File size: 25,202 Bytes
eedd550 |
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 11:42:52,153 ----------------------------------------------------------------------------------------------------
2023-10-11 11:42:52,155 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 11:42:52,155 ----------------------------------------------------------------------------------------------------
2023-10-11 11:42:52,156 MultiCorpus: 1085 train + 148 dev + 364 test sentences
- NER_HIPE_2022 Corpus: 1085 train + 148 dev + 364 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/sv/with_doc_seperator
2023-10-11 11:42:52,156 ----------------------------------------------------------------------------------------------------
2023-10-11 11:42:52,156 Train: 1085 sentences
2023-10-11 11:42:52,156 (train_with_dev=False, train_with_test=False)
2023-10-11 11:42:52,156 ----------------------------------------------------------------------------------------------------
2023-10-11 11:42:52,156 Training Params:
2023-10-11 11:42:52,156 - learning_rate: "0.00016"
2023-10-11 11:42:52,156 - mini_batch_size: "8"
2023-10-11 11:42:52,156 - max_epochs: "10"
2023-10-11 11:42:52,156 - shuffle: "True"
2023-10-11 11:42:52,156 ----------------------------------------------------------------------------------------------------
2023-10-11 11:42:52,156 Plugins:
2023-10-11 11:42:52,156 - TensorboardLogger
2023-10-11 11:42:52,157 - LinearScheduler | warmup_fraction: '0.1'
2023-10-11 11:42:52,157 ----------------------------------------------------------------------------------------------------
2023-10-11 11:42:52,157 Final evaluation on model from best epoch (best-model.pt)
2023-10-11 11:42:52,157 - metric: "('micro avg', 'f1-score')"
2023-10-11 11:42:52,157 ----------------------------------------------------------------------------------------------------
2023-10-11 11:42:52,157 Computation:
2023-10-11 11:42:52,157 - compute on device: cuda:0
2023-10-11 11:42:52,157 - embedding storage: none
2023-10-11 11:42:52,157 ----------------------------------------------------------------------------------------------------
2023-10-11 11:42:52,157 Model training base path: "hmbench-newseye/sv-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs8-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-4"
2023-10-11 11:42:52,157 ----------------------------------------------------------------------------------------------------
2023-10-11 11:42:52,157 ----------------------------------------------------------------------------------------------------
2023-10-11 11:42:52,157 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-11 11:43:00,927 epoch 1 - iter 13/136 - loss 2.84089892 - time (sec): 8.77 - samples/sec: 615.10 - lr: 0.000014 - momentum: 0.000000
2023-10-11 11:43:10,129 epoch 1 - iter 26/136 - loss 2.83498049 - time (sec): 17.97 - samples/sec: 615.85 - lr: 0.000029 - momentum: 0.000000
2023-10-11 11:43:18,889 epoch 1 - iter 39/136 - loss 2.82406903 - time (sec): 26.73 - samples/sec: 601.04 - lr: 0.000045 - momentum: 0.000000
2023-10-11 11:43:27,568 epoch 1 - iter 52/136 - loss 2.80264022 - time (sec): 35.41 - samples/sec: 593.72 - lr: 0.000060 - momentum: 0.000000
2023-10-11 11:43:35,878 epoch 1 - iter 65/136 - loss 2.76592580 - time (sec): 43.72 - samples/sec: 582.84 - lr: 0.000075 - momentum: 0.000000
2023-10-11 11:43:44,185 epoch 1 - iter 78/136 - loss 2.71169051 - time (sec): 52.03 - samples/sec: 577.85 - lr: 0.000091 - momentum: 0.000000
2023-10-11 11:43:52,478 epoch 1 - iter 91/136 - loss 2.64797024 - time (sec): 60.32 - samples/sec: 571.71 - lr: 0.000106 - momentum: 0.000000
2023-10-11 11:44:01,392 epoch 1 - iter 104/136 - loss 2.56715392 - time (sec): 69.23 - samples/sec: 576.23 - lr: 0.000121 - momentum: 0.000000
2023-10-11 11:44:09,241 epoch 1 - iter 117/136 - loss 2.50180341 - time (sec): 77.08 - samples/sec: 572.72 - lr: 0.000136 - momentum: 0.000000
2023-10-11 11:44:18,110 epoch 1 - iter 130/136 - loss 2.41593244 - time (sec): 85.95 - samples/sec: 571.04 - lr: 0.000152 - momentum: 0.000000
2023-10-11 11:44:22,531 ----------------------------------------------------------------------------------------------------
2023-10-11 11:44:22,532 EPOCH 1 done: loss 2.3622 - lr: 0.000152
2023-10-11 11:44:27,578 DEV : loss 1.331149935722351 - f1-score (micro avg) 0.0
2023-10-11 11:44:27,587 ----------------------------------------------------------------------------------------------------
2023-10-11 11:44:36,368 epoch 2 - iter 13/136 - loss 1.32887985 - time (sec): 8.78 - samples/sec: 602.57 - lr: 0.000158 - momentum: 0.000000
2023-10-11 11:44:44,636 epoch 2 - iter 26/136 - loss 1.26733038 - time (sec): 17.05 - samples/sec: 574.22 - lr: 0.000157 - momentum: 0.000000
2023-10-11 11:44:52,303 epoch 2 - iter 39/136 - loss 1.20119028 - time (sec): 24.71 - samples/sec: 550.00 - lr: 0.000155 - momentum: 0.000000
2023-10-11 11:45:01,401 epoch 2 - iter 52/136 - loss 1.10723903 - time (sec): 33.81 - samples/sec: 560.91 - lr: 0.000153 - momentum: 0.000000
2023-10-11 11:45:11,107 epoch 2 - iter 65/136 - loss 1.02852991 - time (sec): 43.52 - samples/sec: 563.78 - lr: 0.000152 - momentum: 0.000000
2023-10-11 11:45:20,058 epoch 2 - iter 78/136 - loss 0.96147737 - time (sec): 52.47 - samples/sec: 564.12 - lr: 0.000150 - momentum: 0.000000
2023-10-11 11:45:28,380 epoch 2 - iter 91/136 - loss 0.90827453 - time (sec): 60.79 - samples/sec: 560.43 - lr: 0.000148 - momentum: 0.000000
2023-10-11 11:45:36,838 epoch 2 - iter 104/136 - loss 0.85911401 - time (sec): 69.25 - samples/sec: 560.82 - lr: 0.000147 - momentum: 0.000000
2023-10-11 11:45:44,748 epoch 2 - iter 117/136 - loss 0.83120687 - time (sec): 77.16 - samples/sec: 559.18 - lr: 0.000145 - momentum: 0.000000
2023-10-11 11:45:54,150 epoch 2 - iter 130/136 - loss 0.79282972 - time (sec): 86.56 - samples/sec: 570.10 - lr: 0.000143 - momentum: 0.000000
2023-10-11 11:45:58,234 ----------------------------------------------------------------------------------------------------
2023-10-11 11:45:58,235 EPOCH 2 done: loss 0.7761 - lr: 0.000143
2023-10-11 11:46:03,660 DEV : loss 0.3906053304672241 - f1-score (micro avg) 0.0
2023-10-11 11:46:03,668 ----------------------------------------------------------------------------------------------------
2023-10-11 11:46:12,037 epoch 3 - iter 13/136 - loss 0.34739828 - time (sec): 8.37 - samples/sec: 556.71 - lr: 0.000141 - momentum: 0.000000
2023-10-11 11:46:21,281 epoch 3 - iter 26/136 - loss 0.39816021 - time (sec): 17.61 - samples/sec: 600.01 - lr: 0.000139 - momentum: 0.000000
2023-10-11 11:46:29,696 epoch 3 - iter 39/136 - loss 0.40218390 - time (sec): 26.03 - samples/sec: 601.13 - lr: 0.000137 - momentum: 0.000000
2023-10-11 11:46:38,051 epoch 3 - iter 52/136 - loss 0.40576799 - time (sec): 34.38 - samples/sec: 595.88 - lr: 0.000136 - momentum: 0.000000
2023-10-11 11:46:46,295 epoch 3 - iter 65/136 - loss 0.39917278 - time (sec): 42.63 - samples/sec: 592.30 - lr: 0.000134 - momentum: 0.000000
2023-10-11 11:46:54,812 epoch 3 - iter 78/136 - loss 0.39934212 - time (sec): 51.14 - samples/sec: 592.05 - lr: 0.000132 - momentum: 0.000000
2023-10-11 11:47:03,667 epoch 3 - iter 91/136 - loss 0.39725495 - time (sec): 60.00 - samples/sec: 596.39 - lr: 0.000131 - momentum: 0.000000
2023-10-11 11:47:11,817 epoch 3 - iter 104/136 - loss 0.39381516 - time (sec): 68.15 - samples/sec: 589.08 - lr: 0.000129 - momentum: 0.000000
2023-10-11 11:47:20,374 epoch 3 - iter 117/136 - loss 0.38541873 - time (sec): 76.70 - samples/sec: 590.97 - lr: 0.000127 - momentum: 0.000000
2023-10-11 11:47:28,386 epoch 3 - iter 130/136 - loss 0.38481317 - time (sec): 84.72 - samples/sec: 587.10 - lr: 0.000126 - momentum: 0.000000
2023-10-11 11:47:32,136 ----------------------------------------------------------------------------------------------------
2023-10-11 11:47:32,136 EPOCH 3 done: loss 0.3826 - lr: 0.000126
2023-10-11 11:47:37,658 DEV : loss 0.2841358780860901 - f1-score (micro avg) 0.303
2023-10-11 11:47:37,666 saving best model
2023-10-11 11:47:38,500 ----------------------------------------------------------------------------------------------------
2023-10-11 11:47:46,931 epoch 4 - iter 13/136 - loss 0.30981951 - time (sec): 8.43 - samples/sec: 565.84 - lr: 0.000123 - momentum: 0.000000
2023-10-11 11:47:55,313 epoch 4 - iter 26/136 - loss 0.26209209 - time (sec): 16.81 - samples/sec: 572.91 - lr: 0.000121 - momentum: 0.000000
2023-10-11 11:48:04,432 epoch 4 - iter 39/136 - loss 0.29742145 - time (sec): 25.93 - samples/sec: 602.75 - lr: 0.000120 - momentum: 0.000000
2023-10-11 11:48:12,946 epoch 4 - iter 52/136 - loss 0.30087458 - time (sec): 34.44 - samples/sec: 600.55 - lr: 0.000118 - momentum: 0.000000
2023-10-11 11:48:21,575 epoch 4 - iter 65/136 - loss 0.29630507 - time (sec): 43.07 - samples/sec: 597.25 - lr: 0.000116 - momentum: 0.000000
2023-10-11 11:48:30,045 epoch 4 - iter 78/136 - loss 0.29026697 - time (sec): 51.54 - samples/sec: 593.86 - lr: 0.000115 - momentum: 0.000000
2023-10-11 11:48:38,662 epoch 4 - iter 91/136 - loss 0.29000751 - time (sec): 60.16 - samples/sec: 596.55 - lr: 0.000113 - momentum: 0.000000
2023-10-11 11:48:46,927 epoch 4 - iter 104/136 - loss 0.28862524 - time (sec): 68.42 - samples/sec: 591.68 - lr: 0.000111 - momentum: 0.000000
2023-10-11 11:48:55,506 epoch 4 - iter 117/136 - loss 0.29291057 - time (sec): 77.00 - samples/sec: 588.96 - lr: 0.000109 - momentum: 0.000000
2023-10-11 11:49:03,284 epoch 4 - iter 130/136 - loss 0.29357331 - time (sec): 84.78 - samples/sec: 580.03 - lr: 0.000108 - momentum: 0.000000
2023-10-11 11:49:07,528 ----------------------------------------------------------------------------------------------------
2023-10-11 11:49:07,528 EPOCH 4 done: loss 0.2924 - lr: 0.000108
2023-10-11 11:49:13,047 DEV : loss 0.24194754660129547 - f1-score (micro avg) 0.361
2023-10-11 11:49:13,055 saving best model
2023-10-11 11:49:15,756 ----------------------------------------------------------------------------------------------------
2023-10-11 11:49:24,571 epoch 5 - iter 13/136 - loss 0.25599446 - time (sec): 8.81 - samples/sec: 607.84 - lr: 0.000105 - momentum: 0.000000
2023-10-11 11:49:33,041 epoch 5 - iter 26/136 - loss 0.26637160 - time (sec): 17.28 - samples/sec: 592.68 - lr: 0.000104 - momentum: 0.000000
2023-10-11 11:49:41,925 epoch 5 - iter 39/136 - loss 0.26414833 - time (sec): 26.16 - samples/sec: 599.21 - lr: 0.000102 - momentum: 0.000000
2023-10-11 11:49:51,071 epoch 5 - iter 52/136 - loss 0.25660484 - time (sec): 35.31 - samples/sec: 606.04 - lr: 0.000100 - momentum: 0.000000
2023-10-11 11:49:59,386 epoch 5 - iter 65/136 - loss 0.25490717 - time (sec): 43.63 - samples/sec: 600.20 - lr: 0.000099 - momentum: 0.000000
2023-10-11 11:50:07,873 epoch 5 - iter 78/136 - loss 0.25012610 - time (sec): 52.11 - samples/sec: 591.76 - lr: 0.000097 - momentum: 0.000000
2023-10-11 11:50:16,091 epoch 5 - iter 91/136 - loss 0.24408605 - time (sec): 60.33 - samples/sec: 585.86 - lr: 0.000095 - momentum: 0.000000
2023-10-11 11:50:24,451 epoch 5 - iter 104/136 - loss 0.24082052 - time (sec): 68.69 - samples/sec: 583.79 - lr: 0.000093 - momentum: 0.000000
2023-10-11 11:50:32,782 epoch 5 - iter 117/136 - loss 0.23933921 - time (sec): 77.02 - samples/sec: 581.82 - lr: 0.000092 - momentum: 0.000000
2023-10-11 11:50:41,368 epoch 5 - iter 130/136 - loss 0.23509442 - time (sec): 85.61 - samples/sec: 581.75 - lr: 0.000090 - momentum: 0.000000
2023-10-11 11:50:45,040 ----------------------------------------------------------------------------------------------------
2023-10-11 11:50:45,041 EPOCH 5 done: loss 0.2334 - lr: 0.000090
2023-10-11 11:50:50,569 DEV : loss 0.2023283690214157 - f1-score (micro avg) 0.5352
2023-10-11 11:50:50,577 saving best model
2023-10-11 11:50:53,072 ----------------------------------------------------------------------------------------------------
2023-10-11 11:51:01,826 epoch 6 - iter 13/136 - loss 0.18517967 - time (sec): 8.75 - samples/sec: 571.85 - lr: 0.000088 - momentum: 0.000000
2023-10-11 11:51:10,387 epoch 6 - iter 26/136 - loss 0.18845994 - time (sec): 17.31 - samples/sec: 580.12 - lr: 0.000086 - momentum: 0.000000
2023-10-11 11:51:19,918 epoch 6 - iter 39/136 - loss 0.18613737 - time (sec): 26.84 - samples/sec: 598.85 - lr: 0.000084 - momentum: 0.000000
2023-10-11 11:51:27,768 epoch 6 - iter 52/136 - loss 0.18257001 - time (sec): 34.69 - samples/sec: 587.24 - lr: 0.000083 - momentum: 0.000000
2023-10-11 11:51:35,920 epoch 6 - iter 65/136 - loss 0.17864499 - time (sec): 42.85 - samples/sec: 585.99 - lr: 0.000081 - momentum: 0.000000
2023-10-11 11:51:44,330 epoch 6 - iter 78/136 - loss 0.17689256 - time (sec): 51.26 - samples/sec: 584.99 - lr: 0.000079 - momentum: 0.000000
2023-10-11 11:51:52,541 epoch 6 - iter 91/136 - loss 0.18469652 - time (sec): 59.47 - samples/sec: 586.24 - lr: 0.000077 - momentum: 0.000000
2023-10-11 11:52:00,605 epoch 6 - iter 104/136 - loss 0.18264529 - time (sec): 67.53 - samples/sec: 582.48 - lr: 0.000076 - momentum: 0.000000
2023-10-11 11:52:09,270 epoch 6 - iter 117/136 - loss 0.18078056 - time (sec): 76.19 - samples/sec: 585.17 - lr: 0.000074 - momentum: 0.000000
2023-10-11 11:52:18,305 epoch 6 - iter 130/136 - loss 0.17744831 - time (sec): 85.23 - samples/sec: 589.03 - lr: 0.000072 - momentum: 0.000000
2023-10-11 11:52:21,741 ----------------------------------------------------------------------------------------------------
2023-10-11 11:52:21,741 EPOCH 6 done: loss 0.1785 - lr: 0.000072
2023-10-11 11:52:27,298 DEV : loss 0.18196691572666168 - f1-score (micro avg) 0.6184
2023-10-11 11:52:27,306 saving best model
2023-10-11 11:52:29,802 ----------------------------------------------------------------------------------------------------
2023-10-11 11:52:38,343 epoch 7 - iter 13/136 - loss 0.13573850 - time (sec): 8.54 - samples/sec: 550.59 - lr: 0.000070 - momentum: 0.000000
2023-10-11 11:52:46,362 epoch 7 - iter 26/136 - loss 0.15154756 - time (sec): 16.56 - samples/sec: 539.09 - lr: 0.000068 - momentum: 0.000000
2023-10-11 11:52:54,607 epoch 7 - iter 39/136 - loss 0.14147926 - time (sec): 24.80 - samples/sec: 545.99 - lr: 0.000067 - momentum: 0.000000
2023-10-11 11:53:02,804 epoch 7 - iter 52/136 - loss 0.14239010 - time (sec): 33.00 - samples/sec: 544.29 - lr: 0.000065 - momentum: 0.000000
2023-10-11 11:53:13,375 epoch 7 - iter 65/136 - loss 0.13702337 - time (sec): 43.57 - samples/sec: 546.13 - lr: 0.000063 - momentum: 0.000000
2023-10-11 11:53:22,841 epoch 7 - iter 78/136 - loss 0.13307308 - time (sec): 53.03 - samples/sec: 556.81 - lr: 0.000061 - momentum: 0.000000
2023-10-11 11:53:31,166 epoch 7 - iter 91/136 - loss 0.13621321 - time (sec): 61.36 - samples/sec: 553.56 - lr: 0.000060 - momentum: 0.000000
2023-10-11 11:53:39,831 epoch 7 - iter 104/136 - loss 0.13813726 - time (sec): 70.02 - samples/sec: 553.89 - lr: 0.000058 - momentum: 0.000000
2023-10-11 11:53:48,877 epoch 7 - iter 117/136 - loss 0.13858223 - time (sec): 79.07 - samples/sec: 560.08 - lr: 0.000056 - momentum: 0.000000
2023-10-11 11:53:57,820 epoch 7 - iter 130/136 - loss 0.13942011 - time (sec): 88.01 - samples/sec: 563.86 - lr: 0.000055 - momentum: 0.000000
2023-10-11 11:54:01,781 ----------------------------------------------------------------------------------------------------
2023-10-11 11:54:01,781 EPOCH 7 done: loss 0.1413 - lr: 0.000055
2023-10-11 11:54:07,495 DEV : loss 0.1657494753599167 - f1-score (micro avg) 0.6118
2023-10-11 11:54:07,504 ----------------------------------------------------------------------------------------------------
2023-10-11 11:54:16,368 epoch 8 - iter 13/136 - loss 0.09943176 - time (sec): 8.86 - samples/sec: 569.40 - lr: 0.000052 - momentum: 0.000000
2023-10-11 11:54:25,369 epoch 8 - iter 26/136 - loss 0.12241690 - time (sec): 17.86 - samples/sec: 583.50 - lr: 0.000051 - momentum: 0.000000
2023-10-11 11:54:34,298 epoch 8 - iter 39/136 - loss 0.12160070 - time (sec): 26.79 - samples/sec: 591.13 - lr: 0.000049 - momentum: 0.000000
2023-10-11 11:54:42,211 epoch 8 - iter 52/136 - loss 0.11904427 - time (sec): 34.71 - samples/sec: 571.27 - lr: 0.000047 - momentum: 0.000000
2023-10-11 11:54:51,653 epoch 8 - iter 65/136 - loss 0.11543585 - time (sec): 44.15 - samples/sec: 581.03 - lr: 0.000045 - momentum: 0.000000
2023-10-11 11:55:01,140 epoch 8 - iter 78/136 - loss 0.11519079 - time (sec): 53.63 - samples/sec: 590.70 - lr: 0.000044 - momentum: 0.000000
2023-10-11 11:55:09,607 epoch 8 - iter 91/136 - loss 0.11768833 - time (sec): 62.10 - samples/sec: 583.56 - lr: 0.000042 - momentum: 0.000000
2023-10-11 11:55:17,593 epoch 8 - iter 104/136 - loss 0.11737793 - time (sec): 70.09 - samples/sec: 577.78 - lr: 0.000040 - momentum: 0.000000
2023-10-11 11:55:26,457 epoch 8 - iter 117/136 - loss 0.11880016 - time (sec): 78.95 - samples/sec: 580.16 - lr: 0.000039 - momentum: 0.000000
2023-10-11 11:55:34,585 epoch 8 - iter 130/136 - loss 0.11635025 - time (sec): 87.08 - samples/sec: 577.11 - lr: 0.000037 - momentum: 0.000000
2023-10-11 11:55:37,961 ----------------------------------------------------------------------------------------------------
2023-10-11 11:55:37,961 EPOCH 8 done: loss 0.1166 - lr: 0.000037
2023-10-11 11:55:43,939 DEV : loss 0.1629943549633026 - f1-score (micro avg) 0.6403
2023-10-11 11:55:43,947 saving best model
2023-10-11 11:55:46,470 ----------------------------------------------------------------------------------------------------
2023-10-11 11:55:54,803 epoch 9 - iter 13/136 - loss 0.12409007 - time (sec): 8.33 - samples/sec: 568.60 - lr: 0.000034 - momentum: 0.000000
2023-10-11 11:56:02,461 epoch 9 - iter 26/136 - loss 0.12338939 - time (sec): 15.99 - samples/sec: 549.75 - lr: 0.000033 - momentum: 0.000000
2023-10-11 11:56:10,655 epoch 9 - iter 39/136 - loss 0.11349678 - time (sec): 24.18 - samples/sec: 555.76 - lr: 0.000031 - momentum: 0.000000
2023-10-11 11:56:19,197 epoch 9 - iter 52/136 - loss 0.10693226 - time (sec): 32.72 - samples/sec: 566.11 - lr: 0.000029 - momentum: 0.000000
2023-10-11 11:56:28,242 epoch 9 - iter 65/136 - loss 0.10839615 - time (sec): 41.77 - samples/sec: 576.28 - lr: 0.000028 - momentum: 0.000000
2023-10-11 11:56:36,534 epoch 9 - iter 78/136 - loss 0.10371986 - time (sec): 50.06 - samples/sec: 575.29 - lr: 0.000026 - momentum: 0.000000
2023-10-11 11:56:45,697 epoch 9 - iter 91/136 - loss 0.09994810 - time (sec): 59.22 - samples/sec: 578.63 - lr: 0.000024 - momentum: 0.000000
2023-10-11 11:56:54,262 epoch 9 - iter 104/136 - loss 0.10282509 - time (sec): 67.79 - samples/sec: 577.37 - lr: 0.000023 - momentum: 0.000000
2023-10-11 11:57:03,462 epoch 9 - iter 117/136 - loss 0.10471070 - time (sec): 76.99 - samples/sec: 580.27 - lr: 0.000021 - momentum: 0.000000
2023-10-11 11:57:12,384 epoch 9 - iter 130/136 - loss 0.10371980 - time (sec): 85.91 - samples/sec: 582.84 - lr: 0.000019 - momentum: 0.000000
2023-10-11 11:57:15,903 ----------------------------------------------------------------------------------------------------
2023-10-11 11:57:15,903 EPOCH 9 done: loss 0.1028 - lr: 0.000019
2023-10-11 11:57:21,970 DEV : loss 0.15900003910064697 - f1-score (micro avg) 0.6486
2023-10-11 11:57:21,979 saving best model
2023-10-11 11:57:24,526 ----------------------------------------------------------------------------------------------------
2023-10-11 11:57:33,259 epoch 10 - iter 13/136 - loss 0.08673914 - time (sec): 8.73 - samples/sec: 577.96 - lr: 0.000017 - momentum: 0.000000
2023-10-11 11:57:41,971 epoch 10 - iter 26/136 - loss 0.10283299 - time (sec): 17.44 - samples/sec: 581.89 - lr: 0.000015 - momentum: 0.000000
2023-10-11 11:57:50,068 epoch 10 - iter 39/136 - loss 0.10096336 - time (sec): 25.54 - samples/sec: 576.52 - lr: 0.000013 - momentum: 0.000000
2023-10-11 11:57:58,268 epoch 10 - iter 52/136 - loss 0.10508712 - time (sec): 33.74 - samples/sec: 576.09 - lr: 0.000012 - momentum: 0.000000
2023-10-11 11:58:07,497 epoch 10 - iter 65/136 - loss 0.09663908 - time (sec): 42.97 - samples/sec: 582.26 - lr: 0.000010 - momentum: 0.000000
2023-10-11 11:58:15,641 epoch 10 - iter 78/136 - loss 0.09515246 - time (sec): 51.11 - samples/sec: 576.29 - lr: 0.000008 - momentum: 0.000000
2023-10-11 11:58:24,729 epoch 10 - iter 91/136 - loss 0.09381065 - time (sec): 60.20 - samples/sec: 577.73 - lr: 0.000007 - momentum: 0.000000
2023-10-11 11:58:33,569 epoch 10 - iter 104/136 - loss 0.09477746 - time (sec): 69.04 - samples/sec: 575.19 - lr: 0.000005 - momentum: 0.000000
2023-10-11 11:58:42,433 epoch 10 - iter 117/136 - loss 0.09519411 - time (sec): 77.90 - samples/sec: 577.00 - lr: 0.000003 - momentum: 0.000000
2023-10-11 11:58:51,065 epoch 10 - iter 130/136 - loss 0.09571016 - time (sec): 86.54 - samples/sec: 576.79 - lr: 0.000002 - momentum: 0.000000
2023-10-11 11:58:54,747 ----------------------------------------------------------------------------------------------------
2023-10-11 11:58:54,747 EPOCH 10 done: loss 0.0960 - lr: 0.000002
2023-10-11 11:59:00,673 DEV : loss 0.16201113164424896 - f1-score (micro avg) 0.6306
2023-10-11 11:59:01,568 ----------------------------------------------------------------------------------------------------
2023-10-11 11:59:01,570 Loading model from best epoch ...
2023-10-11 11:59:05,259 SequenceTagger predicts: Dictionary with 17 tags: O, S-LOC, B-LOC, E-LOC, I-LOC, S-PER, B-PER, E-PER, I-PER, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd, S-ORG, B-ORG, E-ORG, I-ORG
2023-10-11 11:59:17,559
Results:
- F-score (micro) 0.6285
- F-score (macro) 0.4257
- Accuracy 0.5154
By class:
precision recall f1-score support
LOC 0.6256 0.8462 0.7193 312
PER 0.6557 0.5769 0.6138 208
HumanProd 0.2429 0.7727 0.3696 22
ORG 0.0000 0.0000 0.0000 55
micro avg 0.5906 0.6717 0.6285 597
macro avg 0.3810 0.5490 0.4257 597
weighted avg 0.5644 0.6717 0.6034 597
2023-10-11 11:59:17,559 ----------------------------------------------------------------------------------------------------
|