File size: 25,388 Bytes
be84496 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 |
2023-10-10 23:54:05,281 ----------------------------------------------------------------------------------------------------
2023-10-10 23:54:05,283 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): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
(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): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
(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): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
(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): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
(final_layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=1472, out_features=17, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-10 23:54:05,283 ----------------------------------------------------------------------------------------------------
2023-10-10 23:54:05,283 MultiCorpus: 1166 train + 165 dev + 415 test sentences
- NER_HIPE_2022 Corpus: 1166 train + 165 dev + 415 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/fi/with_doc_seperator
2023-10-10 23:54:05,283 ----------------------------------------------------------------------------------------------------
2023-10-10 23:54:05,283 Train: 1166 sentences
2023-10-10 23:54:05,283 (train_with_dev=False, train_with_test=False)
2023-10-10 23:54:05,283 ----------------------------------------------------------------------------------------------------
2023-10-10 23:54:05,283 Training Params:
2023-10-10 23:54:05,283 - learning_rate: "0.00016"
2023-10-10 23:54:05,284 - mini_batch_size: "4"
2023-10-10 23:54:05,284 - max_epochs: "10"
2023-10-10 23:54:05,284 - shuffle: "True"
2023-10-10 23:54:05,284 ----------------------------------------------------------------------------------------------------
2023-10-10 23:54:05,284 Plugins:
2023-10-10 23:54:05,284 - TensorboardLogger
2023-10-10 23:54:05,284 - LinearScheduler | warmup_fraction: '0.1'
2023-10-10 23:54:05,284 ----------------------------------------------------------------------------------------------------
2023-10-10 23:54:05,284 Final evaluation on model from best epoch (best-model.pt)
2023-10-10 23:54:05,284 - metric: "('micro avg', 'f1-score')"
2023-10-10 23:54:05,284 ----------------------------------------------------------------------------------------------------
2023-10-10 23:54:05,284 Computation:
2023-10-10 23:54:05,284 - compute on device: cuda:0
2023-10-10 23:54:05,284 - embedding storage: none
2023-10-10 23:54:05,284 ----------------------------------------------------------------------------------------------------
2023-10-10 23:54:05,285 Model training base path: "hmbench-newseye/fi-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs4-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-2"
2023-10-10 23:54:05,285 ----------------------------------------------------------------------------------------------------
2023-10-10 23:54:05,285 ----------------------------------------------------------------------------------------------------
2023-10-10 23:54:05,285 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-10 23:54:16,090 epoch 1 - iter 29/292 - loss 2.85292695 - time (sec): 10.80 - samples/sec: 466.34 - lr: 0.000015 - momentum: 0.000000
2023-10-10 23:54:25,760 epoch 1 - iter 58/292 - loss 2.84278330 - time (sec): 20.47 - samples/sec: 443.15 - lr: 0.000031 - momentum: 0.000000
2023-10-10 23:54:36,713 epoch 1 - iter 87/292 - loss 2.81776914 - time (sec): 31.43 - samples/sec: 439.22 - lr: 0.000047 - momentum: 0.000000
2023-10-10 23:54:47,841 epoch 1 - iter 116/292 - loss 2.76831466 - time (sec): 42.55 - samples/sec: 427.25 - lr: 0.000063 - momentum: 0.000000
2023-10-10 23:54:58,018 epoch 1 - iter 145/292 - loss 2.68656274 - time (sec): 52.73 - samples/sec: 415.48 - lr: 0.000079 - momentum: 0.000000
2023-10-10 23:55:08,442 epoch 1 - iter 174/292 - loss 2.57976127 - time (sec): 63.16 - samples/sec: 409.34 - lr: 0.000095 - momentum: 0.000000
2023-10-10 23:55:18,924 epoch 1 - iter 203/292 - loss 2.45236738 - time (sec): 73.64 - samples/sec: 413.26 - lr: 0.000111 - momentum: 0.000000
2023-10-10 23:55:28,867 epoch 1 - iter 232/292 - loss 2.33278661 - time (sec): 83.58 - samples/sec: 414.64 - lr: 0.000127 - momentum: 0.000000
2023-10-10 23:55:39,838 epoch 1 - iter 261/292 - loss 2.18255479 - time (sec): 94.55 - samples/sec: 419.36 - lr: 0.000142 - momentum: 0.000000
2023-10-10 23:55:50,363 epoch 1 - iter 290/292 - loss 2.04497265 - time (sec): 105.08 - samples/sec: 421.95 - lr: 0.000158 - momentum: 0.000000
2023-10-10 23:55:50,796 ----------------------------------------------------------------------------------------------------
2023-10-10 23:55:50,797 EPOCH 1 done: loss 2.0427 - lr: 0.000158
2023-10-10 23:55:56,831 DEV : loss 0.6711924076080322 - f1-score (micro avg) 0.0
2023-10-10 23:55:56,840 ----------------------------------------------------------------------------------------------------
2023-10-10 23:56:06,264 epoch 2 - iter 29/292 - loss 0.72788195 - time (sec): 9.42 - samples/sec: 460.20 - lr: 0.000158 - momentum: 0.000000
2023-10-10 23:56:16,572 epoch 2 - iter 58/292 - loss 0.65066851 - time (sec): 19.73 - samples/sec: 462.75 - lr: 0.000157 - momentum: 0.000000
2023-10-10 23:56:27,835 epoch 2 - iter 87/292 - loss 0.64893825 - time (sec): 30.99 - samples/sec: 455.63 - lr: 0.000155 - momentum: 0.000000
2023-10-10 23:56:38,177 epoch 2 - iter 116/292 - loss 0.63051086 - time (sec): 41.33 - samples/sec: 429.76 - lr: 0.000153 - momentum: 0.000000
2023-10-10 23:56:49,107 epoch 2 - iter 145/292 - loss 0.59200239 - time (sec): 52.27 - samples/sec: 426.17 - lr: 0.000151 - momentum: 0.000000
2023-10-10 23:56:58,355 epoch 2 - iter 174/292 - loss 0.59108378 - time (sec): 61.51 - samples/sec: 414.48 - lr: 0.000149 - momentum: 0.000000
2023-10-10 23:57:08,769 epoch 2 - iter 203/292 - loss 0.56265754 - time (sec): 71.93 - samples/sec: 422.57 - lr: 0.000148 - momentum: 0.000000
2023-10-10 23:57:18,903 epoch 2 - iter 232/292 - loss 0.52815545 - time (sec): 82.06 - samples/sec: 430.30 - lr: 0.000146 - momentum: 0.000000
2023-10-10 23:57:28,177 epoch 2 - iter 261/292 - loss 0.50476180 - time (sec): 91.33 - samples/sec: 432.41 - lr: 0.000144 - momentum: 0.000000
2023-10-10 23:57:38,370 epoch 2 - iter 290/292 - loss 0.51847967 - time (sec): 101.53 - samples/sec: 435.68 - lr: 0.000142 - momentum: 0.000000
2023-10-10 23:57:38,886 ----------------------------------------------------------------------------------------------------
2023-10-10 23:57:38,887 EPOCH 2 done: loss 0.5176 - lr: 0.000142
2023-10-10 23:57:44,863 DEV : loss 0.3161855638027191 - f1-score (micro avg) 0.0623
2023-10-10 23:57:44,872 saving best model
2023-10-10 23:57:45,807 ----------------------------------------------------------------------------------------------------
2023-10-10 23:57:54,679 epoch 3 - iter 29/292 - loss 0.41070537 - time (sec): 8.87 - samples/sec: 414.46 - lr: 0.000141 - momentum: 0.000000
2023-10-10 23:58:03,954 epoch 3 - iter 58/292 - loss 0.34583803 - time (sec): 18.14 - samples/sec: 442.90 - lr: 0.000139 - momentum: 0.000000
2023-10-10 23:58:13,921 epoch 3 - iter 87/292 - loss 0.41258944 - time (sec): 28.11 - samples/sec: 467.28 - lr: 0.000137 - momentum: 0.000000
2023-10-10 23:58:22,598 epoch 3 - iter 116/292 - loss 0.40555208 - time (sec): 36.79 - samples/sec: 453.78 - lr: 0.000135 - momentum: 0.000000
2023-10-10 23:58:32,658 epoch 3 - iter 145/292 - loss 0.38549466 - time (sec): 46.85 - samples/sec: 459.81 - lr: 0.000133 - momentum: 0.000000
2023-10-10 23:58:41,945 epoch 3 - iter 174/292 - loss 0.37158746 - time (sec): 56.14 - samples/sec: 457.91 - lr: 0.000132 - momentum: 0.000000
2023-10-10 23:58:51,749 epoch 3 - iter 203/292 - loss 0.35890713 - time (sec): 65.94 - samples/sec: 458.28 - lr: 0.000130 - momentum: 0.000000
2023-10-10 23:59:01,562 epoch 3 - iter 232/292 - loss 0.34783392 - time (sec): 75.75 - samples/sec: 460.75 - lr: 0.000128 - momentum: 0.000000
2023-10-10 23:59:11,656 epoch 3 - iter 261/292 - loss 0.34033929 - time (sec): 85.85 - samples/sec: 457.48 - lr: 0.000126 - momentum: 0.000000
2023-10-10 23:59:21,987 epoch 3 - iter 290/292 - loss 0.33163979 - time (sec): 96.18 - samples/sec: 459.69 - lr: 0.000125 - momentum: 0.000000
2023-10-10 23:59:22,517 ----------------------------------------------------------------------------------------------------
2023-10-10 23:59:22,517 EPOCH 3 done: loss 0.3376 - lr: 0.000125
2023-10-10 23:59:28,109 DEV : loss 0.25150060653686523 - f1-score (micro avg) 0.2521
2023-10-10 23:59:28,118 saving best model
2023-10-10 23:59:34,996 ----------------------------------------------------------------------------------------------------
2023-10-10 23:59:44,938 epoch 4 - iter 29/292 - loss 0.27546017 - time (sec): 9.94 - samples/sec: 417.80 - lr: 0.000123 - momentum: 0.000000
2023-10-10 23:59:55,490 epoch 4 - iter 58/292 - loss 0.35419864 - time (sec): 20.49 - samples/sec: 421.57 - lr: 0.000121 - momentum: 0.000000
2023-10-11 00:00:05,886 epoch 4 - iter 87/292 - loss 0.28871423 - time (sec): 30.89 - samples/sec: 423.34 - lr: 0.000119 - momentum: 0.000000
2023-10-11 00:00:16,538 epoch 4 - iter 116/292 - loss 0.27766491 - time (sec): 41.54 - samples/sec: 420.92 - lr: 0.000117 - momentum: 0.000000
2023-10-11 00:00:26,064 epoch 4 - iter 145/292 - loss 0.27300843 - time (sec): 51.06 - samples/sec: 417.19 - lr: 0.000116 - momentum: 0.000000
2023-10-11 00:00:36,178 epoch 4 - iter 174/292 - loss 0.26748700 - time (sec): 61.18 - samples/sec: 418.89 - lr: 0.000114 - momentum: 0.000000
2023-10-11 00:00:45,939 epoch 4 - iter 203/292 - loss 0.25642219 - time (sec): 70.94 - samples/sec: 425.81 - lr: 0.000112 - momentum: 0.000000
2023-10-11 00:00:55,685 epoch 4 - iter 232/292 - loss 0.25338406 - time (sec): 80.68 - samples/sec: 426.43 - lr: 0.000110 - momentum: 0.000000
2023-10-11 00:01:06,318 epoch 4 - iter 261/292 - loss 0.25874665 - time (sec): 91.32 - samples/sec: 429.89 - lr: 0.000109 - momentum: 0.000000
2023-10-11 00:01:18,053 epoch 4 - iter 290/292 - loss 0.25322113 - time (sec): 103.05 - samples/sec: 430.08 - lr: 0.000107 - momentum: 0.000000
2023-10-11 00:01:18,507 ----------------------------------------------------------------------------------------------------
2023-10-11 00:01:18,508 EPOCH 4 done: loss 0.2532 - lr: 0.000107
2023-10-11 00:01:24,255 DEV : loss 0.19647738337516785 - f1-score (micro avg) 0.4458
2023-10-11 00:01:24,264 saving best model
2023-10-11 00:01:30,995 ----------------------------------------------------------------------------------------------------
2023-10-11 00:01:41,291 epoch 5 - iter 29/292 - loss 0.20651694 - time (sec): 10.29 - samples/sec: 418.24 - lr: 0.000105 - momentum: 0.000000
2023-10-11 00:01:52,132 epoch 5 - iter 58/292 - loss 0.17543958 - time (sec): 21.13 - samples/sec: 428.09 - lr: 0.000103 - momentum: 0.000000
2023-10-11 00:02:02,418 epoch 5 - iter 87/292 - loss 0.17084180 - time (sec): 31.42 - samples/sec: 420.60 - lr: 0.000101 - momentum: 0.000000
2023-10-11 00:02:12,891 epoch 5 - iter 116/292 - loss 0.16868452 - time (sec): 41.89 - samples/sec: 421.77 - lr: 0.000100 - momentum: 0.000000
2023-10-11 00:02:22,341 epoch 5 - iter 145/292 - loss 0.16960028 - time (sec): 51.34 - samples/sec: 422.58 - lr: 0.000098 - momentum: 0.000000
2023-10-11 00:02:33,183 epoch 5 - iter 174/292 - loss 0.18355358 - time (sec): 62.18 - samples/sec: 440.39 - lr: 0.000096 - momentum: 0.000000
2023-10-11 00:02:42,823 epoch 5 - iter 203/292 - loss 0.17979713 - time (sec): 71.82 - samples/sec: 443.66 - lr: 0.000094 - momentum: 0.000000
2023-10-11 00:02:53,349 epoch 5 - iter 232/292 - loss 0.17660361 - time (sec): 82.35 - samples/sec: 442.96 - lr: 0.000093 - momentum: 0.000000
2023-10-11 00:03:02,616 epoch 5 - iter 261/292 - loss 0.17489136 - time (sec): 91.62 - samples/sec: 438.13 - lr: 0.000091 - momentum: 0.000000
2023-10-11 00:03:12,400 epoch 5 - iter 290/292 - loss 0.17466389 - time (sec): 101.40 - samples/sec: 436.72 - lr: 0.000089 - momentum: 0.000000
2023-10-11 00:03:12,893 ----------------------------------------------------------------------------------------------------
2023-10-11 00:03:12,893 EPOCH 5 done: loss 0.1748 - lr: 0.000089
2023-10-11 00:03:18,827 DEV : loss 0.16777318716049194 - f1-score (micro avg) 0.5745
2023-10-11 00:03:18,837 saving best model
2023-10-11 00:03:26,400 ----------------------------------------------------------------------------------------------------
2023-10-11 00:03:36,472 epoch 6 - iter 29/292 - loss 0.11628886 - time (sec): 10.07 - samples/sec: 478.62 - lr: 0.000087 - momentum: 0.000000
2023-10-11 00:03:45,925 epoch 6 - iter 58/292 - loss 0.12246444 - time (sec): 19.52 - samples/sec: 460.68 - lr: 0.000085 - momentum: 0.000000
2023-10-11 00:03:55,483 epoch 6 - iter 87/292 - loss 0.11954585 - time (sec): 29.08 - samples/sec: 456.69 - lr: 0.000084 - momentum: 0.000000
2023-10-11 00:04:05,383 epoch 6 - iter 116/292 - loss 0.12489577 - time (sec): 38.98 - samples/sec: 449.30 - lr: 0.000082 - momentum: 0.000000
2023-10-11 00:04:14,806 epoch 6 - iter 145/292 - loss 0.12818154 - time (sec): 48.40 - samples/sec: 450.02 - lr: 0.000080 - momentum: 0.000000
2023-10-11 00:04:24,229 epoch 6 - iter 174/292 - loss 0.13073488 - time (sec): 57.82 - samples/sec: 448.58 - lr: 0.000078 - momentum: 0.000000
2023-10-11 00:04:34,811 epoch 6 - iter 203/292 - loss 0.13015890 - time (sec): 68.41 - samples/sec: 457.91 - lr: 0.000077 - momentum: 0.000000
2023-10-11 00:04:44,607 epoch 6 - iter 232/292 - loss 0.13300589 - time (sec): 78.20 - samples/sec: 452.20 - lr: 0.000075 - momentum: 0.000000
2023-10-11 00:04:54,487 epoch 6 - iter 261/292 - loss 0.12882566 - time (sec): 88.08 - samples/sec: 451.32 - lr: 0.000073 - momentum: 0.000000
2023-10-11 00:05:03,962 epoch 6 - iter 290/292 - loss 0.12651322 - time (sec): 97.56 - samples/sec: 451.68 - lr: 0.000071 - momentum: 0.000000
2023-10-11 00:05:04,643 ----------------------------------------------------------------------------------------------------
2023-10-11 00:05:04,643 EPOCH 6 done: loss 0.1256 - lr: 0.000071
2023-10-11 00:05:10,396 DEV : loss 0.16338036954402924 - f1-score (micro avg) 0.6681
2023-10-11 00:05:10,405 saving best model
2023-10-11 00:05:17,046 ----------------------------------------------------------------------------------------------------
2023-10-11 00:05:26,299 epoch 7 - iter 29/292 - loss 0.10546819 - time (sec): 9.25 - samples/sec: 474.01 - lr: 0.000069 - momentum: 0.000000
2023-10-11 00:05:36,757 epoch 7 - iter 58/292 - loss 0.09332528 - time (sec): 19.71 - samples/sec: 479.39 - lr: 0.000068 - momentum: 0.000000
2023-10-11 00:05:45,596 epoch 7 - iter 87/292 - loss 0.08784883 - time (sec): 28.55 - samples/sec: 458.94 - lr: 0.000066 - momentum: 0.000000
2023-10-11 00:05:55,071 epoch 7 - iter 116/292 - loss 0.09824998 - time (sec): 38.02 - samples/sec: 454.65 - lr: 0.000064 - momentum: 0.000000
2023-10-11 00:06:04,287 epoch 7 - iter 145/292 - loss 0.09962287 - time (sec): 47.24 - samples/sec: 437.76 - lr: 0.000062 - momentum: 0.000000
2023-10-11 00:06:14,804 epoch 7 - iter 174/292 - loss 0.09942096 - time (sec): 57.75 - samples/sec: 440.28 - lr: 0.000061 - momentum: 0.000000
2023-10-11 00:06:25,173 epoch 7 - iter 203/292 - loss 0.09994166 - time (sec): 68.12 - samples/sec: 444.26 - lr: 0.000059 - momentum: 0.000000
2023-10-11 00:06:35,887 epoch 7 - iter 232/292 - loss 0.09925328 - time (sec): 78.84 - samples/sec: 445.93 - lr: 0.000057 - momentum: 0.000000
2023-10-11 00:06:45,736 epoch 7 - iter 261/292 - loss 0.09753226 - time (sec): 88.69 - samples/sec: 445.91 - lr: 0.000055 - momentum: 0.000000
2023-10-11 00:06:56,594 epoch 7 - iter 290/292 - loss 0.09556655 - time (sec): 99.54 - samples/sec: 444.81 - lr: 0.000054 - momentum: 0.000000
2023-10-11 00:06:57,108 ----------------------------------------------------------------------------------------------------
2023-10-11 00:06:57,109 EPOCH 7 done: loss 0.0954 - lr: 0.000054
2023-10-11 00:07:03,477 DEV : loss 0.15695013105869293 - f1-score (micro avg) 0.7137
2023-10-11 00:07:03,487 saving best model
2023-10-11 00:07:11,348 ----------------------------------------------------------------------------------------------------
2023-10-11 00:07:22,503 epoch 8 - iter 29/292 - loss 0.07758372 - time (sec): 11.15 - samples/sec: 390.65 - lr: 0.000052 - momentum: 0.000000
2023-10-11 00:07:32,880 epoch 8 - iter 58/292 - loss 0.09159419 - time (sec): 21.53 - samples/sec: 397.40 - lr: 0.000050 - momentum: 0.000000
2023-10-11 00:07:44,317 epoch 8 - iter 87/292 - loss 0.07999343 - time (sec): 32.96 - samples/sec: 396.79 - lr: 0.000048 - momentum: 0.000000
2023-10-11 00:07:55,122 epoch 8 - iter 116/292 - loss 0.08349239 - time (sec): 43.77 - samples/sec: 399.24 - lr: 0.000046 - momentum: 0.000000
2023-10-11 00:08:05,305 epoch 8 - iter 145/292 - loss 0.08024406 - time (sec): 53.95 - samples/sec: 404.48 - lr: 0.000045 - momentum: 0.000000
2023-10-11 00:08:14,966 epoch 8 - iter 174/292 - loss 0.08260764 - time (sec): 63.61 - samples/sec: 397.69 - lr: 0.000043 - momentum: 0.000000
2023-10-11 00:08:26,513 epoch 8 - iter 203/292 - loss 0.07913294 - time (sec): 75.16 - samples/sec: 407.83 - lr: 0.000041 - momentum: 0.000000
2023-10-11 00:08:36,850 epoch 8 - iter 232/292 - loss 0.07779059 - time (sec): 85.50 - samples/sec: 406.23 - lr: 0.000039 - momentum: 0.000000
2023-10-11 00:08:47,631 epoch 8 - iter 261/292 - loss 0.07763753 - time (sec): 96.28 - samples/sec: 415.71 - lr: 0.000038 - momentum: 0.000000
2023-10-11 00:08:57,163 epoch 8 - iter 290/292 - loss 0.07708456 - time (sec): 105.81 - samples/sec: 418.87 - lr: 0.000036 - momentum: 0.000000
2023-10-11 00:08:57,561 ----------------------------------------------------------------------------------------------------
2023-10-11 00:08:57,562 EPOCH 8 done: loss 0.0769 - lr: 0.000036
2023-10-11 00:09:03,594 DEV : loss 0.15200063586235046 - f1-score (micro avg) 0.7158
2023-10-11 00:09:03,603 saving best model
2023-10-11 00:09:10,900 ----------------------------------------------------------------------------------------------------
2023-10-11 00:09:20,671 epoch 9 - iter 29/292 - loss 0.07231048 - time (sec): 9.77 - samples/sec: 450.13 - lr: 0.000034 - momentum: 0.000000
2023-10-11 00:09:29,933 epoch 9 - iter 58/292 - loss 0.06442967 - time (sec): 19.03 - samples/sec: 445.71 - lr: 0.000032 - momentum: 0.000000
2023-10-11 00:09:38,821 epoch 9 - iter 87/292 - loss 0.07412397 - time (sec): 27.92 - samples/sec: 431.82 - lr: 0.000030 - momentum: 0.000000
2023-10-11 00:09:49,204 epoch 9 - iter 116/292 - loss 0.06959765 - time (sec): 38.30 - samples/sec: 441.31 - lr: 0.000029 - momentum: 0.000000
2023-10-11 00:10:00,330 epoch 9 - iter 145/292 - loss 0.06728567 - time (sec): 49.43 - samples/sec: 448.23 - lr: 0.000027 - momentum: 0.000000
2023-10-11 00:10:09,929 epoch 9 - iter 174/292 - loss 0.06657777 - time (sec): 59.03 - samples/sec: 442.53 - lr: 0.000025 - momentum: 0.000000
2023-10-11 00:10:19,909 epoch 9 - iter 203/292 - loss 0.06542487 - time (sec): 69.01 - samples/sec: 446.84 - lr: 0.000023 - momentum: 0.000000
2023-10-11 00:10:29,409 epoch 9 - iter 232/292 - loss 0.06463835 - time (sec): 78.51 - samples/sec: 444.23 - lr: 0.000022 - momentum: 0.000000
2023-10-11 00:10:39,706 epoch 9 - iter 261/292 - loss 0.06426085 - time (sec): 88.80 - samples/sec: 446.58 - lr: 0.000020 - momentum: 0.000000
2023-10-11 00:10:50,193 epoch 9 - iter 290/292 - loss 0.06250526 - time (sec): 99.29 - samples/sec: 446.07 - lr: 0.000018 - momentum: 0.000000
2023-10-11 00:10:50,659 ----------------------------------------------------------------------------------------------------
2023-10-11 00:10:50,659 EPOCH 9 done: loss 0.0624 - lr: 0.000018
2023-10-11 00:10:56,556 DEV : loss 0.14497257769107819 - f1-score (micro avg) 0.73
2023-10-11 00:10:56,566 saving best model
2023-10-11 00:11:01,168 ----------------------------------------------------------------------------------------------------
2023-10-11 00:11:11,357 epoch 10 - iter 29/292 - loss 0.06224818 - time (sec): 10.18 - samples/sec: 419.09 - lr: 0.000016 - momentum: 0.000000
2023-10-11 00:11:22,403 epoch 10 - iter 58/292 - loss 0.06532578 - time (sec): 21.23 - samples/sec: 425.84 - lr: 0.000014 - momentum: 0.000000
2023-10-11 00:11:33,271 epoch 10 - iter 87/292 - loss 0.05906433 - time (sec): 32.10 - samples/sec: 404.26 - lr: 0.000013 - momentum: 0.000000
2023-10-11 00:11:44,384 epoch 10 - iter 116/292 - loss 0.05573679 - time (sec): 43.21 - samples/sec: 400.72 - lr: 0.000011 - momentum: 0.000000
2023-10-11 00:11:55,480 epoch 10 - iter 145/292 - loss 0.05252369 - time (sec): 54.31 - samples/sec: 396.67 - lr: 0.000009 - momentum: 0.000000
2023-10-11 00:12:07,290 epoch 10 - iter 174/292 - loss 0.05442054 - time (sec): 66.12 - samples/sec: 398.00 - lr: 0.000007 - momentum: 0.000000
2023-10-11 00:12:19,426 epoch 10 - iter 203/292 - loss 0.05703866 - time (sec): 78.25 - samples/sec: 400.09 - lr: 0.000006 - momentum: 0.000000
2023-10-11 00:12:30,207 epoch 10 - iter 232/292 - loss 0.05634774 - time (sec): 89.03 - samples/sec: 394.96 - lr: 0.000004 - momentum: 0.000000
2023-10-11 00:12:42,095 epoch 10 - iter 261/292 - loss 0.05627346 - time (sec): 100.92 - samples/sec: 399.12 - lr: 0.000002 - momentum: 0.000000
2023-10-11 00:12:52,967 epoch 10 - iter 290/292 - loss 0.05708459 - time (sec): 111.79 - samples/sec: 396.06 - lr: 0.000000 - momentum: 0.000000
2023-10-11 00:12:53,475 ----------------------------------------------------------------------------------------------------
2023-10-11 00:12:53,476 EPOCH 10 done: loss 0.0569 - lr: 0.000000
2023-10-11 00:12:59,917 DEV : loss 0.14803124964237213 - f1-score (micro avg) 0.7406
2023-10-11 00:12:59,927 saving best model
2023-10-11 00:13:03,728 ----------------------------------------------------------------------------------------------------
2023-10-11 00:13:03,730 Loading model from best epoch ...
2023-10-11 00:13:07,606 SequenceTagger predicts: Dictionary with 17 tags: O, S-LOC, B-LOC, E-LOC, I-LOC, S-PER, B-PER, E-PER, I-PER, S-ORG, B-ORG, E-ORG, I-ORG, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd
2023-10-11 00:13:22,595
Results:
- F-score (micro) 0.6983
- F-score (macro) 0.6233
- Accuracy 0.5556
By class:
precision recall f1-score support
PER 0.7487 0.8305 0.7875 348
LOC 0.5710 0.7395 0.6444 261
ORG 0.3830 0.3462 0.3636 52
HumanProd 0.7143 0.6818 0.6977 22
micro avg 0.6503 0.7540 0.6983 683
macro avg 0.6042 0.6495 0.6233 683
weighted avg 0.6518 0.7540 0.6976 683
2023-10-11 00:13:22,595 ----------------------------------------------------------------------------------------------------
|