentum: 0.000000
2023-10-18 18:53:08,399 epoch 1 - iter 657/738 - loss 1.79640058 - time (sec): 15.66 - samples/sec: 9356.30 - lr: 0.000027 - momentum: 0.000000
2023-10-18 18:53:10,201 epoch 1 - iter 730/738 - loss 1.66035043 - time (sec): 17.46 - samples/sec: 9446.21 - lr: 0.000030 - momentum: 0.000000
2023-10-18 18:53:10,377 ----------------------------------------------------------------------------------------------------
2023-10-18 18:53:10,377 EPOCH 1 done: loss 1.6503 - lr: 0.000030
2023-10-18 18:53:13,137 DEV : loss 0.48679542541503906 - f1-score (micro avg) 0.0
2023-10-18 18:53:13,162 ----------------------------------------------------------------------------------------------------
2023-10-18 18:53:14,888 epoch 2 - iter 73/738 - loss 0.53428200 - time (sec): 1.73 - samples/sec: 9434.51 - lr: 0.000030 - momentum: 0.000000
2023-10-18 18:53:16,772 epoch 2 - iter 146/738 - loss 0.53407256 - time (sec): 3.61 - samples/sec: 9471.60 - lr: 0.000029 - momentum: 0.000000
2023-10-18 18:53:18,686 epoch 2 - iter 219/738 - loss 0.52996403 - time (sec): 5.52 - samples/sec: 9495.57 - lr: 0.000029 - momentum: 0.000000
2023-10-18 18:53:20,509 epoch 2 - iter 292/738 - loss 0.53131590 - time (sec): 7.35 - samples/sec: 9415.99 - lr: 0.000029 - momentum: 0.000000
2023-10-18 18:53:22,170 epoch 2 - iter 365/738 - loss 0.52158207 - time (sec): 9.01 - samples/sec: 9329.11 - lr: 0.000028 - momentum: 0.000000
2023-10-18 18:53:23,928 epoch 2 - iter 438/738 - loss 0.51558366 - time (sec): 10.77 - samples/sec: 9362.28 - lr: 0.000028 - momentum: 0.000000
2023-10-18 18:53:25,719 epoch 2 - iter 511/738 - loss 0.49662623 - time (sec): 12.56 - samples/sec: 9450.83 - lr: 0.000028 - momentum: 0.000000
2023-10-18 18:53:27,382 epoch 2 - iter 584/738 - loss 0.49077712 - time (sec): 14.22 - samples/sec: 9407.95 - lr: 0.000027 - momentum: 0.000000
2023-10-18 18:53:29,071 epoch 2 - iter 657/738 - loss 0.48484877 - time (sec): 15.91 - samples/sec: 9359.70 - lr: 0.000027 - momentum: 0.000000
2023-10-18 18:53:30,846 epoch 2 - iter 730/738 - loss 0.47901795 - time (sec): 17.68 - samples/sec: 9310.89 - lr: 0.000027 - momentum: 0.000000
2023-10-18 18:53:31,028 ----------------------------------------------------------------------------------------------------
2023-10-18 18:53:31,029 EPOCH 2 done: loss 0.4785 - lr: 0.000027
2023-10-18 18:53:38,087 DEV : loss 0.33998891711235046 - f1-score (micro avg) 0.2712
2023-10-18 18:53:38,113 saving best model
2023-10-18 18:53:38,141 ----------------------------------------------------------------------------------------------------
2023-10-18 18:53:39,865 epoch 3 - iter 73/738 - loss 0.40935465 - time (sec): 1.72 - samples/sec: 9824.36 - lr: 0.000026 - momentum: 0.000000
2023-10-18 18:53:41,684 epoch 3 - iter 146/738 - loss 0.41668843 - time (sec): 3.54 - samples/sec: 9634.44 - lr: 0.000026 - momentum: 0.000000
2023-10-18 18:53:43,431 epoch 3 - iter 219/738 - loss 0.41715597 - time (sec): 5.29 - samples/sec: 9480.32 - lr: 0.000026 - momentum: 0.000000
2023-10-18 18:53:45,146 epoch 3 - iter 292/738 - loss 0.41227420 - time (sec): 7.00 - samples/sec: 9447.71 - lr: 0.000025 - momentum: 0.000000
2023-10-18 18:53:46,726 epoch 3 - iter 365/738 - loss 0.41105667 - time (sec): 8.58 - samples/sec: 9313.13 - lr: 0.000025 - momentum: 0.000000
2023-10-18 18:53:48,446 epoch 3 - iter 438/738 - loss 0.40654280 - time (sec): 10.30 - samples/sec: 9335.83 - lr: 0.000025 - momentum: 0.000000
2023-10-18 18:53:50,257 epoch 3 - iter 511/738 - loss 0.39785791 - time (sec): 12.11 - samples/sec: 9328.86 - lr: 0.000024 - momentum: 0.000000
2023-10-18 18:53:52,055 epoch 3 - iter 584/738 - loss 0.39819890 - time (sec): 13.91 - samples/sec: 9430.47 - lr: 0.000024 - momentum: 0.000000
2023-10-18 18:53:53,814 epoch 3 - iter 657/738 - loss 0.39752472 - time (sec): 15.67 - samples/sec: 9377.86 - lr: 0.000024 - momentum: 0.000000
2023-10-18 18:53:55,587 epoch 3 - iter 730/738 - loss 0.39992408 - time (sec): 17.45 - samples/sec: 9426.87 - lr: 0.000023 - momentum: 0.000000
2023-10-18 18:53:55,783 ----------------------------------------------------------------------------------------------------
2023-10-18 18:53:55,784 EPOCH 3 done: loss 0.3998 - lr: 0.000023
2023-10-18 18:54:02,909 DEV : loss 0.30542612075805664 - f1-score (micro avg) 0.32
2023-10-18 18:54:02,934 saving best model
2023-10-18 18:54:02,977 ----------------------------------------------------------------------------------------------------
2023-10-18 18:54:04,826 epoch 4 - iter 73/738 - loss 0.36944310 - time (sec): 1.85 - samples/sec: 9510.06 - lr: 0.000023 - momentum: 0.000000
2023-10-18 18:54:06,476 epoch 4 - iter 146/738 - loss 0.35416470 - time (sec): 3.50 - samples/sec: 9315.70 - lr: 0.000023 - momentum: 0.000000
2023-10-18 18:54:08,180 epoch 4 - iter 219/738 - loss 0.36472416 - time (sec): 5.20 - samples/sec: 9343.76 - lr: 0.000022 - momentum: 0.000000
2023-10-18 18:54:09,909 epoch 4 - iter 292/738 - loss 0.36539814 - time (sec): 6.93 - samples/sec: 9519.65 - lr: 0.000022 - momentum: 0.000000
2023-10-18 18:54:11,751 epoch 4 - iter 365/738 - loss 0.36058865 - time (sec): 8.77 - samples/sec: 9569.18 - lr: 0.000022 - momentum: 0.000000
2023-10-18 18:54:13,564 epoch 4 - iter 438/738 - loss 0.36988378 - time (sec): 10.59 - samples/sec: 9494.32 - lr: 0.000021 - momentum: 0.000000
2023-10-18 18:54:15,301 epoch 4 - iter 511/738 - loss 0.36884095 - time (sec): 12.32 - samples/sec: 9483.53 - lr: 0.000021 - momentum: 0.000000
2023-10-18 18:54:16,957 epoch 4 - iter 584/738 - loss 0.36969941 - time (sec): 13.98 - samples/sec: 9494.13 - lr: 0.000021 - momentum: 0.000000
2023-10-18 18:54:18,684 epoch 4 - iter 657/738 - loss 0.36761614 - time (sec): 15.71 - samples/sec: 9486.58 - lr: 0.000020 - momentum: 0.000000
2023-10-18 18:54:20,383 epoch 4 - iter 730/738 - loss 0.36714089 - time (sec): 17.40 - samples/sec: 9473.64 - lr: 0.000020 - momentum: 0.000000
2023-10-18 18:54:20,563 ----------------------------------------------------------------------------------------------------
2023-10-18 18:54:20,563 EPOCH 4 done: loss 0.3663 - lr: 0.000020
2023-10-18 18:54:27,702 DEV : loss 0.2950303554534912 - f1-score (micro avg) 0.3587
2023-10-18 18:54:27,728 saving best model
2023-10-18 18:54:27,766 ----------------------------------------------------------------------------------------------------
2023-10-18 18:54:29,457 epoch 5 - iter 73/738 - loss 0.34146707 - time (sec): 1.69 - samples/sec: 9529.99 - lr: 0.000020 - momentum: 0.000000
2023-10-18 18:54:31,179 epoch 5 - iter 146/738 - loss 0.34642612 - time (sec): 3.41 - samples/sec: 9364.92 - lr: 0.000019 - momentum: 0.000000
2023-10-18 18:54:32,907 epoch 5 - iter 219/738 - loss 0.34668670 - time (sec): 5.14 - samples/sec: 9395.92 - lr: 0.000019 - momentum: 0.000000
2023-10-18 18:54:35,219 epoch 5 - iter 292/738 - loss 0.34194245 - time (sec): 7.45 - samples/sec: 9100.75 - lr: 0.000019 - momentum: 0.000000
2023-10-18 18:54:37,017 epoch 5 - iter 365/738 - loss 0.34187389 - time (sec): 9.25 - samples/sec: 9123.73 - lr: 0.000018 - momentum: 0.000000
2023-10-18 18:54:38,755 epoch 5 - iter 438/738 - loss 0.34124463 - time (sec): 10.99 - samples/sec: 9147.79 - lr: 0.000018 - momentum: 0.000000
2023-10-18 18:54:40,488 epoch 5 - iter 511/738 - loss 0.34019308 - time (sec): 12.72 - samples/sec: 8986.48 - lr: 0.000018 - momentum: 0.000000
2023-10-18 18:54:42,281 epoch 5 - iter 584/738 - loss 0.34038501 - time (sec): 14.51 - samples/sec: 8887.32 - lr: 0.000017 - momentum: 0.000000
2023-10-18 18:54:44,097 epoch 5 - iter 657/738 - loss 0.33829006 - time (sec): 16.33 - samples/sec: 8947.38 - lr: 0.000017 - momentum: 0.000000
2023-10-18 18:54:45,929 epoch 5 - iter 730/738 - loss 0.33961410 - time (sec): 18.16 - samples/sec: 9066.69 - lr: 0.000017 - momentum: 0.000000
2023-10-18 18:54:46,122 ----------------------------------------------------------------------------------------------------
2023-10-18 18:54:46,122 EPOCH 5 done: loss 0.3400 - lr: 0.000017
2023-10-18 18:54:53,313 DEV : loss 0.2756083011627197 - f1-score (micro avg) 0.3862
2023-10-18 18:54:53,339 saving best model
2023-10-18 18:54:53,378 ----------------------------------------------------------------------------------------------------
2023-10-18 18:54:55,141 epoch 6 - iter 73/738 - loss 0.34120287 - time (sec): 1.76 - samples/sec: 10190.28 - lr: 0.000016 - momentum: 0.000000
2023-10-18 18:54:56,920 epoch 6 - iter 146/738 - loss 0.32162573 - time (sec): 3.54 - samples/sec: 9838.24 - lr: 0.000016 - momentum: 0.000000
2023-10-18 18:54:58,625 epoch 6 - iter 219/738 - loss 0.31874927 - time (sec): 5.25 - samples/sec: 9710.50 - lr: 0.000016 - momentum: 0.000000
2023-10-18 18:55:00,318 epoch 6 - iter 292/738 - loss 0.32065056 - time (sec): 6.94 - samples/sec: 9693.25 - lr: 0.000015 - momentum: 0.000000
2023-10-18 18:55:02,163 epoch 6 - iter 365/738 - loss 0.31480678 - time (sec): 8.78 - samples/sec: 9569.85 - lr: 0.000015 - momentum: 0.000000
2023-10-18 18:55:03,941 epoch 6 - iter 438/738 - loss 0.32409156 - time (sec): 10.56 - samples/sec: 9331.67 - lr: 0.000015 - momentum: 0.000000
2023-10-18 18:55:05,700 epoch 6 - iter 511/738 - loss 0.32737315 - time (sec): 12.32 - samples/sec: 9351.69 - lr: 0.000014 - momentum: 0.000000
2023-10-18 18:55:07,401 epoch 6 - iter 584/738 - loss 0.32330551 - time (sec): 14.02 - samples/sec: 9309.21 - lr: 0.000014 - momentum: 0.000000
2023-10-18 18:55:09,120 epoch 6 - iter 657/738 - loss 0.32258843 - time (sec): 15.74 - samples/sec: 9300.41 - lr: 0.000014 - momentum: 0.000000
2023-10-18 18:55:10,901 epoch 6 - iter 730/738 - loss 0.32208614 - time (sec): 17.52 - samples/sec: 9395.06 - lr: 0.000013 - momentum: 0.000000
2023-10-18 18:55:11,098 ----------------------------------------------------------------------------------------------------
2023-10-18 18:55:11,098 EPOCH 6 done: loss 0.3219 - lr: 0.000013
2023-10-18 18:55:18,261 DEV : loss 0.26639610528945923 - f1-score (micro avg) 0.4146
2023-10-18 18:55:18,288 saving best model
2023-10-18 18:55:18,322 ----------------------------------------------------------------------------------------------------
2023-10-18 18:55:20,066 epoch 7 - iter 73/738 - loss 0.30556597 - time (sec): 1.74 - samples/sec: 9564.01 - lr: 0.000013 - momentum: 0.000000
2023-10-18 18:55:21,775 epoch 7 - iter 146/738 - loss 0.31726978 - time (sec): 3.45 - samples/sec: 9546.50 - lr: 0.000013 - momentum: 0.000000
2023-10-18 18:55:23,473 epoch 7 - iter 219/738 - loss 0.30091048 - time (sec): 5.15 - samples/sec: 9444.26 - lr: 0.000012 - momentum: 0.000000
2023-10-18 18:55:25,177 epoch 7 - iter 292/738 - loss 0.30981691 - time (sec): 6.85 - samples/sec: 9502.79 - lr: 0.000012 - momentum: 0.000000
2023-10-18 18:55:26,970 epoch 7 - iter 365/738 - loss 0.30604870 - time (sec): 8.65 - samples/sec: 9520.74 - lr: 0.000012 - momentum: 0.000000
2023-10-18 18:55:28,704 epoch 7 - iter 438/738 - loss 0.30386617 - time (sec): 10.38 - samples/sec: 9526.86 - lr: 0.000011 - momentum: 0.000000
2023-10-18 18:55:30,392 epoch 7 - iter 511/738 - loss 0.30690882 - time (sec): 12.07 - samples/sec: 9522.71 - lr: 0.000011 - momentum: 0.000000
2023-10-18 18:55:32,166 epoch 7 - iter 584/738 - loss 0.30363679 - time (sec): 13.84 - samples/sec: 9488.93 - lr: 0.000011 - momentum: 0.000000
2023-10-18 18:55:33,875 epoch 7 - iter 657/738 - loss 0.30303827 - time (sec): 15.55 - samples/sec: 9521.30 - lr: 0.000010 - momentum: 0.000000
2023-10-18 18:55:35,616 epoch 7 - iter 730/738 - loss 0.30566278 - time (sec): 17.29 - samples/sec: 9534.15 - lr: 0.000010 - momentum: 0.000000
2023-10-18 18:55:35,806 ----------------------------------------------------------------------------------------------------
2023-10-18 18:55:35,806 EPOCH 7 done: loss 0.3074 - lr: 0.000010
2023-10-18 18:55:42,992 DEV : loss 0.2636333405971527 - f1-score (micro avg) 0.4126
2023-10-18 18:55:43,019 ----------------------------------------------------------------------------------------------------
2023-10-18 18:55:44,858 epoch 8 - iter 73/738 - loss 0.30261683 - time (sec): 1.84 - samples/sec: 11006.58 - lr: 0.000010 - momentum: 0.000000
2023-10-18 18:55:46,596 epoch 8 - iter 146/738 - loss 0.30342876 - time (sec): 3.58 - samples/sec: 10365.01 - lr: 0.000009 - momentum: 0.000000
2023-10-18 18:55:48,328 epoch 8 - iter 219/738 - loss 0.29957239 - time (sec): 5.31 - samples/sec: 9959.68 - lr: 0.000009 - momentum: 0.000000
2023-10-18 18:55:50,041 epoch 8 - iter 292/738 - loss 0.30420435 - time (sec): 7.02 - samples/sec: 9882.01 - lr: 0.000009 - momentum: 0.000000
2023-10-18 18:55:51,706 epoch 8 - iter 365/738 - loss 0.29943806 - time (sec): 8.69 - samples/sec: 9704.42 - lr: 0.000008 - momentum: 0.000000
2023-10-18 18:55:53,371 epoch 8 - iter 438/738 - loss 0.30018714 - time (sec): 10.35 - samples/sec: 9670.71 - lr: 0.000008 - momentum: 0.000000
2023-10-18 18:55:55,084 epoch 8 - iter 511/738 - loss 0.29946570 - time (sec): 12.06 - samples/sec: 9639.16 - lr: 0.000008 - momentum: 0.000000
2023-10-18 18:55:56,861 epoch 8 - iter 584/738 - loss 0.29967547 - time (sec): 13.84 - samples/sec: 9609.30 - lr: 0.000007 - momentum: 0.000000
2023-10-18 18:55:58,534 epoch 8 - iter 657/738 - loss 0.30042341 - time (sec): 15.51 - samples/sec: 9584.72 - lr: 0.000007 - momentum: 0.000000
2023-10-18 18:56:00,204 epoch 8 - iter 730/738 - loss 0.29829974 - time (sec): 17.18 - samples/sec: 9567.93 - lr: 0.000007 - momentum: 0.000000
2023-10-18 18:56:00,391 ----------------------------------------------------------------------------------------------------
2023-10-18 18:56:00,391 EPOCH 8 done: loss 0.2999 - lr: 0.000007
2023-10-18 18:56:07,599 DEV : loss 0.2585228383541107 - f1-score (micro avg) 0.4277
2023-10-18 18:56:07,625 saving best model
2023-10-18 18:56:07,658 ----------------------------------------------------------------------------------------------------
2023-10-18 18:56:09,344 epoch 9 - iter 73/738 - loss 0.30758024 - time (sec): 1.69 - samples/sec: 9099.32 - lr: 0.000006 - momentum: 0.000000
2023-10-18 18:56:11,067 epoch 9 - iter 146/738 - loss 0.28960177 - time (sec): 3.41 - samples/sec: 9174.55 - lr: 0.000006 - momentum: 0.000000
2023-10-18 18:56:12,916 epoch 9 - iter 219/738 - loss 0.30272308 - time (sec): 5.26 - samples/sec: 9519.35 - lr: 0.000006 - momentum: 0.000000
2023-10-18 18:56:15,107 epoch 9 - iter 292/738 - loss 0.30045146 - time (sec): 7.45 - samples/sec: 8945.26 - lr: 0.000005 - momentum: 0.000000
2023-10-18 18:56:16,846 epoch 9 - iter 365/738 - loss 0.30003631 - time (sec): 9.19 - samples/sec: 8958.35 - lr: 0.000005 - momentum: 0.000000
2023-10-18 18:56:18,557 epoch 9 - iter 438/738 - loss 0.30160690 - time (sec): 10.90 - samples/sec: 8970.13 - lr: 0.000005 - momentum: 0.000000
2023-10-18 18:56:20,297 epoch 9 - iter 511/738 - loss 0.29885683 - time (sec): 12.64 - samples/sec: 9090.97 - lr: 0.000004 - momentum: 0.000000
2023-10-18 18:56:22,022 epoch 9 - iter 584/738 - loss 0.29861281 - time (sec): 14.36 - samples/sec: 9134.26 - lr: 0.000004 - momentum: 0.000000
2023-10-18 18:56:23,671 epoch 9 - iter 657/738 - loss 0.29753874 - time (sec): 16.01 - samples/sec: 9118.57 - lr: 0.000004 - momentum: 0.000000
2023-10-18 18:56:25,423 epoch 9 - iter 730/738 - loss 0.29722536 - time (sec): 17.76 - samples/sec: 9159.83 - lr: 0.000003 - momentum: 0.000000
2023-10-18 18:56:25,709 ----------------------------------------------------------------------------------------------------
2023-10-18 18:56:25,709 EPOCH 9 done: loss 0.2950 - lr: 0.000003
2023-10-18 18:56:32,950 DEV : loss 0.2576240301132202 - f1-score (micro avg) 0.431
2023-10-18 18:56:32,976 saving best model
2023-10-18 18:56:33,008 ----------------------------------------------------------------------------------------------------
2023-10-18 18:56:34,836 epoch 10 - iter 73/738 - loss 0.25004751 - time (sec): 1.83 - samples/sec: 9649.60 - lr: 0.000003 - momentum: 0.000000
2023-10-18 18:56:36,574 epoch 10 - iter 146/738 - loss 0.26018308 - time (sec): 3.57 - samples/sec: 9518.95 - lr: 0.000003 - momentum: 0.000000
2023-10-18 18:56:38,376 epoch 10 - iter 219/738 - loss 0.27025184 - time (sec): 5.37 - samples/sec: 9358.69 - lr: 0.000002 - momentum: 0.000000
2023-10-18 18:56:40,175 epoch 10 - iter 292/738 - loss 0.27857312 - time (sec): 7.17 - samples/sec: 9197.61 - lr: 0.000002 - momentum: 0.000000
2023-10-18 18:56:41,871 epoch 10 - iter 365/738 - loss 0.28303110 - time (sec): 8.86 - samples/sec: 9296.23 - lr: 0.000002 - momentum: 0.000000
2023-10-18 18:56:43,610 epoch 10 - iter 438/738 - loss 0.28434751 - time (sec): 10.60 - samples/sec: 9421.95 - lr: 0.000001 - momentum: 0.000000
2023-10-18 18:56:45,319 epoch 10 - iter 511/738 - loss 0.28539794 - time (sec): 12.31 - samples/sec: 9386.12 - lr: 0.000001 - momentum: 0.000000
2023-10-18 18:56:47,129 epoch 10 - iter 584/738 - loss 0.28911825 - time (sec): 14.12 - samples/sec: 9443.73 - lr: 0.000001 - momentum: 0.000000
2023-10-18 18:56:48,852 epoch 10 - iter 657/738 - loss 0.28832023 - time (sec): 15.84 - samples/sec: 9435.15 - lr: 0.000000 - momentum: 0.000000
2023-10-18 18:56:50,534 epoch 10 - iter 730/738 - loss 0.28826766 - time (sec): 17.52 - samples/sec: 9417.66 - lr: 0.000000 - momentum: 0.000000
2023-10-18 18:56:50,717 ----------------------------------------------------------------------------------------------------
2023-10-18 18:56:50,718 EPOCH 10 done: loss 0.2885 - lr: 0.000000
2023-10-18 18:56:57,957 DEV : loss 0.2570817172527313 - f1-score (micro avg) 0.4335
2023-10-18 18:56:57,984 saving best model
2023-10-18 18:56:58,046 ----------------------------------------------------------------------------------------------------
2023-10-18 18:56:58,046 Loading model from best epoch ...
2023-10-18 18:56:58,126 SequenceTagger predicts: Dictionary with 21 tags: O, S-loc, B-loc, E-loc, I-loc, S-pers, B-pers, E-pers, I-pers, S-org, B-org, E-org, I-org, S-time, B-time, E-time, I-time, S-prod, B-prod, E-prod, I-prod
2023-10-18 18:57:00,796
Results:
- F-score (micro) 0.4619
- F-score (macro) 0.2331
- Accuracy 0.3203
By class:
precision recall f1-score support
loc 0.4967 0.6935 0.5788 858
pers 0.3154 0.4246 0.3619 537
org 0.1429 0.0076 0.0144 132
time 0.2439 0.1852 0.2105 54
prod 0.0000 0.0000 0.0000 61
micro avg 0.4236 0.5079 0.4619 1642
macro avg 0.2398 0.2622 0.2331 1642
weighted avg 0.3822 0.5079 0.4289 1642
2023-10-18 18:57:00,796 ----------------------------------------------------------------------------------------------------
|