File size: 25,153 Bytes
1ab8560 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 |
2023-10-06 15:36:54,881 ----------------------------------------------------------------------------------------------------
2023-10-06 15:36:54,882 Model: "SequenceTagger(
(embeddings): ByT5Embeddings(
(model): T5EncoderModel(
(shared): Embedding(384, 1472)
(encoder): T5Stack(
(embed_tokens): Embedding(384, 1472)
(block): ModuleList(
(0): T5Block(
(layer): ModuleList(
(0): T5LayerSelfAttention(
(SelfAttention): T5Attention(
(q): Linear(in_features=1472, out_features=384, bias=False)
(k): Linear(in_features=1472, out_features=384, bias=False)
(v): Linear(in_features=1472, out_features=384, bias=False)
(o): Linear(in_features=384, out_features=1472, bias=False)
(relative_attention_bias): Embedding(32, 6)
)
(layer_norm): T5LayerNorm()
(dropout): Dropout(p=0.1, inplace=False)
)
(1): T5LayerFF(
(DenseReluDense): T5DenseGatedActDense(
(wi_0): Linear(in_features=1472, out_features=3584, bias=False)
(wi_1): Linear(in_features=1472, out_features=3584, bias=False)
(wo): Linear(in_features=3584, out_features=1472, bias=False)
(dropout): Dropout(p=0.1, inplace=False)
(act): NewGELUActivation()
)
(layer_norm): T5LayerNorm()
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
(1-11): 11 x T5Block(
(layer): ModuleList(
(0): T5LayerSelfAttention(
(SelfAttention): T5Attention(
(q): Linear(in_features=1472, out_features=384, bias=False)
(k): Linear(in_features=1472, out_features=384, bias=False)
(v): Linear(in_features=1472, out_features=384, bias=False)
(o): Linear(in_features=384, out_features=1472, bias=False)
)
(layer_norm): T5LayerNorm()
(dropout): Dropout(p=0.1, inplace=False)
)
(1): T5LayerFF(
(DenseReluDense): T5DenseGatedActDense(
(wi_0): Linear(in_features=1472, out_features=3584, bias=False)
(wi_1): Linear(in_features=1472, out_features=3584, bias=False)
(wo): Linear(in_features=3584, out_features=1472, bias=False)
(dropout): Dropout(p=0.1, inplace=False)
(act): NewGELUActivation()
)
(layer_norm): T5LayerNorm()
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
(final_layer_norm): T5LayerNorm()
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=1472, out_features=25, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-06 15:36:54,882 ----------------------------------------------------------------------------------------------------
2023-10-06 15:36:54,882 MultiCorpus: 1214 train + 266 dev + 251 test sentences
- NER_HIPE_2022 Corpus: 1214 train + 266 dev + 251 test sentences - /app/.flair/datasets/ner_hipe_2022/v2.1/ajmc/en/with_doc_seperator
2023-10-06 15:36:54,882 ----------------------------------------------------------------------------------------------------
2023-10-06 15:36:54,882 Train: 1214 sentences
2023-10-06 15:36:54,882 (train_with_dev=False, train_with_test=False)
2023-10-06 15:36:54,883 ----------------------------------------------------------------------------------------------------
2023-10-06 15:36:54,883 Training Params:
2023-10-06 15:36:54,883 - learning_rate: "0.00016"
2023-10-06 15:36:54,883 - mini_batch_size: "4"
2023-10-06 15:36:54,883 - max_epochs: "10"
2023-10-06 15:36:54,883 - shuffle: "True"
2023-10-06 15:36:54,883 ----------------------------------------------------------------------------------------------------
2023-10-06 15:36:54,883 Plugins:
2023-10-06 15:36:54,883 - TensorboardLogger
2023-10-06 15:36:54,883 - LinearScheduler | warmup_fraction: '0.1'
2023-10-06 15:36:54,883 ----------------------------------------------------------------------------------------------------
2023-10-06 15:36:54,883 Final evaluation on model from best epoch (best-model.pt)
2023-10-06 15:36:54,883 - metric: "('micro avg', 'f1-score')"
2023-10-06 15:36:54,883 ----------------------------------------------------------------------------------------------------
2023-10-06 15:36:54,883 Computation:
2023-10-06 15:36:54,883 - compute on device: cuda:0
2023-10-06 15:36:54,883 - embedding storage: none
2023-10-06 15:36:54,884 ----------------------------------------------------------------------------------------------------
2023-10-06 15:36:54,884 Model training base path: "hmbench-ajmc/en-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs4-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-5"
2023-10-06 15:36:54,884 ----------------------------------------------------------------------------------------------------
2023-10-06 15:36:54,884 ----------------------------------------------------------------------------------------------------
2023-10-06 15:36:54,884 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-06 15:37:06,733 epoch 1 - iter 30/304 - loss 3.23322094 - time (sec): 11.85 - samples/sec: 269.84 - lr: 0.000015 - momentum: 0.000000
2023-10-06 15:37:18,012 epoch 1 - iter 60/304 - loss 3.22188262 - time (sec): 23.13 - samples/sec: 266.48 - lr: 0.000031 - momentum: 0.000000
2023-10-06 15:37:29,580 epoch 1 - iter 90/304 - loss 3.19289241 - time (sec): 34.69 - samples/sec: 258.25 - lr: 0.000047 - momentum: 0.000000
2023-10-06 15:37:41,327 epoch 1 - iter 120/304 - loss 3.12149688 - time (sec): 46.44 - samples/sec: 256.97 - lr: 0.000063 - momentum: 0.000000
2023-10-06 15:37:53,113 epoch 1 - iter 150/304 - loss 3.01764722 - time (sec): 58.23 - samples/sec: 255.74 - lr: 0.000078 - momentum: 0.000000
2023-10-06 15:38:05,859 epoch 1 - iter 180/304 - loss 2.88619671 - time (sec): 70.97 - samples/sec: 258.03 - lr: 0.000094 - momentum: 0.000000
2023-10-06 15:38:16,746 epoch 1 - iter 210/304 - loss 2.77838030 - time (sec): 81.86 - samples/sec: 254.96 - lr: 0.000110 - momentum: 0.000000
2023-10-06 15:38:29,102 epoch 1 - iter 240/304 - loss 2.63128465 - time (sec): 94.22 - samples/sec: 255.26 - lr: 0.000126 - momentum: 0.000000
2023-10-06 15:38:41,231 epoch 1 - iter 270/304 - loss 2.47848072 - time (sec): 106.35 - samples/sec: 255.99 - lr: 0.000142 - momentum: 0.000000
2023-10-06 15:38:53,817 epoch 1 - iter 300/304 - loss 2.32263470 - time (sec): 118.93 - samples/sec: 257.03 - lr: 0.000157 - momentum: 0.000000
2023-10-06 15:38:55,380 ----------------------------------------------------------------------------------------------------
2023-10-06 15:38:55,380 EPOCH 1 done: loss 2.3027 - lr: 0.000157
2023-10-06 15:39:03,100 DEV : loss 0.8800114989280701 - f1-score (micro avg) 0.0
2023-10-06 15:39:03,107 ----------------------------------------------------------------------------------------------------
2023-10-06 15:39:15,065 epoch 2 - iter 30/304 - loss 0.78322681 - time (sec): 11.96 - samples/sec: 256.26 - lr: 0.000158 - momentum: 0.000000
2023-10-06 15:39:26,739 epoch 2 - iter 60/304 - loss 0.72767057 - time (sec): 23.63 - samples/sec: 258.01 - lr: 0.000157 - momentum: 0.000000
2023-10-06 15:39:39,354 epoch 2 - iter 90/304 - loss 0.71491699 - time (sec): 36.25 - samples/sec: 256.99 - lr: 0.000155 - momentum: 0.000000
2023-10-06 15:39:51,537 epoch 2 - iter 120/304 - loss 0.66298949 - time (sec): 48.43 - samples/sec: 258.50 - lr: 0.000153 - momentum: 0.000000
2023-10-06 15:40:01,824 epoch 2 - iter 150/304 - loss 0.62149707 - time (sec): 58.72 - samples/sec: 256.76 - lr: 0.000151 - momentum: 0.000000
2023-10-06 15:40:12,871 epoch 2 - iter 180/304 - loss 0.57555139 - time (sec): 69.76 - samples/sec: 259.55 - lr: 0.000150 - momentum: 0.000000
2023-10-06 15:40:24,343 epoch 2 - iter 210/304 - loss 0.55783175 - time (sec): 81.23 - samples/sec: 260.05 - lr: 0.000148 - momentum: 0.000000
2023-10-06 15:40:35,599 epoch 2 - iter 240/304 - loss 0.52723917 - time (sec): 92.49 - samples/sec: 262.39 - lr: 0.000146 - momentum: 0.000000
2023-10-06 15:40:47,204 epoch 2 - iter 270/304 - loss 0.50036839 - time (sec): 104.10 - samples/sec: 264.01 - lr: 0.000144 - momentum: 0.000000
2023-10-06 15:40:58,458 epoch 2 - iter 300/304 - loss 0.47802664 - time (sec): 115.35 - samples/sec: 265.43 - lr: 0.000143 - momentum: 0.000000
2023-10-06 15:40:59,845 ----------------------------------------------------------------------------------------------------
2023-10-06 15:40:59,845 EPOCH 2 done: loss 0.4776 - lr: 0.000143
2023-10-06 15:41:06,852 DEV : loss 0.3180970847606659 - f1-score (micro avg) 0.4756
2023-10-06 15:41:06,858 saving best model
2023-10-06 15:41:07,687 ----------------------------------------------------------------------------------------------------
2023-10-06 15:41:18,649 epoch 3 - iter 30/304 - loss 0.31951679 - time (sec): 10.96 - samples/sec: 263.60 - lr: 0.000141 - momentum: 0.000000
2023-10-06 15:41:29,646 epoch 3 - iter 60/304 - loss 0.27399777 - time (sec): 21.96 - samples/sec: 266.88 - lr: 0.000139 - momentum: 0.000000
2023-10-06 15:41:40,074 epoch 3 - iter 90/304 - loss 0.24835641 - time (sec): 32.38 - samples/sec: 265.71 - lr: 0.000137 - momentum: 0.000000
2023-10-06 15:41:51,842 epoch 3 - iter 120/304 - loss 0.24689103 - time (sec): 44.15 - samples/sec: 271.15 - lr: 0.000135 - momentum: 0.000000
2023-10-06 15:42:02,829 epoch 3 - iter 150/304 - loss 0.23461259 - time (sec): 55.14 - samples/sec: 270.08 - lr: 0.000134 - momentum: 0.000000
2023-10-06 15:42:14,182 epoch 3 - iter 180/304 - loss 0.23128760 - time (sec): 66.49 - samples/sec: 270.04 - lr: 0.000132 - momentum: 0.000000
2023-10-06 15:42:25,602 epoch 3 - iter 210/304 - loss 0.22156764 - time (sec): 77.91 - samples/sec: 270.77 - lr: 0.000130 - momentum: 0.000000
2023-10-06 15:42:37,128 epoch 3 - iter 240/304 - loss 0.21366743 - time (sec): 89.44 - samples/sec: 271.39 - lr: 0.000128 - momentum: 0.000000
2023-10-06 15:42:48,275 epoch 3 - iter 270/304 - loss 0.20493924 - time (sec): 100.59 - samples/sec: 271.47 - lr: 0.000127 - momentum: 0.000000
2023-10-06 15:42:59,999 epoch 3 - iter 300/304 - loss 0.19796880 - time (sec): 112.31 - samples/sec: 271.94 - lr: 0.000125 - momentum: 0.000000
2023-10-06 15:43:01,558 ----------------------------------------------------------------------------------------------------
2023-10-06 15:43:01,558 EPOCH 3 done: loss 0.1975 - lr: 0.000125
2023-10-06 15:43:08,750 DEV : loss 0.184345543384552 - f1-score (micro avg) 0.6873
2023-10-06 15:43:08,757 saving best model
2023-10-06 15:43:13,063 ----------------------------------------------------------------------------------------------------
2023-10-06 15:43:24,872 epoch 4 - iter 30/304 - loss 0.15311988 - time (sec): 11.81 - samples/sec: 268.15 - lr: 0.000123 - momentum: 0.000000
2023-10-06 15:43:35,640 epoch 4 - iter 60/304 - loss 0.14346239 - time (sec): 22.57 - samples/sec: 262.24 - lr: 0.000121 - momentum: 0.000000
2023-10-06 15:43:46,925 epoch 4 - iter 90/304 - loss 0.13496503 - time (sec): 33.86 - samples/sec: 264.56 - lr: 0.000119 - momentum: 0.000000
2023-10-06 15:43:58,485 epoch 4 - iter 120/304 - loss 0.13053640 - time (sec): 45.42 - samples/sec: 264.62 - lr: 0.000118 - momentum: 0.000000
2023-10-06 15:44:10,259 epoch 4 - iter 150/304 - loss 0.12961165 - time (sec): 57.19 - samples/sec: 264.22 - lr: 0.000116 - momentum: 0.000000
2023-10-06 15:44:21,869 epoch 4 - iter 180/304 - loss 0.12156178 - time (sec): 68.80 - samples/sec: 264.02 - lr: 0.000114 - momentum: 0.000000
2023-10-06 15:44:33,279 epoch 4 - iter 210/304 - loss 0.11898423 - time (sec): 80.21 - samples/sec: 262.79 - lr: 0.000112 - momentum: 0.000000
2023-10-06 15:44:45,927 epoch 4 - iter 240/304 - loss 0.11525286 - time (sec): 92.86 - samples/sec: 265.18 - lr: 0.000111 - momentum: 0.000000
2023-10-06 15:44:58,587 epoch 4 - iter 270/304 - loss 0.11334956 - time (sec): 105.52 - samples/sec: 264.46 - lr: 0.000109 - momentum: 0.000000
2023-10-06 15:45:09,784 epoch 4 - iter 300/304 - loss 0.10875073 - time (sec): 116.72 - samples/sec: 262.60 - lr: 0.000107 - momentum: 0.000000
2023-10-06 15:45:11,182 ----------------------------------------------------------------------------------------------------
2023-10-06 15:45:11,183 EPOCH 4 done: loss 0.1081 - lr: 0.000107
2023-10-06 15:45:19,184 DEV : loss 0.1357320249080658 - f1-score (micro avg) 0.816
2023-10-06 15:45:19,192 saving best model
2023-10-06 15:45:23,511 ----------------------------------------------------------------------------------------------------
2023-10-06 15:45:35,491 epoch 5 - iter 30/304 - loss 0.04930699 - time (sec): 11.98 - samples/sec: 256.70 - lr: 0.000105 - momentum: 0.000000
2023-10-06 15:45:47,151 epoch 5 - iter 60/304 - loss 0.07126682 - time (sec): 23.64 - samples/sec: 256.67 - lr: 0.000103 - momentum: 0.000000
2023-10-06 15:45:59,609 epoch 5 - iter 90/304 - loss 0.07174582 - time (sec): 36.10 - samples/sec: 256.34 - lr: 0.000102 - momentum: 0.000000
2023-10-06 15:46:11,788 epoch 5 - iter 120/304 - loss 0.06935646 - time (sec): 48.27 - samples/sec: 257.19 - lr: 0.000100 - momentum: 0.000000
2023-10-06 15:46:24,076 epoch 5 - iter 150/304 - loss 0.07373860 - time (sec): 60.56 - samples/sec: 257.80 - lr: 0.000098 - momentum: 0.000000
2023-10-06 15:46:35,831 epoch 5 - iter 180/304 - loss 0.06852216 - time (sec): 72.32 - samples/sec: 256.45 - lr: 0.000096 - momentum: 0.000000
2023-10-06 15:46:48,151 epoch 5 - iter 210/304 - loss 0.06505765 - time (sec): 84.64 - samples/sec: 256.89 - lr: 0.000094 - momentum: 0.000000
2023-10-06 15:47:00,070 epoch 5 - iter 240/304 - loss 0.06581915 - time (sec): 96.56 - samples/sec: 257.94 - lr: 0.000093 - momentum: 0.000000
2023-10-06 15:47:11,797 epoch 5 - iter 270/304 - loss 0.06376206 - time (sec): 108.28 - samples/sec: 257.20 - lr: 0.000091 - momentum: 0.000000
2023-10-06 15:47:23,426 epoch 5 - iter 300/304 - loss 0.06521827 - time (sec): 119.91 - samples/sec: 256.54 - lr: 0.000089 - momentum: 0.000000
2023-10-06 15:47:24,581 ----------------------------------------------------------------------------------------------------
2023-10-06 15:47:24,581 EPOCH 5 done: loss 0.0649 - lr: 0.000089
2023-10-06 15:47:32,312 DEV : loss 0.14172478020191193 - f1-score (micro avg) 0.7991
2023-10-06 15:47:32,318 ----------------------------------------------------------------------------------------------------
2023-10-06 15:47:44,395 epoch 6 - iter 30/304 - loss 0.06900995 - time (sec): 12.08 - samples/sec: 263.68 - lr: 0.000087 - momentum: 0.000000
2023-10-06 15:47:56,682 epoch 6 - iter 60/304 - loss 0.06261892 - time (sec): 24.36 - samples/sec: 264.26 - lr: 0.000085 - momentum: 0.000000
2023-10-06 15:48:09,164 epoch 6 - iter 90/304 - loss 0.05549560 - time (sec): 36.84 - samples/sec: 264.22 - lr: 0.000084 - momentum: 0.000000
2023-10-06 15:48:21,345 epoch 6 - iter 120/304 - loss 0.05286815 - time (sec): 49.03 - samples/sec: 261.99 - lr: 0.000082 - momentum: 0.000000
2023-10-06 15:48:33,228 epoch 6 - iter 150/304 - loss 0.04820134 - time (sec): 60.91 - samples/sec: 261.39 - lr: 0.000080 - momentum: 0.000000
2023-10-06 15:48:44,239 epoch 6 - iter 180/304 - loss 0.04438212 - time (sec): 71.92 - samples/sec: 258.41 - lr: 0.000078 - momentum: 0.000000
2023-10-06 15:48:56,652 epoch 6 - iter 210/304 - loss 0.04595305 - time (sec): 84.33 - samples/sec: 258.91 - lr: 0.000077 - momentum: 0.000000
2023-10-06 15:49:08,204 epoch 6 - iter 240/304 - loss 0.04631832 - time (sec): 95.88 - samples/sec: 257.53 - lr: 0.000075 - momentum: 0.000000
2023-10-06 15:49:20,203 epoch 6 - iter 270/304 - loss 0.04490989 - time (sec): 107.88 - samples/sec: 255.86 - lr: 0.000073 - momentum: 0.000000
2023-10-06 15:49:32,333 epoch 6 - iter 300/304 - loss 0.04721332 - time (sec): 120.01 - samples/sec: 255.61 - lr: 0.000071 - momentum: 0.000000
2023-10-06 15:49:33,641 ----------------------------------------------------------------------------------------------------
2023-10-06 15:49:33,641 EPOCH 6 done: loss 0.0484 - lr: 0.000071
2023-10-06 15:49:41,541 DEV : loss 0.14471642673015594 - f1-score (micro avg) 0.8208
2023-10-06 15:49:41,548 saving best model
2023-10-06 15:49:45,868 ----------------------------------------------------------------------------------------------------
2023-10-06 15:49:57,395 epoch 7 - iter 30/304 - loss 0.03112703 - time (sec): 11.53 - samples/sec: 248.84 - lr: 0.000069 - momentum: 0.000000
2023-10-06 15:50:09,035 epoch 7 - iter 60/304 - loss 0.03500998 - time (sec): 23.16 - samples/sec: 248.39 - lr: 0.000068 - momentum: 0.000000
2023-10-06 15:50:21,004 epoch 7 - iter 90/304 - loss 0.02979522 - time (sec): 35.13 - samples/sec: 249.33 - lr: 0.000066 - momentum: 0.000000
2023-10-06 15:50:32,990 epoch 7 - iter 120/304 - loss 0.03007276 - time (sec): 47.12 - samples/sec: 251.04 - lr: 0.000064 - momentum: 0.000000
2023-10-06 15:50:44,973 epoch 7 - iter 150/304 - loss 0.02904381 - time (sec): 59.10 - samples/sec: 250.68 - lr: 0.000062 - momentum: 0.000000
2023-10-06 15:50:57,542 epoch 7 - iter 180/304 - loss 0.03211479 - time (sec): 71.67 - samples/sec: 252.55 - lr: 0.000061 - momentum: 0.000000
2023-10-06 15:51:09,551 epoch 7 - iter 210/304 - loss 0.03134543 - time (sec): 83.68 - samples/sec: 253.38 - lr: 0.000059 - momentum: 0.000000
2023-10-06 15:51:21,545 epoch 7 - iter 240/304 - loss 0.03216970 - time (sec): 95.67 - samples/sec: 252.58 - lr: 0.000057 - momentum: 0.000000
2023-10-06 15:51:34,108 epoch 7 - iter 270/304 - loss 0.03210527 - time (sec): 108.24 - samples/sec: 254.12 - lr: 0.000055 - momentum: 0.000000
2023-10-06 15:51:46,256 epoch 7 - iter 300/304 - loss 0.03929255 - time (sec): 120.39 - samples/sec: 254.28 - lr: 0.000054 - momentum: 0.000000
2023-10-06 15:51:47,718 ----------------------------------------------------------------------------------------------------
2023-10-06 15:51:47,719 EPOCH 7 done: loss 0.0388 - lr: 0.000054
2023-10-06 15:51:55,710 DEV : loss 0.15623363852500916 - f1-score (micro avg) 0.8316
2023-10-06 15:51:55,718 saving best model
2023-10-06 15:52:00,038 ----------------------------------------------------------------------------------------------------
2023-10-06 15:52:12,017 epoch 8 - iter 30/304 - loss 0.04538668 - time (sec): 11.98 - samples/sec: 252.89 - lr: 0.000052 - momentum: 0.000000
2023-10-06 15:52:23,734 epoch 8 - iter 60/304 - loss 0.04340644 - time (sec): 23.69 - samples/sec: 252.12 - lr: 0.000050 - momentum: 0.000000
2023-10-06 15:52:35,070 epoch 8 - iter 90/304 - loss 0.03841906 - time (sec): 35.03 - samples/sec: 248.27 - lr: 0.000048 - momentum: 0.000000
2023-10-06 15:52:47,018 epoch 8 - iter 120/304 - loss 0.03922015 - time (sec): 46.98 - samples/sec: 250.61 - lr: 0.000046 - momentum: 0.000000
2023-10-06 15:52:59,419 epoch 8 - iter 150/304 - loss 0.03629845 - time (sec): 59.38 - samples/sec: 252.41 - lr: 0.000045 - momentum: 0.000000
2023-10-06 15:53:11,296 epoch 8 - iter 180/304 - loss 0.03414241 - time (sec): 71.26 - samples/sec: 251.94 - lr: 0.000043 - momentum: 0.000000
2023-10-06 15:53:23,608 epoch 8 - iter 210/304 - loss 0.03144849 - time (sec): 83.57 - samples/sec: 253.58 - lr: 0.000041 - momentum: 0.000000
2023-10-06 15:53:36,172 epoch 8 - iter 240/304 - loss 0.03112508 - time (sec): 96.13 - samples/sec: 255.89 - lr: 0.000039 - momentum: 0.000000
2023-10-06 15:53:48,275 epoch 8 - iter 270/304 - loss 0.03257122 - time (sec): 108.24 - samples/sec: 256.62 - lr: 0.000038 - momentum: 0.000000
2023-10-06 15:53:59,628 epoch 8 - iter 300/304 - loss 0.03300308 - time (sec): 119.59 - samples/sec: 255.66 - lr: 0.000036 - momentum: 0.000000
2023-10-06 15:54:01,265 ----------------------------------------------------------------------------------------------------
2023-10-06 15:54:01,266 EPOCH 8 done: loss 0.0329 - lr: 0.000036
2023-10-06 15:54:09,139 DEV : loss 0.15997229516506195 - f1-score (micro avg) 0.8379
2023-10-06 15:54:09,147 saving best model
2023-10-06 15:54:13,485 ----------------------------------------------------------------------------------------------------
2023-10-06 15:54:25,850 epoch 9 - iter 30/304 - loss 0.03849688 - time (sec): 12.36 - samples/sec: 273.29 - lr: 0.000034 - momentum: 0.000000
2023-10-06 15:54:37,879 epoch 9 - iter 60/304 - loss 0.03428552 - time (sec): 24.39 - samples/sec: 269.06 - lr: 0.000032 - momentum: 0.000000
2023-10-06 15:54:49,432 epoch 9 - iter 90/304 - loss 0.03229835 - time (sec): 35.95 - samples/sec: 264.90 - lr: 0.000030 - momentum: 0.000000
2023-10-06 15:55:01,065 epoch 9 - iter 120/304 - loss 0.02975385 - time (sec): 47.58 - samples/sec: 260.62 - lr: 0.000029 - momentum: 0.000000
2023-10-06 15:55:13,022 epoch 9 - iter 150/304 - loss 0.02627761 - time (sec): 59.54 - samples/sec: 257.51 - lr: 0.000027 - momentum: 0.000000
2023-10-06 15:55:25,259 epoch 9 - iter 180/304 - loss 0.02710223 - time (sec): 71.77 - samples/sec: 257.30 - lr: 0.000025 - momentum: 0.000000
2023-10-06 15:55:37,771 epoch 9 - iter 210/304 - loss 0.02532150 - time (sec): 84.28 - samples/sec: 257.25 - lr: 0.000023 - momentum: 0.000000
2023-10-06 15:55:49,592 epoch 9 - iter 240/304 - loss 0.02867009 - time (sec): 96.11 - samples/sec: 255.81 - lr: 0.000022 - momentum: 0.000000
2023-10-06 15:56:01,310 epoch 9 - iter 270/304 - loss 0.02617301 - time (sec): 107.82 - samples/sec: 255.88 - lr: 0.000020 - momentum: 0.000000
2023-10-06 15:56:13,354 epoch 9 - iter 300/304 - loss 0.02569570 - time (sec): 119.87 - samples/sec: 255.38 - lr: 0.000018 - momentum: 0.000000
2023-10-06 15:56:14,735 ----------------------------------------------------------------------------------------------------
2023-10-06 15:56:14,736 EPOCH 9 done: loss 0.0263 - lr: 0.000018
2023-10-06 15:56:22,592 DEV : loss 0.16357897222042084 - f1-score (micro avg) 0.8443
2023-10-06 15:56:22,599 saving best model
2023-10-06 15:56:26,917 ----------------------------------------------------------------------------------------------------
2023-10-06 15:56:38,528 epoch 10 - iter 30/304 - loss 0.03508030 - time (sec): 11.61 - samples/sec: 247.56 - lr: 0.000016 - momentum: 0.000000
2023-10-06 15:56:51,462 epoch 10 - iter 60/304 - loss 0.03055439 - time (sec): 24.54 - samples/sec: 259.78 - lr: 0.000014 - momentum: 0.000000
2023-10-06 15:57:03,159 epoch 10 - iter 90/304 - loss 0.02707793 - time (sec): 36.24 - samples/sec: 258.47 - lr: 0.000013 - momentum: 0.000000
2023-10-06 15:57:14,883 epoch 10 - iter 120/304 - loss 0.02382844 - time (sec): 47.96 - samples/sec: 257.65 - lr: 0.000011 - momentum: 0.000000
2023-10-06 15:57:26,168 epoch 10 - iter 150/304 - loss 0.02697956 - time (sec): 59.25 - samples/sec: 255.31 - lr: 0.000009 - momentum: 0.000000
2023-10-06 15:57:38,175 epoch 10 - iter 180/304 - loss 0.02428988 - time (sec): 71.26 - samples/sec: 256.39 - lr: 0.000007 - momentum: 0.000000
2023-10-06 15:57:49,550 epoch 10 - iter 210/304 - loss 0.02197756 - time (sec): 82.63 - samples/sec: 255.17 - lr: 0.000006 - momentum: 0.000000
2023-10-06 15:58:01,485 epoch 10 - iter 240/304 - loss 0.02195724 - time (sec): 94.57 - samples/sec: 257.10 - lr: 0.000004 - momentum: 0.000000
2023-10-06 15:58:13,044 epoch 10 - iter 270/304 - loss 0.02154227 - time (sec): 106.13 - samples/sec: 259.09 - lr: 0.000002 - momentum: 0.000000
2023-10-06 15:58:24,617 epoch 10 - iter 300/304 - loss 0.02330141 - time (sec): 117.70 - samples/sec: 261.07 - lr: 0.000000 - momentum: 0.000000
2023-10-06 15:58:25,767 ----------------------------------------------------------------------------------------------------
2023-10-06 15:58:25,767 EPOCH 10 done: loss 0.0240 - lr: 0.000000
2023-10-06 15:58:32,973 DEV : loss 0.16605204343795776 - f1-score (micro avg) 0.8426
2023-10-06 15:58:33,817 ----------------------------------------------------------------------------------------------------
2023-10-06 15:58:33,819 Loading model from best epoch ...
2023-10-06 15:58:37,445 SequenceTagger predicts: Dictionary with 25 tags: O, S-scope, B-scope, E-scope, I-scope, S-pers, B-pers, E-pers, I-pers, S-work, B-work, E-work, I-work, S-loc, B-loc, E-loc, I-loc, S-date, B-date, E-date, I-date, S-object, B-object, E-object, I-object
2023-10-06 15:58:44,032
Results:
- F-score (micro) 0.8
- F-score (macro) 0.6473
- Accuracy 0.6744
By class:
precision recall f1-score support
scope 0.7372 0.7616 0.7492 151
pers 0.8198 0.9479 0.8792 96
work 0.7593 0.8632 0.8079 95
loc 1.0000 0.6667 0.8000 3
date 0.0000 0.0000 0.0000 3
micro avg 0.7692 0.8333 0.8000 348
macro avg 0.6633 0.6479 0.6473 348
weighted avg 0.7619 0.8333 0.7951 348
2023-10-06 15:58:44,032 ----------------------------------------------------------------------------------------------------
|