File size: 25,249 Bytes
6b10b52 |
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 |
2023-10-11 03:35:43,703 ----------------------------------------------------------------------------------------------------
2023-10-11 03:35:43,706 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()
)"
2023-10-11 03:35:43,706 ----------------------------------------------------------------------------------------------------
2023-10-11 03:35:43,707 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
2023-10-11 03:35:43,707 ----------------------------------------------------------------------------------------------------
2023-10-11 03:35:43,707 Train: 1166 sentences
2023-10-11 03:35:43,707 (train_with_dev=False, train_with_test=False)
2023-10-11 03:35:43,707 ----------------------------------------------------------------------------------------------------
2023-10-11 03:35:43,707 Training Params:
2023-10-11 03:35:43,707 - learning_rate: "0.00016"
2023-10-11 03:35:43,707 - mini_batch_size: "4"
2023-10-11 03:35:43,707 - max_epochs: "10"
2023-10-11 03:35:43,707 - shuffle: "True"
2023-10-11 03:35:43,707 ----------------------------------------------------------------------------------------------------
2023-10-11 03:35:43,707 Plugins:
2023-10-11 03:35:43,707 - TensorboardLogger
2023-10-11 03:35:43,707 - LinearScheduler | warmup_fraction: '0.1'
2023-10-11 03:35:43,708 ----------------------------------------------------------------------------------------------------
2023-10-11 03:35:43,708 Final evaluation on model from best epoch (best-model.pt)
2023-10-11 03:35:43,708 - metric: "('micro avg', 'f1-score')"
2023-10-11 03:35:43,708 ----------------------------------------------------------------------------------------------------
2023-10-11 03:35:43,708 Computation:
2023-10-11 03:35:43,708 - compute on device: cuda:0
2023-10-11 03:35:43,708 - embedding storage: none
2023-10-11 03:35:43,708 ----------------------------------------------------------------------------------------------------
2023-10-11 03:35:43,708 Model training base path: "hmbench-newseye/fi-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs4-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-5"
2023-10-11 03:35:43,708 ----------------------------------------------------------------------------------------------------
2023-10-11 03:35:43,708 ----------------------------------------------------------------------------------------------------
2023-10-11 03:35:43,708 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-11 03:35:54,860 epoch 1 - iter 29/292 - loss 2.81957970 - time (sec): 11.15 - samples/sec: 386.73 - lr: 0.000015 - momentum: 0.000000
2023-10-11 03:36:04,709 epoch 1 - iter 58/292 - loss 2.80589817 - time (sec): 21.00 - samples/sec: 393.07 - lr: 0.000031 - momentum: 0.000000
2023-10-11 03:36:14,324 epoch 1 - iter 87/292 - loss 2.77951816 - time (sec): 30.61 - samples/sec: 399.20 - lr: 0.000047 - momentum: 0.000000
2023-10-11 03:36:26,204 epoch 1 - iter 116/292 - loss 2.70206612 - time (sec): 42.49 - samples/sec: 411.23 - lr: 0.000063 - momentum: 0.000000
2023-10-11 03:36:35,857 epoch 1 - iter 145/292 - loss 2.61368049 - time (sec): 52.15 - samples/sec: 418.15 - lr: 0.000079 - momentum: 0.000000
2023-10-11 03:36:46,142 epoch 1 - iter 174/292 - loss 2.50306929 - time (sec): 62.43 - samples/sec: 428.31 - lr: 0.000095 - momentum: 0.000000
2023-10-11 03:36:57,372 epoch 1 - iter 203/292 - loss 2.37448119 - time (sec): 73.66 - samples/sec: 438.29 - lr: 0.000111 - momentum: 0.000000
2023-10-11 03:37:06,746 epoch 1 - iter 232/292 - loss 2.26594986 - time (sec): 83.04 - samples/sec: 438.73 - lr: 0.000127 - momentum: 0.000000
2023-10-11 03:37:16,335 epoch 1 - iter 261/292 - loss 2.14370781 - time (sec): 92.63 - samples/sec: 439.30 - lr: 0.000142 - momentum: 0.000000
2023-10-11 03:37:25,653 epoch 1 - iter 290/292 - loss 2.03951691 - time (sec): 101.94 - samples/sec: 434.93 - lr: 0.000158 - momentum: 0.000000
2023-10-11 03:37:26,081 ----------------------------------------------------------------------------------------------------
2023-10-11 03:37:26,082 EPOCH 1 done: loss 2.0362 - lr: 0.000158
2023-10-11 03:37:31,472 DEV : loss 0.6394248604774475 - f1-score (micro avg) 0.0
2023-10-11 03:37:31,482 ----------------------------------------------------------------------------------------------------
2023-10-11 03:37:41,271 epoch 2 - iter 29/292 - loss 0.68301608 - time (sec): 9.79 - samples/sec: 447.23 - lr: 0.000158 - momentum: 0.000000
2023-10-11 03:37:50,950 epoch 2 - iter 58/292 - loss 0.63527545 - time (sec): 19.47 - samples/sec: 440.97 - lr: 0.000157 - momentum: 0.000000
2023-10-11 03:38:00,592 epoch 2 - iter 87/292 - loss 0.63132550 - time (sec): 29.11 - samples/sec: 436.41 - lr: 0.000155 - momentum: 0.000000
2023-10-11 03:38:10,670 epoch 2 - iter 116/292 - loss 0.58196696 - time (sec): 39.19 - samples/sec: 443.97 - lr: 0.000153 - momentum: 0.000000
2023-10-11 03:38:20,932 epoch 2 - iter 145/292 - loss 0.60854308 - time (sec): 49.45 - samples/sec: 450.09 - lr: 0.000151 - momentum: 0.000000
2023-10-11 03:38:31,538 epoch 2 - iter 174/292 - loss 0.57661614 - time (sec): 60.05 - samples/sec: 445.20 - lr: 0.000149 - momentum: 0.000000
2023-10-11 03:38:41,306 epoch 2 - iter 203/292 - loss 0.54824891 - time (sec): 69.82 - samples/sec: 447.23 - lr: 0.000148 - momentum: 0.000000
2023-10-11 03:38:50,123 epoch 2 - iter 232/292 - loss 0.52349347 - time (sec): 78.64 - samples/sec: 443.56 - lr: 0.000146 - momentum: 0.000000
2023-10-11 03:38:59,450 epoch 2 - iter 261/292 - loss 0.51475348 - time (sec): 87.97 - samples/sec: 443.06 - lr: 0.000144 - momentum: 0.000000
2023-10-11 03:39:09,649 epoch 2 - iter 290/292 - loss 0.49892401 - time (sec): 98.16 - samples/sec: 450.27 - lr: 0.000142 - momentum: 0.000000
2023-10-11 03:39:10,190 ----------------------------------------------------------------------------------------------------
2023-10-11 03:39:10,190 EPOCH 2 done: loss 0.4983 - lr: 0.000142
2023-10-11 03:39:16,070 DEV : loss 0.29269105195999146 - f1-score (micro avg) 0.0
2023-10-11 03:39:16,079 ----------------------------------------------------------------------------------------------------
2023-10-11 03:39:25,841 epoch 3 - iter 29/292 - loss 0.35386531 - time (sec): 9.76 - samples/sec: 408.81 - lr: 0.000141 - momentum: 0.000000
2023-10-11 03:39:35,418 epoch 3 - iter 58/292 - loss 0.35190101 - time (sec): 19.34 - samples/sec: 411.64 - lr: 0.000139 - momentum: 0.000000
2023-10-11 03:39:45,503 epoch 3 - iter 87/292 - loss 0.33548860 - time (sec): 29.42 - samples/sec: 425.46 - lr: 0.000137 - momentum: 0.000000
2023-10-11 03:39:55,406 epoch 3 - iter 116/292 - loss 0.31617685 - time (sec): 39.32 - samples/sec: 437.00 - lr: 0.000135 - momentum: 0.000000
2023-10-11 03:40:05,107 epoch 3 - iter 145/292 - loss 0.30835528 - time (sec): 49.03 - samples/sec: 434.75 - lr: 0.000133 - momentum: 0.000000
2023-10-11 03:40:15,147 epoch 3 - iter 174/292 - loss 0.29080066 - time (sec): 59.07 - samples/sec: 441.22 - lr: 0.000132 - momentum: 0.000000
2023-10-11 03:40:25,273 epoch 3 - iter 203/292 - loss 0.30465050 - time (sec): 69.19 - samples/sec: 443.61 - lr: 0.000130 - momentum: 0.000000
2023-10-11 03:40:34,384 epoch 3 - iter 232/292 - loss 0.30244144 - time (sec): 78.30 - samples/sec: 442.32 - lr: 0.000128 - momentum: 0.000000
2023-10-11 03:40:45,048 epoch 3 - iter 261/292 - loss 0.29585035 - time (sec): 88.97 - samples/sec: 446.66 - lr: 0.000126 - momentum: 0.000000
2023-10-11 03:40:55,141 epoch 3 - iter 290/292 - loss 0.29281659 - time (sec): 99.06 - samples/sec: 445.33 - lr: 0.000125 - momentum: 0.000000
2023-10-11 03:40:55,768 ----------------------------------------------------------------------------------------------------
2023-10-11 03:40:55,768 EPOCH 3 done: loss 0.2912 - lr: 0.000125
2023-10-11 03:41:01,875 DEV : loss 0.19170591235160828 - f1-score (micro avg) 0.477
2023-10-11 03:41:01,885 saving best model
2023-10-11 03:41:02,794 ----------------------------------------------------------------------------------------------------
2023-10-11 03:41:12,772 epoch 4 - iter 29/292 - loss 0.19259566 - time (sec): 9.98 - samples/sec: 473.64 - lr: 0.000123 - momentum: 0.000000
2023-10-11 03:41:23,316 epoch 4 - iter 58/292 - loss 0.16708108 - time (sec): 20.52 - samples/sec: 490.02 - lr: 0.000121 - momentum: 0.000000
2023-10-11 03:41:33,104 epoch 4 - iter 87/292 - loss 0.19487845 - time (sec): 30.31 - samples/sec: 484.26 - lr: 0.000119 - momentum: 0.000000
2023-10-11 03:41:42,943 epoch 4 - iter 116/292 - loss 0.19940283 - time (sec): 40.15 - samples/sec: 483.38 - lr: 0.000117 - momentum: 0.000000
2023-10-11 03:41:52,477 epoch 4 - iter 145/292 - loss 0.20218637 - time (sec): 49.68 - samples/sec: 479.92 - lr: 0.000116 - momentum: 0.000000
2023-10-11 03:42:01,818 epoch 4 - iter 174/292 - loss 0.19572403 - time (sec): 59.02 - samples/sec: 475.35 - lr: 0.000114 - momentum: 0.000000
2023-10-11 03:42:10,954 epoch 4 - iter 203/292 - loss 0.19603038 - time (sec): 68.16 - samples/sec: 468.68 - lr: 0.000112 - momentum: 0.000000
2023-10-11 03:42:20,751 epoch 4 - iter 232/292 - loss 0.19062971 - time (sec): 77.95 - samples/sec: 463.96 - lr: 0.000110 - momentum: 0.000000
2023-10-11 03:42:29,640 epoch 4 - iter 261/292 - loss 0.18725691 - time (sec): 86.84 - samples/sec: 455.07 - lr: 0.000109 - momentum: 0.000000
2023-10-11 03:42:39,637 epoch 4 - iter 290/292 - loss 0.18640083 - time (sec): 96.84 - samples/sec: 457.68 - lr: 0.000107 - momentum: 0.000000
2023-10-11 03:42:40,049 ----------------------------------------------------------------------------------------------------
2023-10-11 03:42:40,050 EPOCH 4 done: loss 0.1868 - lr: 0.000107
2023-10-11 03:42:45,788 DEV : loss 0.1456957757472992 - f1-score (micro avg) 0.6624
2023-10-11 03:42:45,798 saving best model
2023-10-11 03:42:46,754 ----------------------------------------------------------------------------------------------------
2023-10-11 03:42:56,570 epoch 5 - iter 29/292 - loss 0.13994720 - time (sec): 9.81 - samples/sec: 456.00 - lr: 0.000105 - momentum: 0.000000
2023-10-11 03:43:06,990 epoch 5 - iter 58/292 - loss 0.11776309 - time (sec): 20.23 - samples/sec: 472.13 - lr: 0.000103 - momentum: 0.000000
2023-10-11 03:43:17,064 epoch 5 - iter 87/292 - loss 0.13368301 - time (sec): 30.31 - samples/sec: 472.38 - lr: 0.000101 - momentum: 0.000000
2023-10-11 03:43:27,262 epoch 5 - iter 116/292 - loss 0.13287921 - time (sec): 40.51 - samples/sec: 471.51 - lr: 0.000100 - momentum: 0.000000
2023-10-11 03:43:36,983 epoch 5 - iter 145/292 - loss 0.13471707 - time (sec): 50.23 - samples/sec: 469.55 - lr: 0.000098 - momentum: 0.000000
2023-10-11 03:43:46,317 epoch 5 - iter 174/292 - loss 0.13352509 - time (sec): 59.56 - samples/sec: 467.45 - lr: 0.000096 - momentum: 0.000000
2023-10-11 03:43:56,006 epoch 5 - iter 203/292 - loss 0.13150760 - time (sec): 69.25 - samples/sec: 458.57 - lr: 0.000094 - momentum: 0.000000
2023-10-11 03:44:04,831 epoch 5 - iter 232/292 - loss 0.12973404 - time (sec): 78.07 - samples/sec: 454.17 - lr: 0.000093 - momentum: 0.000000
2023-10-11 03:44:14,924 epoch 5 - iter 261/292 - loss 0.12458178 - time (sec): 88.17 - samples/sec: 456.46 - lr: 0.000091 - momentum: 0.000000
2023-10-11 03:44:24,115 epoch 5 - iter 290/292 - loss 0.12278117 - time (sec): 97.36 - samples/sec: 455.52 - lr: 0.000089 - momentum: 0.000000
2023-10-11 03:44:24,506 ----------------------------------------------------------------------------------------------------
2023-10-11 03:44:24,506 EPOCH 5 done: loss 0.1229 - lr: 0.000089
2023-10-11 03:44:30,384 DEV : loss 0.12754610180854797 - f1-score (micro avg) 0.7325
2023-10-11 03:44:30,395 saving best model
2023-10-11 03:44:33,053 ----------------------------------------------------------------------------------------------------
2023-10-11 03:44:41,909 epoch 6 - iter 29/292 - loss 0.08298524 - time (sec): 8.85 - samples/sec: 421.75 - lr: 0.000087 - momentum: 0.000000
2023-10-11 03:44:51,064 epoch 6 - iter 58/292 - loss 0.11175773 - time (sec): 18.01 - samples/sec: 432.68 - lr: 0.000085 - momentum: 0.000000
2023-10-11 03:45:00,824 epoch 6 - iter 87/292 - loss 0.10165883 - time (sec): 27.77 - samples/sec: 447.56 - lr: 0.000084 - momentum: 0.000000
2023-10-11 03:45:10,403 epoch 6 - iter 116/292 - loss 0.09678991 - time (sec): 37.35 - samples/sec: 445.89 - lr: 0.000082 - momentum: 0.000000
2023-10-11 03:45:20,476 epoch 6 - iter 145/292 - loss 0.09397512 - time (sec): 47.42 - samples/sec: 455.48 - lr: 0.000080 - momentum: 0.000000
2023-10-11 03:45:30,725 epoch 6 - iter 174/292 - loss 0.08697011 - time (sec): 57.67 - samples/sec: 461.02 - lr: 0.000078 - momentum: 0.000000
2023-10-11 03:45:39,853 epoch 6 - iter 203/292 - loss 0.08810664 - time (sec): 66.80 - samples/sec: 455.49 - lr: 0.000077 - momentum: 0.000000
2023-10-11 03:45:48,933 epoch 6 - iter 232/292 - loss 0.08610263 - time (sec): 75.88 - samples/sec: 448.75 - lr: 0.000075 - momentum: 0.000000
2023-10-11 03:46:00,055 epoch 6 - iter 261/292 - loss 0.08625545 - time (sec): 87.00 - samples/sec: 453.30 - lr: 0.000073 - momentum: 0.000000
2023-10-11 03:46:09,886 epoch 6 - iter 290/292 - loss 0.08521640 - time (sec): 96.83 - samples/sec: 455.25 - lr: 0.000071 - momentum: 0.000000
2023-10-11 03:46:10,525 ----------------------------------------------------------------------------------------------------
2023-10-11 03:46:10,525 EPOCH 6 done: loss 0.0846 - lr: 0.000071
2023-10-11 03:46:16,096 DEV : loss 0.13130022585391998 - f1-score (micro avg) 0.7613
2023-10-11 03:46:16,105 saving best model
2023-10-11 03:46:18,785 ----------------------------------------------------------------------------------------------------
2023-10-11 03:46:27,822 epoch 7 - iter 29/292 - loss 0.07101005 - time (sec): 9.03 - samples/sec: 422.44 - lr: 0.000069 - momentum: 0.000000
2023-10-11 03:46:36,891 epoch 7 - iter 58/292 - loss 0.06386797 - time (sec): 18.10 - samples/sec: 437.68 - lr: 0.000068 - momentum: 0.000000
2023-10-11 03:46:45,457 epoch 7 - iter 87/292 - loss 0.06618248 - time (sec): 26.67 - samples/sec: 427.44 - lr: 0.000066 - momentum: 0.000000
2023-10-11 03:46:56,357 epoch 7 - iter 116/292 - loss 0.06714686 - time (sec): 37.57 - samples/sec: 453.34 - lr: 0.000064 - momentum: 0.000000
2023-10-11 03:47:06,080 epoch 7 - iter 145/292 - loss 0.06385049 - time (sec): 47.29 - samples/sec: 446.11 - lr: 0.000062 - momentum: 0.000000
2023-10-11 03:47:15,959 epoch 7 - iter 174/292 - loss 0.06762135 - time (sec): 57.17 - samples/sec: 442.09 - lr: 0.000061 - momentum: 0.000000
2023-10-11 03:47:25,666 epoch 7 - iter 203/292 - loss 0.06778033 - time (sec): 66.88 - samples/sec: 443.83 - lr: 0.000059 - momentum: 0.000000
2023-10-11 03:47:36,174 epoch 7 - iter 232/292 - loss 0.06382795 - time (sec): 77.38 - samples/sec: 445.91 - lr: 0.000057 - momentum: 0.000000
2023-10-11 03:47:46,819 epoch 7 - iter 261/292 - loss 0.06495314 - time (sec): 88.03 - samples/sec: 446.36 - lr: 0.000055 - momentum: 0.000000
2023-10-11 03:47:57,307 epoch 7 - iter 290/292 - loss 0.06399569 - time (sec): 98.52 - samples/sec: 449.10 - lr: 0.000054 - momentum: 0.000000
2023-10-11 03:47:57,797 ----------------------------------------------------------------------------------------------------
2023-10-11 03:47:57,798 EPOCH 7 done: loss 0.0640 - lr: 0.000054
2023-10-11 03:48:03,890 DEV : loss 0.11788583546876907 - f1-score (micro avg) 0.7922
2023-10-11 03:48:03,899 saving best model
2023-10-11 03:48:06,447 ----------------------------------------------------------------------------------------------------
2023-10-11 03:48:15,755 epoch 8 - iter 29/292 - loss 0.06045212 - time (sec): 9.30 - samples/sec: 440.49 - lr: 0.000052 - momentum: 0.000000
2023-10-11 03:48:25,290 epoch 8 - iter 58/292 - loss 0.05668848 - time (sec): 18.84 - samples/sec: 448.24 - lr: 0.000050 - momentum: 0.000000
2023-10-11 03:48:34,932 epoch 8 - iter 87/292 - loss 0.06064471 - time (sec): 28.48 - samples/sec: 448.98 - lr: 0.000048 - momentum: 0.000000
2023-10-11 03:48:45,261 epoch 8 - iter 116/292 - loss 0.06304311 - time (sec): 38.81 - samples/sec: 458.96 - lr: 0.000046 - momentum: 0.000000
2023-10-11 03:48:54,956 epoch 8 - iter 145/292 - loss 0.06183104 - time (sec): 48.50 - samples/sec: 456.93 - lr: 0.000045 - momentum: 0.000000
2023-10-11 03:49:03,980 epoch 8 - iter 174/292 - loss 0.05807061 - time (sec): 57.53 - samples/sec: 450.09 - lr: 0.000043 - momentum: 0.000000
2023-10-11 03:49:14,058 epoch 8 - iter 203/292 - loss 0.05673632 - time (sec): 67.61 - samples/sec: 454.78 - lr: 0.000041 - momentum: 0.000000
2023-10-11 03:49:24,114 epoch 8 - iter 232/292 - loss 0.05801900 - time (sec): 77.66 - samples/sec: 454.79 - lr: 0.000039 - momentum: 0.000000
2023-10-11 03:49:33,999 epoch 8 - iter 261/292 - loss 0.05350185 - time (sec): 87.55 - samples/sec: 447.75 - lr: 0.000038 - momentum: 0.000000
2023-10-11 03:49:44,408 epoch 8 - iter 290/292 - loss 0.05081514 - time (sec): 97.96 - samples/sec: 451.22 - lr: 0.000036 - momentum: 0.000000
2023-10-11 03:49:44,954 ----------------------------------------------------------------------------------------------------
2023-10-11 03:49:44,955 EPOCH 8 done: loss 0.0507 - lr: 0.000036
2023-10-11 03:49:50,893 DEV : loss 0.1216011717915535 - f1-score (micro avg) 0.7983
2023-10-11 03:49:50,903 saving best model
2023-10-11 03:49:53,505 ----------------------------------------------------------------------------------------------------
2023-10-11 03:50:03,543 epoch 9 - iter 29/292 - loss 0.03943985 - time (sec): 10.03 - samples/sec: 458.19 - lr: 0.000034 - momentum: 0.000000
2023-10-11 03:50:12,469 epoch 9 - iter 58/292 - loss 0.04844555 - time (sec): 18.96 - samples/sec: 434.67 - lr: 0.000032 - momentum: 0.000000
2023-10-11 03:50:21,870 epoch 9 - iter 87/292 - loss 0.04471746 - time (sec): 28.36 - samples/sec: 437.51 - lr: 0.000030 - momentum: 0.000000
2023-10-11 03:50:31,201 epoch 9 - iter 116/292 - loss 0.04485485 - time (sec): 37.69 - samples/sec: 436.29 - lr: 0.000029 - momentum: 0.000000
2023-10-11 03:50:41,656 epoch 9 - iter 145/292 - loss 0.04694636 - time (sec): 48.15 - samples/sec: 446.77 - lr: 0.000027 - momentum: 0.000000
2023-10-11 03:50:51,729 epoch 9 - iter 174/292 - loss 0.04399638 - time (sec): 58.22 - samples/sec: 449.01 - lr: 0.000025 - momentum: 0.000000
2023-10-11 03:51:02,209 epoch 9 - iter 203/292 - loss 0.04168589 - time (sec): 68.70 - samples/sec: 447.28 - lr: 0.000023 - momentum: 0.000000
2023-10-11 03:51:12,835 epoch 9 - iter 232/292 - loss 0.04315952 - time (sec): 79.33 - samples/sec: 451.79 - lr: 0.000022 - momentum: 0.000000
2023-10-11 03:51:22,130 epoch 9 - iter 261/292 - loss 0.04201939 - time (sec): 88.62 - samples/sec: 451.08 - lr: 0.000020 - momentum: 0.000000
2023-10-11 03:51:31,943 epoch 9 - iter 290/292 - loss 0.04206582 - time (sec): 98.43 - samples/sec: 449.63 - lr: 0.000018 - momentum: 0.000000
2023-10-11 03:51:32,417 ----------------------------------------------------------------------------------------------------
2023-10-11 03:51:32,417 EPOCH 9 done: loss 0.0420 - lr: 0.000018
2023-10-11 03:51:38,147 DEV : loss 0.12305182963609695 - f1-score (micro avg) 0.794
2023-10-11 03:51:38,156 ----------------------------------------------------------------------------------------------------
2023-10-11 03:51:47,466 epoch 10 - iter 29/292 - loss 0.03708349 - time (sec): 9.31 - samples/sec: 452.73 - lr: 0.000016 - momentum: 0.000000
2023-10-11 03:51:56,648 epoch 10 - iter 58/292 - loss 0.04525890 - time (sec): 18.49 - samples/sec: 451.53 - lr: 0.000014 - momentum: 0.000000
2023-10-11 03:52:06,950 epoch 10 - iter 87/292 - loss 0.03777979 - time (sec): 28.79 - samples/sec: 458.84 - lr: 0.000013 - momentum: 0.000000
2023-10-11 03:52:16,910 epoch 10 - iter 116/292 - loss 0.03392100 - time (sec): 38.75 - samples/sec: 455.20 - lr: 0.000011 - momentum: 0.000000
2023-10-11 03:52:26,359 epoch 10 - iter 145/292 - loss 0.03634942 - time (sec): 48.20 - samples/sec: 457.11 - lr: 0.000009 - momentum: 0.000000
2023-10-11 03:52:35,578 epoch 10 - iter 174/292 - loss 0.03671887 - time (sec): 57.42 - samples/sec: 449.89 - lr: 0.000007 - momentum: 0.000000
2023-10-11 03:52:45,924 epoch 10 - iter 203/292 - loss 0.03799343 - time (sec): 67.77 - samples/sec: 457.28 - lr: 0.000006 - momentum: 0.000000
2023-10-11 03:52:55,496 epoch 10 - iter 232/292 - loss 0.03741065 - time (sec): 77.34 - samples/sec: 455.74 - lr: 0.000004 - momentum: 0.000000
2023-10-11 03:53:05,434 epoch 10 - iter 261/292 - loss 0.03813911 - time (sec): 87.28 - samples/sec: 456.28 - lr: 0.000002 - momentum: 0.000000
2023-10-11 03:53:15,261 epoch 10 - iter 290/292 - loss 0.03854453 - time (sec): 97.10 - samples/sec: 453.32 - lr: 0.000000 - momentum: 0.000000
2023-10-11 03:53:15,951 ----------------------------------------------------------------------------------------------------
2023-10-11 03:53:15,951 EPOCH 10 done: loss 0.0383 - lr: 0.000000
2023-10-11 03:53:21,696 DEV : loss 0.12359649688005447 - f1-score (micro avg) 0.7863
2023-10-11 03:53:22,642 ----------------------------------------------------------------------------------------------------
2023-10-11 03:53:22,644 Loading model from best epoch ...
2023-10-11 03:53:27,154 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 03:53:40,165
Results:
- F-score (micro) 0.7459
- F-score (macro) 0.7006
- Accuracy 0.6127
By class:
precision recall f1-score support
PER 0.8209 0.8563 0.8383 348
LOC 0.5972 0.8123 0.6883 261
ORG 0.4000 0.3846 0.3922 52
HumanProd 0.9048 0.8636 0.8837 22
micro avg 0.6958 0.8038 0.7459 683
macro avg 0.6807 0.7292 0.7006 683
weighted avg 0.7061 0.8038 0.7485 683
2023-10-11 03:53:40,165 ----------------------------------------------------------------------------------------------------
|