File size: 25,252 Bytes
86697a4 |
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 21:41:37,288 ----------------------------------------------------------------------------------------------------
2023-10-10 21:41:37,290 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 21:41:37,290 ----------------------------------------------------------------------------------------------------
2023-10-10 21:41:37,291 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 21:41:37,291 ----------------------------------------------------------------------------------------------------
2023-10-10 21:41:37,291 Train: 1166 sentences
2023-10-10 21:41:37,291 (train_with_dev=False, train_with_test=False)
2023-10-10 21:41:37,291 ----------------------------------------------------------------------------------------------------
2023-10-10 21:41:37,291 Training Params:
2023-10-10 21:41:37,291 - learning_rate: "0.00015"
2023-10-10 21:41:37,291 - mini_batch_size: "8"
2023-10-10 21:41:37,291 - max_epochs: "10"
2023-10-10 21:41:37,291 - shuffle: "True"
2023-10-10 21:41:37,291 ----------------------------------------------------------------------------------------------------
2023-10-10 21:41:37,291 Plugins:
2023-10-10 21:41:37,291 - TensorboardLogger
2023-10-10 21:41:37,292 - LinearScheduler | warmup_fraction: '0.1'
2023-10-10 21:41:37,292 ----------------------------------------------------------------------------------------------------
2023-10-10 21:41:37,292 Final evaluation on model from best epoch (best-model.pt)
2023-10-10 21:41:37,292 - metric: "('micro avg', 'f1-score')"
2023-10-10 21:41:37,292 ----------------------------------------------------------------------------------------------------
2023-10-10 21:41:37,292 Computation:
2023-10-10 21:41:37,292 - compute on device: cuda:0
2023-10-10 21:41:37,292 - embedding storage: none
2023-10-10 21:41:37,292 ----------------------------------------------------------------------------------------------------
2023-10-10 21:41:37,292 Model training base path: "hmbench-newseye/fi-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs8-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-1"
2023-10-10 21:41:37,292 ----------------------------------------------------------------------------------------------------
2023-10-10 21:41:37,292 ----------------------------------------------------------------------------------------------------
2023-10-10 21:41:37,292 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-10 21:41:47,892 epoch 1 - iter 14/146 - loss 2.83099006 - time (sec): 10.60 - samples/sec: 454.83 - lr: 0.000013 - momentum: 0.000000
2023-10-10 21:41:56,307 epoch 1 - iter 28/146 - loss 2.82672987 - time (sec): 19.01 - samples/sec: 453.39 - lr: 0.000028 - momentum: 0.000000
2023-10-10 21:42:05,412 epoch 1 - iter 42/146 - loss 2.81735922 - time (sec): 28.12 - samples/sec: 487.02 - lr: 0.000042 - momentum: 0.000000
2023-10-10 21:42:13,913 epoch 1 - iter 56/146 - loss 2.80389264 - time (sec): 36.62 - samples/sec: 492.89 - lr: 0.000057 - momentum: 0.000000
2023-10-10 21:42:21,745 epoch 1 - iter 70/146 - loss 2.78350842 - time (sec): 44.45 - samples/sec: 488.75 - lr: 0.000071 - momentum: 0.000000
2023-10-10 21:42:30,315 epoch 1 - iter 84/146 - loss 2.74176859 - time (sec): 53.02 - samples/sec: 484.59 - lr: 0.000085 - momentum: 0.000000
2023-10-10 21:42:38,433 epoch 1 - iter 98/146 - loss 2.68567253 - time (sec): 61.14 - samples/sec: 478.65 - lr: 0.000100 - momentum: 0.000000
2023-10-10 21:42:48,723 epoch 1 - iter 112/146 - loss 2.58862607 - time (sec): 71.43 - samples/sec: 486.62 - lr: 0.000114 - momentum: 0.000000
2023-10-10 21:42:57,392 epoch 1 - iter 126/146 - loss 2.51465264 - time (sec): 80.10 - samples/sec: 484.34 - lr: 0.000128 - momentum: 0.000000
2023-10-10 21:43:07,025 epoch 1 - iter 140/146 - loss 2.43562024 - time (sec): 89.73 - samples/sec: 478.39 - lr: 0.000143 - momentum: 0.000000
2023-10-10 21:43:10,647 ----------------------------------------------------------------------------------------------------
2023-10-10 21:43:10,648 EPOCH 1 done: loss 2.4069 - lr: 0.000143
2023-10-10 21:43:16,880 DEV : loss 1.356826901435852 - f1-score (micro avg) 0.0
2023-10-10 21:43:16,889 ----------------------------------------------------------------------------------------------------
2023-10-10 21:43:26,905 epoch 2 - iter 14/146 - loss 1.39370672 - time (sec): 10.01 - samples/sec: 457.36 - lr: 0.000149 - momentum: 0.000000
2023-10-10 21:43:36,699 epoch 2 - iter 28/146 - loss 1.35742229 - time (sec): 19.81 - samples/sec: 449.72 - lr: 0.000147 - momentum: 0.000000
2023-10-10 21:43:46,625 epoch 2 - iter 42/146 - loss 1.22339354 - time (sec): 29.73 - samples/sec: 441.39 - lr: 0.000145 - momentum: 0.000000
2023-10-10 21:43:56,477 epoch 2 - iter 56/146 - loss 1.12844539 - time (sec): 39.59 - samples/sec: 446.10 - lr: 0.000144 - momentum: 0.000000
2023-10-10 21:44:07,430 epoch 2 - iter 70/146 - loss 1.05603913 - time (sec): 50.54 - samples/sec: 453.57 - lr: 0.000142 - momentum: 0.000000
2023-10-10 21:44:16,779 epoch 2 - iter 84/146 - loss 0.99801684 - time (sec): 59.89 - samples/sec: 448.64 - lr: 0.000141 - momentum: 0.000000
2023-10-10 21:44:26,181 epoch 2 - iter 98/146 - loss 0.95042167 - time (sec): 69.29 - samples/sec: 443.73 - lr: 0.000139 - momentum: 0.000000
2023-10-10 21:44:35,483 epoch 2 - iter 112/146 - loss 0.91179574 - time (sec): 78.59 - samples/sec: 439.91 - lr: 0.000137 - momentum: 0.000000
2023-10-10 21:44:44,970 epoch 2 - iter 126/146 - loss 0.87949814 - time (sec): 88.08 - samples/sec: 440.06 - lr: 0.000136 - momentum: 0.000000
2023-10-10 21:44:54,351 epoch 2 - iter 140/146 - loss 0.84964841 - time (sec): 97.46 - samples/sec: 440.66 - lr: 0.000134 - momentum: 0.000000
2023-10-10 21:44:58,079 ----------------------------------------------------------------------------------------------------
2023-10-10 21:44:58,079 EPOCH 2 done: loss 0.8457 - lr: 0.000134
2023-10-10 21:45:04,659 DEV : loss 0.4102368950843811 - f1-score (micro avg) 0.0
2023-10-10 21:45:04,669 ----------------------------------------------------------------------------------------------------
2023-10-10 21:45:13,948 epoch 3 - iter 14/146 - loss 0.53677705 - time (sec): 9.28 - samples/sec: 380.64 - lr: 0.000132 - momentum: 0.000000
2023-10-10 21:45:23,837 epoch 3 - iter 28/146 - loss 0.43791549 - time (sec): 19.17 - samples/sec: 446.68 - lr: 0.000130 - momentum: 0.000000
2023-10-10 21:45:32,873 epoch 3 - iter 42/146 - loss 0.45683813 - time (sec): 28.20 - samples/sec: 453.06 - lr: 0.000129 - momentum: 0.000000
2023-10-10 21:45:41,710 epoch 3 - iter 56/146 - loss 0.44634295 - time (sec): 37.04 - samples/sec: 451.44 - lr: 0.000127 - momentum: 0.000000
2023-10-10 21:45:50,849 epoch 3 - iter 70/146 - loss 0.43680565 - time (sec): 46.18 - samples/sec: 454.80 - lr: 0.000126 - momentum: 0.000000
2023-10-10 21:45:59,653 epoch 3 - iter 84/146 - loss 0.43710977 - time (sec): 54.98 - samples/sec: 447.84 - lr: 0.000124 - momentum: 0.000000
2023-10-10 21:46:09,907 epoch 3 - iter 98/146 - loss 0.46074815 - time (sec): 65.24 - samples/sec: 457.86 - lr: 0.000122 - momentum: 0.000000
2023-10-10 21:46:19,746 epoch 3 - iter 112/146 - loss 0.44530575 - time (sec): 75.08 - samples/sec: 462.35 - lr: 0.000121 - momentum: 0.000000
2023-10-10 21:46:29,509 epoch 3 - iter 126/146 - loss 0.43292114 - time (sec): 84.84 - samples/sec: 461.28 - lr: 0.000119 - momentum: 0.000000
2023-10-10 21:46:37,992 epoch 3 - iter 140/146 - loss 0.43074177 - time (sec): 93.32 - samples/sec: 456.80 - lr: 0.000118 - momentum: 0.000000
2023-10-10 21:46:41,672 ----------------------------------------------------------------------------------------------------
2023-10-10 21:46:41,672 EPOCH 3 done: loss 0.4255 - lr: 0.000118
2023-10-10 21:46:47,458 DEV : loss 0.29022911190986633 - f1-score (micro avg) 0.0078
2023-10-10 21:46:47,467 saving best model
2023-10-10 21:46:48,888 ----------------------------------------------------------------------------------------------------
2023-10-10 21:46:59,134 epoch 4 - iter 14/146 - loss 0.30368856 - time (sec): 10.24 - samples/sec: 462.62 - lr: 0.000115 - momentum: 0.000000
2023-10-10 21:47:08,517 epoch 4 - iter 28/146 - loss 0.28383264 - time (sec): 19.63 - samples/sec: 438.24 - lr: 0.000114 - momentum: 0.000000
2023-10-10 21:47:18,673 epoch 4 - iter 42/146 - loss 0.34186303 - time (sec): 29.78 - samples/sec: 452.85 - lr: 0.000112 - momentum: 0.000000
2023-10-10 21:47:28,077 epoch 4 - iter 56/146 - loss 0.34206921 - time (sec): 39.19 - samples/sec: 446.05 - lr: 0.000111 - momentum: 0.000000
2023-10-10 21:47:38,755 epoch 4 - iter 70/146 - loss 0.33835857 - time (sec): 49.87 - samples/sec: 443.34 - lr: 0.000109 - momentum: 0.000000
2023-10-10 21:47:48,562 epoch 4 - iter 84/146 - loss 0.33883733 - time (sec): 59.67 - samples/sec: 441.55 - lr: 0.000107 - momentum: 0.000000
2023-10-10 21:47:59,381 epoch 4 - iter 98/146 - loss 0.33112961 - time (sec): 70.49 - samples/sec: 439.76 - lr: 0.000106 - momentum: 0.000000
2023-10-10 21:48:08,664 epoch 4 - iter 112/146 - loss 0.32683644 - time (sec): 79.77 - samples/sec: 436.20 - lr: 0.000104 - momentum: 0.000000
2023-10-10 21:48:18,314 epoch 4 - iter 126/146 - loss 0.32139242 - time (sec): 89.42 - samples/sec: 434.36 - lr: 0.000103 - momentum: 0.000000
2023-10-10 21:48:27,775 epoch 4 - iter 140/146 - loss 0.32278398 - time (sec): 98.88 - samples/sec: 429.61 - lr: 0.000101 - momentum: 0.000000
2023-10-10 21:48:32,032 ----------------------------------------------------------------------------------------------------
2023-10-10 21:48:32,033 EPOCH 4 done: loss 0.3173 - lr: 0.000101
2023-10-10 21:48:38,292 DEV : loss 0.23751258850097656 - f1-score (micro avg) 0.3686
2023-10-10 21:48:38,302 saving best model
2023-10-10 21:48:47,130 ----------------------------------------------------------------------------------------------------
2023-10-10 21:48:56,208 epoch 5 - iter 14/146 - loss 0.26938515 - time (sec): 9.07 - samples/sec: 453.04 - lr: 0.000099 - momentum: 0.000000
2023-10-10 21:49:06,203 epoch 5 - iter 28/146 - loss 0.33101405 - time (sec): 19.07 - samples/sec: 475.22 - lr: 0.000097 - momentum: 0.000000
2023-10-10 21:49:15,596 epoch 5 - iter 42/146 - loss 0.31691507 - time (sec): 28.46 - samples/sec: 466.90 - lr: 0.000096 - momentum: 0.000000
2023-10-10 21:49:24,647 epoch 5 - iter 56/146 - loss 0.28600023 - time (sec): 37.51 - samples/sec: 461.87 - lr: 0.000094 - momentum: 0.000000
2023-10-10 21:49:34,897 epoch 5 - iter 70/146 - loss 0.27334427 - time (sec): 47.76 - samples/sec: 459.23 - lr: 0.000092 - momentum: 0.000000
2023-10-10 21:49:44,423 epoch 5 - iter 84/146 - loss 0.26564120 - time (sec): 57.29 - samples/sec: 452.58 - lr: 0.000091 - momentum: 0.000000
2023-10-10 21:49:53,632 epoch 5 - iter 98/146 - loss 0.26142888 - time (sec): 66.50 - samples/sec: 454.26 - lr: 0.000089 - momentum: 0.000000
2023-10-10 21:50:02,850 epoch 5 - iter 112/146 - loss 0.25808288 - time (sec): 75.72 - samples/sec: 458.61 - lr: 0.000088 - momentum: 0.000000
2023-10-10 21:50:11,451 epoch 5 - iter 126/146 - loss 0.25610218 - time (sec): 84.32 - samples/sec: 458.55 - lr: 0.000086 - momentum: 0.000000
2023-10-10 21:50:20,211 epoch 5 - iter 140/146 - loss 0.25265262 - time (sec): 93.08 - samples/sec: 458.37 - lr: 0.000084 - momentum: 0.000000
2023-10-10 21:50:23,964 ----------------------------------------------------------------------------------------------------
2023-10-10 21:50:23,964 EPOCH 5 done: loss 0.2526 - lr: 0.000084
2023-10-10 21:50:29,941 DEV : loss 0.20196418464183807 - f1-score (micro avg) 0.4746
2023-10-10 21:50:29,950 saving best model
2023-10-10 21:50:37,460 ----------------------------------------------------------------------------------------------------
2023-10-10 21:50:47,309 epoch 6 - iter 14/146 - loss 0.21125154 - time (sec): 9.84 - samples/sec: 508.80 - lr: 0.000082 - momentum: 0.000000
2023-10-10 21:50:56,084 epoch 6 - iter 28/146 - loss 0.22445195 - time (sec): 18.62 - samples/sec: 481.00 - lr: 0.000081 - momentum: 0.000000
2023-10-10 21:51:05,040 epoch 6 - iter 42/146 - loss 0.21454292 - time (sec): 27.58 - samples/sec: 476.11 - lr: 0.000079 - momentum: 0.000000
2023-10-10 21:51:13,750 epoch 6 - iter 56/146 - loss 0.20511075 - time (sec): 36.29 - samples/sec: 485.70 - lr: 0.000077 - momentum: 0.000000
2023-10-10 21:51:23,059 epoch 6 - iter 70/146 - loss 0.20516919 - time (sec): 45.59 - samples/sec: 488.85 - lr: 0.000076 - momentum: 0.000000
2023-10-10 21:51:31,377 epoch 6 - iter 84/146 - loss 0.20221448 - time (sec): 53.91 - samples/sec: 480.18 - lr: 0.000074 - momentum: 0.000000
2023-10-10 21:51:40,433 epoch 6 - iter 98/146 - loss 0.20584764 - time (sec): 62.97 - samples/sec: 474.79 - lr: 0.000073 - momentum: 0.000000
2023-10-10 21:51:49,118 epoch 6 - iter 112/146 - loss 0.19933841 - time (sec): 71.65 - samples/sec: 474.95 - lr: 0.000071 - momentum: 0.000000
2023-10-10 21:51:58,017 epoch 6 - iter 126/146 - loss 0.20436836 - time (sec): 80.55 - samples/sec: 480.15 - lr: 0.000069 - momentum: 0.000000
2023-10-10 21:52:05,954 epoch 6 - iter 140/146 - loss 0.20428475 - time (sec): 88.49 - samples/sec: 475.67 - lr: 0.000068 - momentum: 0.000000
2023-10-10 21:52:10,321 ----------------------------------------------------------------------------------------------------
2023-10-10 21:52:10,322 EPOCH 6 done: loss 0.2022 - lr: 0.000068
2023-10-10 21:52:16,082 DEV : loss 0.17842039465904236 - f1-score (micro avg) 0.5636
2023-10-10 21:52:16,091 saving best model
2023-10-10 21:52:24,741 ----------------------------------------------------------------------------------------------------
2023-10-10 21:52:34,632 epoch 7 - iter 14/146 - loss 0.16519836 - time (sec): 9.89 - samples/sec: 484.44 - lr: 0.000066 - momentum: 0.000000
2023-10-10 21:52:43,455 epoch 7 - iter 28/146 - loss 0.15491464 - time (sec): 18.71 - samples/sec: 459.41 - lr: 0.000064 - momentum: 0.000000
2023-10-10 21:52:52,480 epoch 7 - iter 42/146 - loss 0.17736530 - time (sec): 27.74 - samples/sec: 460.60 - lr: 0.000062 - momentum: 0.000000
2023-10-10 21:53:01,686 epoch 7 - iter 56/146 - loss 0.16335559 - time (sec): 36.94 - samples/sec: 477.54 - lr: 0.000061 - momentum: 0.000000
2023-10-10 21:53:10,637 epoch 7 - iter 70/146 - loss 0.15852353 - time (sec): 45.89 - samples/sec: 477.37 - lr: 0.000059 - momentum: 0.000000
2023-10-10 21:53:18,935 epoch 7 - iter 84/146 - loss 0.16471171 - time (sec): 54.19 - samples/sec: 468.14 - lr: 0.000058 - momentum: 0.000000
2023-10-10 21:53:28,006 epoch 7 - iter 98/146 - loss 0.16359888 - time (sec): 63.26 - samples/sec: 467.46 - lr: 0.000056 - momentum: 0.000000
2023-10-10 21:53:37,417 epoch 7 - iter 112/146 - loss 0.15896974 - time (sec): 72.67 - samples/sec: 473.77 - lr: 0.000054 - momentum: 0.000000
2023-10-10 21:53:46,512 epoch 7 - iter 126/146 - loss 0.15746455 - time (sec): 81.77 - samples/sec: 474.51 - lr: 0.000053 - momentum: 0.000000
2023-10-10 21:53:56,111 epoch 7 - iter 140/146 - loss 0.16140715 - time (sec): 91.37 - samples/sec: 463.92 - lr: 0.000051 - momentum: 0.000000
2023-10-10 21:54:00,586 ----------------------------------------------------------------------------------------------------
2023-10-10 21:54:00,587 EPOCH 7 done: loss 0.1639 - lr: 0.000051
2023-10-10 21:54:07,600 DEV : loss 0.16299203038215637 - f1-score (micro avg) 0.6061
2023-10-10 21:54:07,610 saving best model
2023-10-10 21:54:15,265 ----------------------------------------------------------------------------------------------------
2023-10-10 21:54:24,807 epoch 8 - iter 14/146 - loss 0.14054633 - time (sec): 9.54 - samples/sec: 515.83 - lr: 0.000049 - momentum: 0.000000
2023-10-10 21:54:33,707 epoch 8 - iter 28/146 - loss 0.15463910 - time (sec): 18.44 - samples/sec: 526.26 - lr: 0.000047 - momentum: 0.000000
2023-10-10 21:54:42,354 epoch 8 - iter 42/146 - loss 0.15471171 - time (sec): 27.08 - samples/sec: 512.50 - lr: 0.000046 - momentum: 0.000000
2023-10-10 21:54:50,606 epoch 8 - iter 56/146 - loss 0.14788860 - time (sec): 35.34 - samples/sec: 498.83 - lr: 0.000044 - momentum: 0.000000
2023-10-10 21:54:59,640 epoch 8 - iter 70/146 - loss 0.14632606 - time (sec): 44.37 - samples/sec: 498.54 - lr: 0.000043 - momentum: 0.000000
2023-10-10 21:55:08,163 epoch 8 - iter 84/146 - loss 0.15098798 - time (sec): 52.89 - samples/sec: 491.80 - lr: 0.000041 - momentum: 0.000000
2023-10-10 21:55:17,245 epoch 8 - iter 98/146 - loss 0.14595901 - time (sec): 61.98 - samples/sec: 483.64 - lr: 0.000039 - momentum: 0.000000
2023-10-10 21:55:27,185 epoch 8 - iter 112/146 - loss 0.14346752 - time (sec): 71.92 - samples/sec: 480.22 - lr: 0.000038 - momentum: 0.000000
2023-10-10 21:55:36,937 epoch 8 - iter 126/146 - loss 0.14157061 - time (sec): 81.67 - samples/sec: 471.85 - lr: 0.000036 - momentum: 0.000000
2023-10-10 21:55:46,970 epoch 8 - iter 140/146 - loss 0.13629684 - time (sec): 91.70 - samples/sec: 467.06 - lr: 0.000035 - momentum: 0.000000
2023-10-10 21:55:50,860 ----------------------------------------------------------------------------------------------------
2023-10-10 21:55:50,860 EPOCH 8 done: loss 0.1398 - lr: 0.000035
2023-10-10 21:55:56,778 DEV : loss 0.15665240585803986 - f1-score (micro avg) 0.6504
2023-10-10 21:55:56,788 saving best model
2023-10-10 21:56:06,636 ----------------------------------------------------------------------------------------------------
2023-10-10 21:56:16,167 epoch 9 - iter 14/146 - loss 0.10274893 - time (sec): 9.53 - samples/sec: 441.91 - lr: 0.000032 - momentum: 0.000000
2023-10-10 21:56:26,757 epoch 9 - iter 28/146 - loss 0.13023131 - time (sec): 20.12 - samples/sec: 454.61 - lr: 0.000031 - momentum: 0.000000
2023-10-10 21:56:36,486 epoch 9 - iter 42/146 - loss 0.12947726 - time (sec): 29.85 - samples/sec: 444.52 - lr: 0.000029 - momentum: 0.000000
2023-10-10 21:56:46,259 epoch 9 - iter 56/146 - loss 0.12059411 - time (sec): 39.62 - samples/sec: 442.36 - lr: 0.000028 - momentum: 0.000000
2023-10-10 21:56:55,148 epoch 9 - iter 70/146 - loss 0.12012299 - time (sec): 48.51 - samples/sec: 450.59 - lr: 0.000026 - momentum: 0.000000
2023-10-10 21:57:03,385 epoch 9 - iter 84/146 - loss 0.12128595 - time (sec): 56.74 - samples/sec: 456.05 - lr: 0.000024 - momentum: 0.000000
2023-10-10 21:57:12,121 epoch 9 - iter 98/146 - loss 0.12559222 - time (sec): 65.48 - samples/sec: 461.22 - lr: 0.000023 - momentum: 0.000000
2023-10-10 21:57:20,492 epoch 9 - iter 112/146 - loss 0.12343906 - time (sec): 73.85 - samples/sec: 465.07 - lr: 0.000021 - momentum: 0.000000
2023-10-10 21:57:29,518 epoch 9 - iter 126/146 - loss 0.12193938 - time (sec): 82.88 - samples/sec: 472.03 - lr: 0.000020 - momentum: 0.000000
2023-10-10 21:57:37,883 epoch 9 - iter 140/146 - loss 0.12379814 - time (sec): 91.24 - samples/sec: 470.00 - lr: 0.000018 - momentum: 0.000000
2023-10-10 21:57:41,307 ----------------------------------------------------------------------------------------------------
2023-10-10 21:57:41,307 EPOCH 9 done: loss 0.1234 - lr: 0.000018
2023-10-10 21:57:47,299 DEV : loss 0.1573924571275711 - f1-score (micro avg) 0.6348
2023-10-10 21:57:47,308 ----------------------------------------------------------------------------------------------------
2023-10-10 21:57:56,128 epoch 10 - iter 14/146 - loss 0.10385368 - time (sec): 8.82 - samples/sec: 508.61 - lr: 0.000016 - momentum: 0.000000
2023-10-10 21:58:04,303 epoch 10 - iter 28/146 - loss 0.12007872 - time (sec): 16.99 - samples/sec: 479.61 - lr: 0.000014 - momentum: 0.000000
2023-10-10 21:58:12,720 epoch 10 - iter 42/146 - loss 0.11224087 - time (sec): 25.41 - samples/sec: 485.72 - lr: 0.000013 - momentum: 0.000000
2023-10-10 21:58:22,256 epoch 10 - iter 56/146 - loss 0.10217048 - time (sec): 34.95 - samples/sec: 503.12 - lr: 0.000011 - momentum: 0.000000
2023-10-10 21:58:31,488 epoch 10 - iter 70/146 - loss 0.10189941 - time (sec): 44.18 - samples/sec: 506.27 - lr: 0.000009 - momentum: 0.000000
2023-10-10 21:58:39,895 epoch 10 - iter 84/146 - loss 0.10134392 - time (sec): 52.59 - samples/sec: 499.85 - lr: 0.000008 - momentum: 0.000000
2023-10-10 21:58:48,328 epoch 10 - iter 98/146 - loss 0.10206351 - time (sec): 61.02 - samples/sec: 498.54 - lr: 0.000006 - momentum: 0.000000
2023-10-10 21:58:57,115 epoch 10 - iter 112/146 - loss 0.10614914 - time (sec): 69.81 - samples/sec: 497.17 - lr: 0.000005 - momentum: 0.000000
2023-10-10 21:59:05,599 epoch 10 - iter 126/146 - loss 0.10946359 - time (sec): 78.29 - samples/sec: 494.50 - lr: 0.000003 - momentum: 0.000000
2023-10-10 21:59:14,323 epoch 10 - iter 140/146 - loss 0.11293547 - time (sec): 87.01 - samples/sec: 494.56 - lr: 0.000001 - momentum: 0.000000
2023-10-10 21:59:17,682 ----------------------------------------------------------------------------------------------------
2023-10-10 21:59:17,683 EPOCH 10 done: loss 0.1160 - lr: 0.000001
2023-10-10 21:59:23,725 DEV : loss 0.15526741743087769 - f1-score (micro avg) 0.6783
2023-10-10 21:59:23,734 saving best model
2023-10-10 21:59:32,858 ----------------------------------------------------------------------------------------------------
2023-10-10 21:59:32,860 Loading model from best epoch ...
2023-10-10 21:59:38,272 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 21:59:51,377
Results:
- F-score (micro) 0.7005
- F-score (macro) 0.6111
- Accuracy 0.5721
By class:
precision recall f1-score support
PER 0.7970 0.7672 0.7818 348
LOC 0.5960 0.7969 0.6820 261
ORG 0.2857 0.3077 0.2963 52
HumanProd 0.8125 0.5909 0.6842 22
micro avg 0.6667 0.7379 0.7005 683
macro avg 0.6228 0.6157 0.6111 683
weighted avg 0.6818 0.7379 0.7036 683
2023-10-10 21:59:51,378 ----------------------------------------------------------------------------------------------------
|