File size: 25,296 Bytes
9accb3b |
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 |
2023-10-10 22:38:50,022 ----------------------------------------------------------------------------------------------------
2023-10-10 22:38:50,024 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 22:38:50,025 ----------------------------------------------------------------------------------------------------
2023-10-10 22:38:50,025 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 22:38:50,025 ----------------------------------------------------------------------------------------------------
2023-10-10 22:38:50,025 Train: 1166 sentences
2023-10-10 22:38:50,025 (train_with_dev=False, train_with_test=False)
2023-10-10 22:38:50,025 ----------------------------------------------------------------------------------------------------
2023-10-10 22:38:50,025 Training Params:
2023-10-10 22:38:50,025 - learning_rate: "0.00016"
2023-10-10 22:38:50,025 - mini_batch_size: "4"
2023-10-10 22:38:50,025 - max_epochs: "10"
2023-10-10 22:38:50,025 - shuffle: "True"
2023-10-10 22:38:50,025 ----------------------------------------------------------------------------------------------------
2023-10-10 22:38:50,025 Plugins:
2023-10-10 22:38:50,026 - TensorboardLogger
2023-10-10 22:38:50,026 - LinearScheduler | warmup_fraction: '0.1'
2023-10-10 22:38:50,026 ----------------------------------------------------------------------------------------------------
2023-10-10 22:38:50,026 Final evaluation on model from best epoch (best-model.pt)
2023-10-10 22:38:50,026 - metric: "('micro avg', 'f1-score')"
2023-10-10 22:38:50,026 ----------------------------------------------------------------------------------------------------
2023-10-10 22:38:50,026 Computation:
2023-10-10 22:38:50,026 - compute on device: cuda:0
2023-10-10 22:38:50,026 - embedding storage: none
2023-10-10 22:38:50,026 ----------------------------------------------------------------------------------------------------
2023-10-10 22:38:50,026 Model training base path: "hmbench-newseye/fi-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs4-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-1"
2023-10-10 22:38:50,026 ----------------------------------------------------------------------------------------------------
2023-10-10 22:38:50,026 ----------------------------------------------------------------------------------------------------
2023-10-10 22:38:50,026 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-10 22:38:59,893 epoch 1 - iter 29/292 - loss 2.82904011 - time (sec): 9.86 - samples/sec: 517.82 - lr: 0.000015 - momentum: 0.000000
2023-10-10 22:39:08,951 epoch 1 - iter 58/292 - loss 2.82009083 - time (sec): 18.92 - samples/sec: 482.18 - lr: 0.000031 - momentum: 0.000000
2023-10-10 22:39:19,259 epoch 1 - iter 87/292 - loss 2.79608502 - time (sec): 29.23 - samples/sec: 501.01 - lr: 0.000047 - momentum: 0.000000
2023-10-10 22:39:28,698 epoch 1 - iter 116/292 - loss 2.75939629 - time (sec): 38.67 - samples/sec: 482.24 - lr: 0.000063 - momentum: 0.000000
2023-10-10 22:39:38,409 epoch 1 - iter 145/292 - loss 2.68289368 - time (sec): 48.38 - samples/sec: 465.91 - lr: 0.000079 - momentum: 0.000000
2023-10-10 22:39:47,654 epoch 1 - iter 174/292 - loss 2.58924155 - time (sec): 57.63 - samples/sec: 455.94 - lr: 0.000095 - momentum: 0.000000
2023-10-10 22:39:57,761 epoch 1 - iter 203/292 - loss 2.45043910 - time (sec): 67.73 - samples/sec: 458.95 - lr: 0.000111 - momentum: 0.000000
2023-10-10 22:40:08,012 epoch 1 - iter 232/292 - loss 2.32335868 - time (sec): 77.98 - samples/sec: 458.42 - lr: 0.000127 - momentum: 0.000000
2023-10-10 22:40:17,693 epoch 1 - iter 261/292 - loss 2.20131381 - time (sec): 87.66 - samples/sec: 455.90 - lr: 0.000142 - momentum: 0.000000
2023-10-10 22:40:27,577 epoch 1 - iter 290/292 - loss 2.08347326 - time (sec): 97.55 - samples/sec: 453.75 - lr: 0.000158 - momentum: 0.000000
2023-10-10 22:40:28,055 ----------------------------------------------------------------------------------------------------
2023-10-10 22:40:28,055 EPOCH 1 done: loss 2.0787 - lr: 0.000158
2023-10-10 22:40:33,670 DEV : loss 0.6691536903381348 - f1-score (micro avg) 0.0
2023-10-10 22:40:33,679 ----------------------------------------------------------------------------------------------------
2023-10-10 22:40:43,569 epoch 2 - iter 29/292 - loss 0.73979818 - time (sec): 9.89 - samples/sec: 473.00 - lr: 0.000158 - momentum: 0.000000
2023-10-10 22:40:53,455 epoch 2 - iter 58/292 - loss 0.79010563 - time (sec): 19.77 - samples/sec: 464.71 - lr: 0.000157 - momentum: 0.000000
2023-10-10 22:41:03,434 epoch 2 - iter 87/292 - loss 0.68326802 - time (sec): 29.75 - samples/sec: 460.46 - lr: 0.000155 - momentum: 0.000000
2023-10-10 22:41:13,798 epoch 2 - iter 116/292 - loss 0.60934981 - time (sec): 40.12 - samples/sec: 460.43 - lr: 0.000153 - momentum: 0.000000
2023-10-10 22:41:24,470 epoch 2 - iter 145/292 - loss 0.58125380 - time (sec): 50.79 - samples/sec: 463.84 - lr: 0.000151 - momentum: 0.000000
2023-10-10 22:41:34,361 epoch 2 - iter 174/292 - loss 0.54661045 - time (sec): 60.68 - samples/sec: 456.43 - lr: 0.000149 - momentum: 0.000000
2023-10-10 22:41:44,174 epoch 2 - iter 203/292 - loss 0.53389626 - time (sec): 70.49 - samples/sec: 447.60 - lr: 0.000148 - momentum: 0.000000
2023-10-10 22:41:54,010 epoch 2 - iter 232/292 - loss 0.51870783 - time (sec): 80.33 - samples/sec: 443.48 - lr: 0.000146 - momentum: 0.000000
2023-10-10 22:42:03,637 epoch 2 - iter 261/292 - loss 0.49765927 - time (sec): 89.96 - samples/sec: 445.87 - lr: 0.000144 - momentum: 0.000000
2023-10-10 22:42:13,850 epoch 2 - iter 290/292 - loss 0.49732765 - time (sec): 100.17 - samples/sec: 442.98 - lr: 0.000142 - momentum: 0.000000
2023-10-10 22:42:14,221 ----------------------------------------------------------------------------------------------------
2023-10-10 22:42:14,222 EPOCH 2 done: loss 0.4968 - lr: 0.000142
2023-10-10 22:42:20,227 DEV : loss 0.28831687569618225 - f1-score (micro avg) 0.1468
2023-10-10 22:42:20,237 saving best model
2023-10-10 22:42:21,190 ----------------------------------------------------------------------------------------------------
2023-10-10 22:42:30,312 epoch 3 - iter 29/292 - loss 0.37250554 - time (sec): 9.12 - samples/sec: 407.06 - lr: 0.000141 - momentum: 0.000000
2023-10-10 22:42:41,368 epoch 3 - iter 58/292 - loss 0.28788479 - time (sec): 20.18 - samples/sec: 440.44 - lr: 0.000139 - momentum: 0.000000
2023-10-10 22:42:50,894 epoch 3 - iter 87/292 - loss 0.31377648 - time (sec): 29.70 - samples/sec: 440.72 - lr: 0.000137 - momentum: 0.000000
2023-10-10 22:43:00,160 epoch 3 - iter 116/292 - loss 0.30860177 - time (sec): 38.97 - samples/sec: 439.06 - lr: 0.000135 - momentum: 0.000000
2023-10-10 22:43:09,738 epoch 3 - iter 145/292 - loss 0.30045066 - time (sec): 48.54 - samples/sec: 444.89 - lr: 0.000133 - momentum: 0.000000
2023-10-10 22:43:18,762 epoch 3 - iter 174/292 - loss 0.30780945 - time (sec): 57.57 - samples/sec: 444.68 - lr: 0.000132 - momentum: 0.000000
2023-10-10 22:43:29,441 epoch 3 - iter 203/292 - loss 0.31651504 - time (sec): 68.25 - samples/sec: 455.95 - lr: 0.000130 - momentum: 0.000000
2023-10-10 22:43:39,300 epoch 3 - iter 232/292 - loss 0.31485026 - time (sec): 78.11 - samples/sec: 456.80 - lr: 0.000128 - momentum: 0.000000
2023-10-10 22:43:49,319 epoch 3 - iter 261/292 - loss 0.30698124 - time (sec): 88.13 - samples/sec: 459.46 - lr: 0.000126 - momentum: 0.000000
2023-10-10 22:43:58,539 epoch 3 - iter 290/292 - loss 0.30215031 - time (sec): 97.35 - samples/sec: 455.07 - lr: 0.000125 - momentum: 0.000000
2023-10-10 22:43:58,987 ----------------------------------------------------------------------------------------------------
2023-10-10 22:43:58,988 EPOCH 3 done: loss 0.3015 - lr: 0.000125
2023-10-10 22:44:05,062 DEV : loss 0.21253274381160736 - f1-score (micro avg) 0.4454
2023-10-10 22:44:05,072 saving best model
2023-10-10 22:44:14,113 ----------------------------------------------------------------------------------------------------
2023-10-10 22:44:24,977 epoch 4 - iter 29/292 - loss 0.20247520 - time (sec): 10.86 - samples/sec: 460.60 - lr: 0.000123 - momentum: 0.000000
2023-10-10 22:44:34,949 epoch 4 - iter 58/292 - loss 0.18905862 - time (sec): 20.83 - samples/sec: 437.66 - lr: 0.000121 - momentum: 0.000000
2023-10-10 22:44:45,648 epoch 4 - iter 87/292 - loss 0.22813786 - time (sec): 31.53 - samples/sec: 444.87 - lr: 0.000119 - momentum: 0.000000
2023-10-10 22:44:55,775 epoch 4 - iter 116/292 - loss 0.23168511 - time (sec): 41.66 - samples/sec: 441.17 - lr: 0.000117 - momentum: 0.000000
2023-10-10 22:45:06,379 epoch 4 - iter 145/292 - loss 0.22417524 - time (sec): 52.26 - samples/sec: 437.63 - lr: 0.000116 - momentum: 0.000000
2023-10-10 22:45:16,848 epoch 4 - iter 174/292 - loss 0.22225430 - time (sec): 62.73 - samples/sec: 431.91 - lr: 0.000114 - momentum: 0.000000
2023-10-10 22:45:27,869 epoch 4 - iter 203/292 - loss 0.21451680 - time (sec): 73.75 - samples/sec: 430.08 - lr: 0.000112 - momentum: 0.000000
2023-10-10 22:45:37,709 epoch 4 - iter 232/292 - loss 0.21129060 - time (sec): 83.59 - samples/sec: 431.92 - lr: 0.000110 - momentum: 0.000000
2023-10-10 22:45:47,385 epoch 4 - iter 261/292 - loss 0.20883694 - time (sec): 93.27 - samples/sec: 425.79 - lr: 0.000109 - momentum: 0.000000
2023-10-10 22:45:57,486 epoch 4 - iter 290/292 - loss 0.20421114 - time (sec): 103.37 - samples/sec: 427.91 - lr: 0.000107 - momentum: 0.000000
2023-10-10 22:45:57,953 ----------------------------------------------------------------------------------------------------
2023-10-10 22:45:57,954 EPOCH 4 done: loss 0.2034 - lr: 0.000107
2023-10-10 22:46:03,949 DEV : loss 0.16762110590934753 - f1-score (micro avg) 0.636
2023-10-10 22:46:03,958 saving best model
2023-10-10 22:46:12,656 ----------------------------------------------------------------------------------------------------
2023-10-10 22:46:22,429 epoch 5 - iter 29/292 - loss 0.15228624 - time (sec): 9.77 - samples/sec: 444.71 - lr: 0.000105 - momentum: 0.000000
2023-10-10 22:46:33,178 epoch 5 - iter 58/292 - loss 0.16824065 - time (sec): 20.52 - samples/sec: 456.34 - lr: 0.000103 - momentum: 0.000000
2023-10-10 22:46:42,550 epoch 5 - iter 87/292 - loss 0.16516610 - time (sec): 29.89 - samples/sec: 454.50 - lr: 0.000101 - momentum: 0.000000
2023-10-10 22:46:52,308 epoch 5 - iter 116/292 - loss 0.14358303 - time (sec): 39.65 - samples/sec: 451.98 - lr: 0.000100 - momentum: 0.000000
2023-10-10 22:47:02,466 epoch 5 - iter 145/292 - loss 0.14189975 - time (sec): 49.81 - samples/sec: 456.77 - lr: 0.000098 - momentum: 0.000000
2023-10-10 22:47:11,868 epoch 5 - iter 174/292 - loss 0.13815073 - time (sec): 59.21 - samples/sec: 447.30 - lr: 0.000096 - momentum: 0.000000
2023-10-10 22:47:22,402 epoch 5 - iter 203/292 - loss 0.13687599 - time (sec): 69.74 - samples/sec: 450.72 - lr: 0.000094 - momentum: 0.000000
2023-10-10 22:47:32,919 epoch 5 - iter 232/292 - loss 0.13663132 - time (sec): 80.26 - samples/sec: 448.91 - lr: 0.000093 - momentum: 0.000000
2023-10-10 22:47:42,044 epoch 5 - iter 261/292 - loss 0.13323399 - time (sec): 89.38 - samples/sec: 444.34 - lr: 0.000091 - momentum: 0.000000
2023-10-10 22:47:52,280 epoch 5 - iter 290/292 - loss 0.13201349 - time (sec): 99.62 - samples/sec: 445.37 - lr: 0.000089 - momentum: 0.000000
2023-10-10 22:47:52,695 ----------------------------------------------------------------------------------------------------
2023-10-10 22:47:52,696 EPOCH 5 done: loss 0.1319 - lr: 0.000089
2023-10-10 22:47:58,790 DEV : loss 0.1449754387140274 - f1-score (micro avg) 0.7834
2023-10-10 22:47:58,800 saving best model
2023-10-10 22:48:08,053 ----------------------------------------------------------------------------------------------------
2023-10-10 22:48:18,620 epoch 6 - iter 29/292 - loss 0.09780191 - time (sec): 10.56 - samples/sec: 508.71 - lr: 0.000087 - momentum: 0.000000
2023-10-10 22:48:27,513 epoch 6 - iter 58/292 - loss 0.10813143 - time (sec): 19.46 - samples/sec: 468.11 - lr: 0.000085 - momentum: 0.000000
2023-10-10 22:48:38,005 epoch 6 - iter 87/292 - loss 0.09566991 - time (sec): 29.95 - samples/sec: 464.73 - lr: 0.000084 - momentum: 0.000000
2023-10-10 22:48:47,930 epoch 6 - iter 116/292 - loss 0.09010732 - time (sec): 39.87 - samples/sec: 463.92 - lr: 0.000082 - momentum: 0.000000
2023-10-10 22:48:57,497 epoch 6 - iter 145/292 - loss 0.09093065 - time (sec): 49.44 - samples/sec: 463.49 - lr: 0.000080 - momentum: 0.000000
2023-10-10 22:49:06,968 epoch 6 - iter 174/292 - loss 0.09436299 - time (sec): 58.91 - samples/sec: 460.52 - lr: 0.000078 - momentum: 0.000000
2023-10-10 22:49:16,957 epoch 6 - iter 203/292 - loss 0.09269008 - time (sec): 68.90 - samples/sec: 448.05 - lr: 0.000077 - momentum: 0.000000
2023-10-10 22:49:27,267 epoch 6 - iter 232/292 - loss 0.09292884 - time (sec): 79.21 - samples/sec: 441.90 - lr: 0.000075 - momentum: 0.000000
2023-10-10 22:49:38,103 epoch 6 - iter 261/292 - loss 0.09295591 - time (sec): 90.05 - samples/sec: 442.24 - lr: 0.000073 - momentum: 0.000000
2023-10-10 22:49:48,309 epoch 6 - iter 290/292 - loss 0.09196060 - time (sec): 100.25 - samples/sec: 440.11 - lr: 0.000071 - momentum: 0.000000
2023-10-10 22:49:48,896 ----------------------------------------------------------------------------------------------------
2023-10-10 22:49:48,897 EPOCH 6 done: loss 0.0923 - lr: 0.000071
2023-10-10 22:49:55,279 DEV : loss 0.1280345916748047 - f1-score (micro avg) 0.7843
2023-10-10 22:49:55,289 saving best model
2023-10-10 22:49:58,465 ----------------------------------------------------------------------------------------------------
2023-10-10 22:50:09,512 epoch 7 - iter 29/292 - loss 0.07453827 - time (sec): 11.04 - samples/sec: 443.98 - lr: 0.000069 - momentum: 0.000000
2023-10-10 22:50:19,617 epoch 7 - iter 58/292 - loss 0.06721636 - time (sec): 21.15 - samples/sec: 411.62 - lr: 0.000068 - momentum: 0.000000
2023-10-10 22:50:29,934 epoch 7 - iter 87/292 - loss 0.07117217 - time (sec): 31.46 - samples/sec: 417.23 - lr: 0.000066 - momentum: 0.000000
2023-10-10 22:50:40,481 epoch 7 - iter 116/292 - loss 0.06362150 - time (sec): 42.01 - samples/sec: 439.08 - lr: 0.000064 - momentum: 0.000000
2023-10-10 22:50:50,693 epoch 7 - iter 145/292 - loss 0.06235246 - time (sec): 52.22 - samples/sec: 430.49 - lr: 0.000062 - momentum: 0.000000
2023-10-10 22:51:00,958 epoch 7 - iter 174/292 - loss 0.06840778 - time (sec): 62.49 - samples/sec: 423.77 - lr: 0.000061 - momentum: 0.000000
2023-10-10 22:51:10,514 epoch 7 - iter 203/292 - loss 0.06625399 - time (sec): 72.05 - samples/sec: 423.90 - lr: 0.000059 - momentum: 0.000000
2023-10-10 22:51:21,312 epoch 7 - iter 232/292 - loss 0.06499604 - time (sec): 82.84 - samples/sec: 428.42 - lr: 0.000057 - momentum: 0.000000
2023-10-10 22:51:30,775 epoch 7 - iter 261/292 - loss 0.06666186 - time (sec): 92.31 - samples/sec: 431.34 - lr: 0.000055 - momentum: 0.000000
2023-10-10 22:51:40,482 epoch 7 - iter 290/292 - loss 0.06904362 - time (sec): 102.01 - samples/sec: 433.76 - lr: 0.000054 - momentum: 0.000000
2023-10-10 22:51:40,947 ----------------------------------------------------------------------------------------------------
2023-10-10 22:51:40,947 EPOCH 7 done: loss 0.0696 - lr: 0.000054
2023-10-10 22:51:46,921 DEV : loss 0.13766349852085114 - f1-score (micro avg) 0.757
2023-10-10 22:51:46,929 ----------------------------------------------------------------------------------------------------
2023-10-10 22:51:56,914 epoch 8 - iter 29/292 - loss 0.04980003 - time (sec): 9.98 - samples/sec: 506.44 - lr: 0.000052 - momentum: 0.000000
2023-10-10 22:52:06,746 epoch 8 - iter 58/292 - loss 0.05123584 - time (sec): 19.82 - samples/sec: 500.27 - lr: 0.000050 - momentum: 0.000000
2023-10-10 22:52:16,597 epoch 8 - iter 87/292 - loss 0.05568077 - time (sec): 29.67 - samples/sec: 485.27 - lr: 0.000048 - momentum: 0.000000
2023-10-10 22:52:25,412 epoch 8 - iter 116/292 - loss 0.05299151 - time (sec): 38.48 - samples/sec: 468.70 - lr: 0.000046 - momentum: 0.000000
2023-10-10 22:52:35,578 epoch 8 - iter 145/292 - loss 0.05256239 - time (sec): 48.65 - samples/sec: 468.99 - lr: 0.000045 - momentum: 0.000000
2023-10-10 22:52:44,728 epoch 8 - iter 174/292 - loss 0.05533329 - time (sec): 57.80 - samples/sec: 467.26 - lr: 0.000043 - momentum: 0.000000
2023-10-10 22:52:53,698 epoch 8 - iter 203/292 - loss 0.05536529 - time (sec): 66.77 - samples/sec: 463.07 - lr: 0.000041 - momentum: 0.000000
2023-10-10 22:53:03,670 epoch 8 - iter 232/292 - loss 0.05502038 - time (sec): 76.74 - samples/sec: 465.90 - lr: 0.000039 - momentum: 0.000000
2023-10-10 22:53:13,047 epoch 8 - iter 261/292 - loss 0.05399818 - time (sec): 86.12 - samples/sec: 461.30 - lr: 0.000038 - momentum: 0.000000
2023-10-10 22:53:23,422 epoch 8 - iter 290/292 - loss 0.05648797 - time (sec): 96.49 - samples/sec: 458.97 - lr: 0.000036 - momentum: 0.000000
2023-10-10 22:53:23,902 ----------------------------------------------------------------------------------------------------
2023-10-10 22:53:23,903 EPOCH 8 done: loss 0.0564 - lr: 0.000036
2023-10-10 22:53:29,955 DEV : loss 0.12616726756095886 - f1-score (micro avg) 0.7716
2023-10-10 22:53:29,964 ----------------------------------------------------------------------------------------------------
2023-10-10 22:53:39,918 epoch 9 - iter 29/292 - loss 0.04779883 - time (sec): 9.95 - samples/sec: 468.37 - lr: 0.000034 - momentum: 0.000000
2023-10-10 22:53:49,660 epoch 9 - iter 58/292 - loss 0.04408728 - time (sec): 19.69 - samples/sec: 475.59 - lr: 0.000032 - momentum: 0.000000
2023-10-10 22:53:59,507 epoch 9 - iter 87/292 - loss 0.04629535 - time (sec): 29.54 - samples/sec: 467.97 - lr: 0.000030 - momentum: 0.000000
2023-10-10 22:54:10,142 epoch 9 - iter 116/292 - loss 0.04202133 - time (sec): 40.18 - samples/sec: 457.87 - lr: 0.000029 - momentum: 0.000000
2023-10-10 22:54:19,541 epoch 9 - iter 145/292 - loss 0.04383999 - time (sec): 49.57 - samples/sec: 451.70 - lr: 0.000027 - momentum: 0.000000
2023-10-10 22:54:30,079 epoch 9 - iter 174/292 - loss 0.04672478 - time (sec): 60.11 - samples/sec: 452.65 - lr: 0.000025 - momentum: 0.000000
2023-10-10 22:54:39,735 epoch 9 - iter 203/292 - loss 0.04641879 - time (sec): 69.77 - samples/sec: 444.73 - lr: 0.000023 - momentum: 0.000000
2023-10-10 22:54:50,153 epoch 9 - iter 232/292 - loss 0.04595521 - time (sec): 80.19 - samples/sec: 449.04 - lr: 0.000022 - momentum: 0.000000
2023-10-10 22:55:00,375 epoch 9 - iter 261/292 - loss 0.04464917 - time (sec): 90.41 - samples/sec: 445.85 - lr: 0.000020 - momentum: 0.000000
2023-10-10 22:55:09,855 epoch 9 - iter 290/292 - loss 0.04722935 - time (sec): 99.89 - samples/sec: 442.99 - lr: 0.000018 - momentum: 0.000000
2023-10-10 22:55:10,343 ----------------------------------------------------------------------------------------------------
2023-10-10 22:55:10,343 EPOCH 9 done: loss 0.0472 - lr: 0.000018
2023-10-10 22:55:16,251 DEV : loss 0.1275288611650467 - f1-score (micro avg) 0.7846
2023-10-10 22:55:16,268 saving best model
2023-10-10 22:55:21,311 ----------------------------------------------------------------------------------------------------
2023-10-10 22:55:31,028 epoch 10 - iter 29/292 - loss 0.03389118 - time (sec): 9.71 - samples/sec: 475.91 - lr: 0.000016 - momentum: 0.000000
2023-10-10 22:55:40,275 epoch 10 - iter 58/292 - loss 0.04001796 - time (sec): 18.96 - samples/sec: 451.56 - lr: 0.000014 - momentum: 0.000000
2023-10-10 22:55:49,745 epoch 10 - iter 87/292 - loss 0.03855770 - time (sec): 28.43 - samples/sec: 451.16 - lr: 0.000013 - momentum: 0.000000
2023-10-10 22:56:00,674 epoch 10 - iter 116/292 - loss 0.03345664 - time (sec): 39.36 - samples/sec: 461.08 - lr: 0.000011 - momentum: 0.000000
2023-10-10 22:56:10,848 epoch 10 - iter 145/292 - loss 0.03410413 - time (sec): 49.53 - samples/sec: 463.14 - lr: 0.000009 - momentum: 0.000000
2023-10-10 22:56:21,318 epoch 10 - iter 174/292 - loss 0.03396462 - time (sec): 60.00 - samples/sec: 455.00 - lr: 0.000007 - momentum: 0.000000
2023-10-10 22:56:32,199 epoch 10 - iter 203/292 - loss 0.03618776 - time (sec): 70.88 - samples/sec: 451.04 - lr: 0.000006 - momentum: 0.000000
2023-10-10 22:56:42,084 epoch 10 - iter 232/292 - loss 0.04153492 - time (sec): 80.77 - samples/sec: 445.45 - lr: 0.000004 - momentum: 0.000000
2023-10-10 22:56:51,564 epoch 10 - iter 261/292 - loss 0.04016965 - time (sec): 90.25 - samples/sec: 442.69 - lr: 0.000002 - momentum: 0.000000
2023-10-10 22:57:01,710 epoch 10 - iter 290/292 - loss 0.04323998 - time (sec): 100.40 - samples/sec: 440.51 - lr: 0.000000 - momentum: 0.000000
2023-10-10 22:57:02,248 ----------------------------------------------------------------------------------------------------
2023-10-10 22:57:02,249 EPOCH 10 done: loss 0.0434 - lr: 0.000000
2023-10-10 22:57:08,238 DEV : loss 0.1296752691268921 - f1-score (micro avg) 0.8017
2023-10-10 22:57:08,247 saving best model
2023-10-10 22:57:13,206 ----------------------------------------------------------------------------------------------------
2023-10-10 22:57:13,209 Loading model from best epoch ...
2023-10-10 22:57:17,160 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
2023-10-10 22:57:30,053
Results:
- F-score (micro) 0.7252
- F-score (macro) 0.6708
- Accuracy 0.587
By class:
precision recall f1-score support
PER 0.7920 0.8534 0.8216 348
LOC 0.5819 0.7625 0.6600 261
ORG 0.4000 0.3846 0.3922 52
HumanProd 0.8500 0.7727 0.8095 22
micro avg 0.6773 0.7804 0.7252 683
macro avg 0.6560 0.6933 0.6708 683
weighted avg 0.6837 0.7804 0.7268 683
2023-10-10 22:57:30,054 ----------------------------------------------------------------------------------------------------
|