File size: 24,982 Bytes
bc7e452 |
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 |
2023-10-08 20:20:45,628 ----------------------------------------------------------------------------------------------------
2023-10-08 20:20:45,629 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): T5LayerNorm()
(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): T5LayerNorm()
(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): T5LayerNorm()
(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): T5LayerNorm()
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
(final_layer_norm): T5LayerNorm()
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=1472, out_features=25, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-08 20:20:45,629 ----------------------------------------------------------------------------------------------------
2023-10-08 20:20:45,630 MultiCorpus: 966 train + 219 dev + 204 test sentences
- NER_HIPE_2022 Corpus: 966 train + 219 dev + 204 test sentences - /app/.flair/datasets/ner_hipe_2022/v2.1/ajmc/fr/with_doc_seperator
2023-10-08 20:20:45,630 ----------------------------------------------------------------------------------------------------
2023-10-08 20:20:45,630 Train: 966 sentences
2023-10-08 20:20:45,630 (train_with_dev=False, train_with_test=False)
2023-10-08 20:20:45,630 ----------------------------------------------------------------------------------------------------
2023-10-08 20:20:45,630 Training Params:
2023-10-08 20:20:45,630 - learning_rate: "0.00016"
2023-10-08 20:20:45,630 - mini_batch_size: "8"
2023-10-08 20:20:45,630 - max_epochs: "10"
2023-10-08 20:20:45,630 - shuffle: "True"
2023-10-08 20:20:45,630 ----------------------------------------------------------------------------------------------------
2023-10-08 20:20:45,630 Plugins:
2023-10-08 20:20:45,630 - TensorboardLogger
2023-10-08 20:20:45,630 - LinearScheduler | warmup_fraction: '0.1'
2023-10-08 20:20:45,630 ----------------------------------------------------------------------------------------------------
2023-10-08 20:20:45,630 Final evaluation on model from best epoch (best-model.pt)
2023-10-08 20:20:45,631 - metric: "('micro avg', 'f1-score')"
2023-10-08 20:20:45,631 ----------------------------------------------------------------------------------------------------
2023-10-08 20:20:45,631 Computation:
2023-10-08 20:20:45,631 - compute on device: cuda:0
2023-10-08 20:20:45,631 - embedding storage: none
2023-10-08 20:20:45,631 ----------------------------------------------------------------------------------------------------
2023-10-08 20:20:45,631 Model training base path: "hmbench-ajmc/fr-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs8-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-2"
2023-10-08 20:20:45,631 ----------------------------------------------------------------------------------------------------
2023-10-08 20:20:45,631 ----------------------------------------------------------------------------------------------------
2023-10-08 20:20:45,631 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-08 20:20:54,662 epoch 1 - iter 12/121 - loss 3.22933883 - time (sec): 9.03 - samples/sec: 288.82 - lr: 0.000015 - momentum: 0.000000
2023-10-08 20:21:04,168 epoch 1 - iter 24/121 - loss 3.22358323 - time (sec): 18.54 - samples/sec: 291.86 - lr: 0.000030 - momentum: 0.000000
2023-10-08 20:21:12,604 epoch 1 - iter 36/121 - loss 3.21405829 - time (sec): 26.97 - samples/sec: 286.52 - lr: 0.000046 - momentum: 0.000000
2023-10-08 20:21:20,504 epoch 1 - iter 48/121 - loss 3.19951121 - time (sec): 34.87 - samples/sec: 283.98 - lr: 0.000062 - momentum: 0.000000
2023-10-08 20:21:28,994 epoch 1 - iter 60/121 - loss 3.16865112 - time (sec): 43.36 - samples/sec: 282.97 - lr: 0.000078 - momentum: 0.000000
2023-10-08 20:21:37,353 epoch 1 - iter 72/121 - loss 3.11383349 - time (sec): 51.72 - samples/sec: 282.21 - lr: 0.000094 - momentum: 0.000000
2023-10-08 20:21:45,864 epoch 1 - iter 84/121 - loss 3.04197392 - time (sec): 60.23 - samples/sec: 283.76 - lr: 0.000110 - momentum: 0.000000
2023-10-08 20:21:54,507 epoch 1 - iter 96/121 - loss 2.96160655 - time (sec): 68.87 - samples/sec: 284.76 - lr: 0.000126 - momentum: 0.000000
2023-10-08 20:22:03,488 epoch 1 - iter 108/121 - loss 2.87149223 - time (sec): 77.86 - samples/sec: 286.53 - lr: 0.000141 - momentum: 0.000000
2023-10-08 20:22:11,628 epoch 1 - iter 120/121 - loss 2.79180261 - time (sec): 86.00 - samples/sec: 285.62 - lr: 0.000157 - momentum: 0.000000
2023-10-08 20:22:12,178 ----------------------------------------------------------------------------------------------------
2023-10-08 20:22:12,179 EPOCH 1 done: loss 2.7843 - lr: 0.000157
2023-10-08 20:22:18,010 DEV : loss 1.8020617961883545 - f1-score (micro avg) 0.0
2023-10-08 20:22:18,016 ----------------------------------------------------------------------------------------------------
2023-10-08 20:22:26,295 epoch 2 - iter 12/121 - loss 1.79034078 - time (sec): 8.28 - samples/sec: 280.51 - lr: 0.000158 - momentum: 0.000000
2023-10-08 20:22:34,693 epoch 2 - iter 24/121 - loss 1.68290825 - time (sec): 16.68 - samples/sec: 279.32 - lr: 0.000157 - momentum: 0.000000
2023-10-08 20:22:43,634 epoch 2 - iter 36/121 - loss 1.57084579 - time (sec): 25.62 - samples/sec: 282.82 - lr: 0.000155 - momentum: 0.000000
2023-10-08 20:22:52,286 epoch 2 - iter 48/121 - loss 1.47431772 - time (sec): 34.27 - samples/sec: 282.90 - lr: 0.000153 - momentum: 0.000000
2023-10-08 20:23:01,117 epoch 2 - iter 60/121 - loss 1.37285982 - time (sec): 43.10 - samples/sec: 281.60 - lr: 0.000151 - momentum: 0.000000
2023-10-08 20:23:09,815 epoch 2 - iter 72/121 - loss 1.28865549 - time (sec): 51.80 - samples/sec: 283.62 - lr: 0.000150 - momentum: 0.000000
2023-10-08 20:23:18,306 epoch 2 - iter 84/121 - loss 1.21968032 - time (sec): 60.29 - samples/sec: 283.73 - lr: 0.000148 - momentum: 0.000000
2023-10-08 20:23:26,781 epoch 2 - iter 96/121 - loss 1.15626234 - time (sec): 68.76 - samples/sec: 283.85 - lr: 0.000146 - momentum: 0.000000
2023-10-08 20:23:34,956 epoch 2 - iter 108/121 - loss 1.10820354 - time (sec): 76.94 - samples/sec: 282.68 - lr: 0.000144 - momentum: 0.000000
2023-10-08 20:23:44,146 epoch 2 - iter 120/121 - loss 1.04246135 - time (sec): 86.13 - samples/sec: 284.29 - lr: 0.000143 - momentum: 0.000000
2023-10-08 20:23:44,922 ----------------------------------------------------------------------------------------------------
2023-10-08 20:23:44,923 EPOCH 2 done: loss 1.0365 - lr: 0.000143
2023-10-08 20:23:50,689 DEV : loss 0.643786609172821 - f1-score (micro avg) 0.0
2023-10-08 20:23:50,695 ----------------------------------------------------------------------------------------------------
2023-10-08 20:23:59,765 epoch 3 - iter 12/121 - loss 0.55356155 - time (sec): 9.07 - samples/sec: 293.97 - lr: 0.000141 - momentum: 0.000000
2023-10-08 20:24:08,933 epoch 3 - iter 24/121 - loss 0.58950453 - time (sec): 18.24 - samples/sec: 292.49 - lr: 0.000139 - momentum: 0.000000
2023-10-08 20:24:17,647 epoch 3 - iter 36/121 - loss 0.58113678 - time (sec): 26.95 - samples/sec: 289.19 - lr: 0.000137 - momentum: 0.000000
2023-10-08 20:24:25,802 epoch 3 - iter 48/121 - loss 0.57384358 - time (sec): 35.11 - samples/sec: 285.06 - lr: 0.000135 - momentum: 0.000000
2023-10-08 20:24:34,033 epoch 3 - iter 60/121 - loss 0.54981326 - time (sec): 43.34 - samples/sec: 285.46 - lr: 0.000134 - momentum: 0.000000
2023-10-08 20:24:41,969 epoch 3 - iter 72/121 - loss 0.53976976 - time (sec): 51.27 - samples/sec: 283.15 - lr: 0.000132 - momentum: 0.000000
2023-10-08 20:24:50,710 epoch 3 - iter 84/121 - loss 0.52639915 - time (sec): 60.01 - samples/sec: 283.82 - lr: 0.000130 - momentum: 0.000000
2023-10-08 20:25:00,185 epoch 3 - iter 96/121 - loss 0.49739756 - time (sec): 69.49 - samples/sec: 284.58 - lr: 0.000128 - momentum: 0.000000
2023-10-08 20:25:08,614 epoch 3 - iter 108/121 - loss 0.48425359 - time (sec): 77.92 - samples/sec: 284.54 - lr: 0.000127 - momentum: 0.000000
2023-10-08 20:25:17,227 epoch 3 - iter 120/121 - loss 0.47509796 - time (sec): 86.53 - samples/sec: 284.26 - lr: 0.000125 - momentum: 0.000000
2023-10-08 20:25:17,777 ----------------------------------------------------------------------------------------------------
2023-10-08 20:25:17,777 EPOCH 3 done: loss 0.4748 - lr: 0.000125
2023-10-08 20:25:23,591 DEV : loss 0.36890920996665955 - f1-score (micro avg) 0.1559
2023-10-08 20:25:23,597 saving best model
2023-10-08 20:25:24,450 ----------------------------------------------------------------------------------------------------
2023-10-08 20:25:33,765 epoch 4 - iter 12/121 - loss 0.27461403 - time (sec): 9.31 - samples/sec: 285.27 - lr: 0.000123 - momentum: 0.000000
2023-10-08 20:25:42,894 epoch 4 - iter 24/121 - loss 0.27755240 - time (sec): 18.44 - samples/sec: 285.97 - lr: 0.000121 - momentum: 0.000000
2023-10-08 20:25:51,154 epoch 4 - iter 36/121 - loss 0.29195280 - time (sec): 26.70 - samples/sec: 283.60 - lr: 0.000120 - momentum: 0.000000
2023-10-08 20:25:59,533 epoch 4 - iter 48/121 - loss 0.29836698 - time (sec): 35.08 - samples/sec: 280.65 - lr: 0.000118 - momentum: 0.000000
2023-10-08 20:26:07,978 epoch 4 - iter 60/121 - loss 0.30533029 - time (sec): 43.53 - samples/sec: 280.70 - lr: 0.000116 - momentum: 0.000000
2023-10-08 20:26:16,709 epoch 4 - iter 72/121 - loss 0.30075527 - time (sec): 52.26 - samples/sec: 282.52 - lr: 0.000114 - momentum: 0.000000
2023-10-08 20:26:25,827 epoch 4 - iter 84/121 - loss 0.30211257 - time (sec): 61.38 - samples/sec: 283.06 - lr: 0.000113 - momentum: 0.000000
2023-10-08 20:26:34,685 epoch 4 - iter 96/121 - loss 0.29564739 - time (sec): 70.23 - samples/sec: 283.31 - lr: 0.000111 - momentum: 0.000000
2023-10-08 20:26:43,003 epoch 4 - iter 108/121 - loss 0.29558956 - time (sec): 78.55 - samples/sec: 281.66 - lr: 0.000109 - momentum: 0.000000
2023-10-08 20:26:51,959 epoch 4 - iter 120/121 - loss 0.29013761 - time (sec): 87.51 - samples/sec: 281.39 - lr: 0.000107 - momentum: 0.000000
2023-10-08 20:26:52,417 ----------------------------------------------------------------------------------------------------
2023-10-08 20:26:52,417 EPOCH 4 done: loss 0.2902 - lr: 0.000107
2023-10-08 20:26:58,347 DEV : loss 0.249277263879776 - f1-score (micro avg) 0.5452
2023-10-08 20:26:58,353 saving best model
2023-10-08 20:27:02,710 ----------------------------------------------------------------------------------------------------
2023-10-08 20:27:11,255 epoch 5 - iter 12/121 - loss 0.27928687 - time (sec): 8.54 - samples/sec: 271.66 - lr: 0.000105 - momentum: 0.000000
2023-10-08 20:27:20,445 epoch 5 - iter 24/121 - loss 0.25512842 - time (sec): 17.73 - samples/sec: 283.64 - lr: 0.000104 - momentum: 0.000000
2023-10-08 20:27:28,658 epoch 5 - iter 36/121 - loss 0.23515344 - time (sec): 25.95 - samples/sec: 279.92 - lr: 0.000102 - momentum: 0.000000
2023-10-08 20:27:37,994 epoch 5 - iter 48/121 - loss 0.22735843 - time (sec): 35.28 - samples/sec: 282.23 - lr: 0.000100 - momentum: 0.000000
2023-10-08 20:27:46,907 epoch 5 - iter 60/121 - loss 0.22192912 - time (sec): 44.20 - samples/sec: 278.96 - lr: 0.000098 - momentum: 0.000000
2023-10-08 20:27:56,382 epoch 5 - iter 72/121 - loss 0.20974637 - time (sec): 53.67 - samples/sec: 279.54 - lr: 0.000097 - momentum: 0.000000
2023-10-08 20:28:05,425 epoch 5 - iter 84/121 - loss 0.21050078 - time (sec): 62.71 - samples/sec: 278.89 - lr: 0.000095 - momentum: 0.000000
2023-10-08 20:28:14,253 epoch 5 - iter 96/121 - loss 0.20891343 - time (sec): 71.54 - samples/sec: 276.93 - lr: 0.000093 - momentum: 0.000000
2023-10-08 20:28:22,864 epoch 5 - iter 108/121 - loss 0.20885785 - time (sec): 80.15 - samples/sec: 275.14 - lr: 0.000091 - momentum: 0.000000
2023-10-08 20:28:32,320 epoch 5 - iter 120/121 - loss 0.20712592 - time (sec): 89.61 - samples/sec: 274.96 - lr: 0.000090 - momentum: 0.000000
2023-10-08 20:28:32,796 ----------------------------------------------------------------------------------------------------
2023-10-08 20:28:32,796 EPOCH 5 done: loss 0.2071 - lr: 0.000090
2023-10-08 20:28:39,095 DEV : loss 0.19593805074691772 - f1-score (micro avg) 0.6249
2023-10-08 20:28:39,101 saving best model
2023-10-08 20:28:43,363 ----------------------------------------------------------------------------------------------------
2023-10-08 20:28:52,053 epoch 6 - iter 12/121 - loss 0.14124031 - time (sec): 8.69 - samples/sec: 257.12 - lr: 0.000088 - momentum: 0.000000
2023-10-08 20:29:01,312 epoch 6 - iter 24/121 - loss 0.16743244 - time (sec): 17.95 - samples/sec: 260.21 - lr: 0.000086 - momentum: 0.000000
2023-10-08 20:29:10,839 epoch 6 - iter 36/121 - loss 0.17141953 - time (sec): 27.47 - samples/sec: 263.19 - lr: 0.000084 - momentum: 0.000000
2023-10-08 20:29:19,263 epoch 6 - iter 48/121 - loss 0.16587158 - time (sec): 35.90 - samples/sec: 261.46 - lr: 0.000082 - momentum: 0.000000
2023-10-08 20:29:29,061 epoch 6 - iter 60/121 - loss 0.16828021 - time (sec): 45.70 - samples/sec: 261.95 - lr: 0.000081 - momentum: 0.000000
2023-10-08 20:29:37,873 epoch 6 - iter 72/121 - loss 0.16848524 - time (sec): 54.51 - samples/sec: 261.28 - lr: 0.000079 - momentum: 0.000000
2023-10-08 20:29:47,353 epoch 6 - iter 84/121 - loss 0.16521968 - time (sec): 63.99 - samples/sec: 261.70 - lr: 0.000077 - momentum: 0.000000
2023-10-08 20:29:57,527 epoch 6 - iter 96/121 - loss 0.16409736 - time (sec): 74.16 - samples/sec: 263.40 - lr: 0.000075 - momentum: 0.000000
2023-10-08 20:30:06,879 epoch 6 - iter 108/121 - loss 0.15927078 - time (sec): 83.51 - samples/sec: 263.15 - lr: 0.000074 - momentum: 0.000000
2023-10-08 20:30:16,701 epoch 6 - iter 120/121 - loss 0.15856885 - time (sec): 93.34 - samples/sec: 263.70 - lr: 0.000072 - momentum: 0.000000
2023-10-08 20:30:17,259 ----------------------------------------------------------------------------------------------------
2023-10-08 20:30:17,260 EPOCH 6 done: loss 0.1587 - lr: 0.000072
2023-10-08 20:30:23,720 DEV : loss 0.16929559409618378 - f1-score (micro avg) 0.808
2023-10-08 20:30:23,726 saving best model
2023-10-08 20:30:28,106 ----------------------------------------------------------------------------------------------------
2023-10-08 20:30:37,857 epoch 7 - iter 12/121 - loss 0.12944446 - time (sec): 9.75 - samples/sec: 272.22 - lr: 0.000070 - momentum: 0.000000
2023-10-08 20:30:47,340 epoch 7 - iter 24/121 - loss 0.14141500 - time (sec): 19.23 - samples/sec: 274.63 - lr: 0.000068 - momentum: 0.000000
2023-10-08 20:30:57,111 epoch 7 - iter 36/121 - loss 0.14016248 - time (sec): 29.00 - samples/sec: 271.62 - lr: 0.000066 - momentum: 0.000000
2023-10-08 20:31:06,846 epoch 7 - iter 48/121 - loss 0.13853899 - time (sec): 38.74 - samples/sec: 272.10 - lr: 0.000065 - momentum: 0.000000
2023-10-08 20:31:15,451 epoch 7 - iter 60/121 - loss 0.13649677 - time (sec): 47.34 - samples/sec: 271.21 - lr: 0.000063 - momentum: 0.000000
2023-10-08 20:31:25,285 epoch 7 - iter 72/121 - loss 0.14153892 - time (sec): 57.18 - samples/sec: 272.03 - lr: 0.000061 - momentum: 0.000000
2023-10-08 20:31:34,769 epoch 7 - iter 84/121 - loss 0.13594019 - time (sec): 66.66 - samples/sec: 270.82 - lr: 0.000059 - momentum: 0.000000
2023-10-08 20:31:43,353 epoch 7 - iter 96/121 - loss 0.13016134 - time (sec): 75.25 - samples/sec: 268.43 - lr: 0.000058 - momentum: 0.000000
2023-10-08 20:31:52,749 epoch 7 - iter 108/121 - loss 0.12900492 - time (sec): 84.64 - samples/sec: 267.03 - lr: 0.000056 - momentum: 0.000000
2023-10-08 20:32:01,387 epoch 7 - iter 120/121 - loss 0.12699129 - time (sec): 93.28 - samples/sec: 264.00 - lr: 0.000054 - momentum: 0.000000
2023-10-08 20:32:01,886 ----------------------------------------------------------------------------------------------------
2023-10-08 20:32:01,887 EPOCH 7 done: loss 0.1270 - lr: 0.000054
2023-10-08 20:32:08,451 DEV : loss 0.1514769345521927 - f1-score (micro avg) 0.7952
2023-10-08 20:32:08,457 ----------------------------------------------------------------------------------------------------
2023-10-08 20:32:17,430 epoch 8 - iter 12/121 - loss 0.11070488 - time (sec): 8.97 - samples/sec: 257.58 - lr: 0.000052 - momentum: 0.000000
2023-10-08 20:32:26,294 epoch 8 - iter 24/121 - loss 0.11582423 - time (sec): 17.84 - samples/sec: 259.71 - lr: 0.000051 - momentum: 0.000000
2023-10-08 20:32:35,894 epoch 8 - iter 36/121 - loss 0.11715067 - time (sec): 27.44 - samples/sec: 263.49 - lr: 0.000049 - momentum: 0.000000
2023-10-08 20:32:45,372 epoch 8 - iter 48/121 - loss 0.11666868 - time (sec): 36.91 - samples/sec: 263.56 - lr: 0.000047 - momentum: 0.000000
2023-10-08 20:32:55,071 epoch 8 - iter 60/121 - loss 0.10825251 - time (sec): 46.61 - samples/sec: 263.79 - lr: 0.000045 - momentum: 0.000000
2023-10-08 20:33:03,724 epoch 8 - iter 72/121 - loss 0.11057782 - time (sec): 55.27 - samples/sec: 261.83 - lr: 0.000044 - momentum: 0.000000
2023-10-08 20:33:13,238 epoch 8 - iter 84/121 - loss 0.10997043 - time (sec): 64.78 - samples/sec: 262.06 - lr: 0.000042 - momentum: 0.000000
2023-10-08 20:33:23,237 epoch 8 - iter 96/121 - loss 0.11055068 - time (sec): 74.78 - samples/sec: 263.56 - lr: 0.000040 - momentum: 0.000000
2023-10-08 20:33:32,720 epoch 8 - iter 108/121 - loss 0.11076208 - time (sec): 84.26 - samples/sec: 263.12 - lr: 0.000038 - momentum: 0.000000
2023-10-08 20:33:41,990 epoch 8 - iter 120/121 - loss 0.10742794 - time (sec): 93.53 - samples/sec: 262.05 - lr: 0.000037 - momentum: 0.000000
2023-10-08 20:33:42,773 ----------------------------------------------------------------------------------------------------
2023-10-08 20:33:42,774 EPOCH 8 done: loss 0.1071 - lr: 0.000037
2023-10-08 20:33:49,188 DEV : loss 0.147888645529747 - f1-score (micro avg) 0.8069
2023-10-08 20:33:49,194 ----------------------------------------------------------------------------------------------------
2023-10-08 20:33:58,303 epoch 9 - iter 12/121 - loss 0.11454532 - time (sec): 9.11 - samples/sec: 265.49 - lr: 0.000035 - momentum: 0.000000
2023-10-08 20:34:07,677 epoch 9 - iter 24/121 - loss 0.10476188 - time (sec): 18.48 - samples/sec: 263.89 - lr: 0.000033 - momentum: 0.000000
2023-10-08 20:34:17,804 epoch 9 - iter 36/121 - loss 0.09375290 - time (sec): 28.61 - samples/sec: 268.21 - lr: 0.000031 - momentum: 0.000000
2023-10-08 20:34:27,231 epoch 9 - iter 48/121 - loss 0.09265415 - time (sec): 38.04 - samples/sec: 267.04 - lr: 0.000029 - momentum: 0.000000
2023-10-08 20:34:35,949 epoch 9 - iter 60/121 - loss 0.08753107 - time (sec): 46.75 - samples/sec: 265.31 - lr: 0.000028 - momentum: 0.000000
2023-10-08 20:34:45,231 epoch 9 - iter 72/121 - loss 0.08796092 - time (sec): 56.04 - samples/sec: 263.85 - lr: 0.000026 - momentum: 0.000000
2023-10-08 20:34:54,799 epoch 9 - iter 84/121 - loss 0.08585326 - time (sec): 65.60 - samples/sec: 262.63 - lr: 0.000024 - momentum: 0.000000
2023-10-08 20:35:03,974 epoch 9 - iter 96/121 - loss 0.08754369 - time (sec): 74.78 - samples/sec: 262.20 - lr: 0.000022 - momentum: 0.000000
2023-10-08 20:35:13,424 epoch 9 - iter 108/121 - loss 0.09096343 - time (sec): 84.23 - samples/sec: 262.17 - lr: 0.000021 - momentum: 0.000000
2023-10-08 20:35:23,051 epoch 9 - iter 120/121 - loss 0.09238782 - time (sec): 93.86 - samples/sec: 262.33 - lr: 0.000019 - momentum: 0.000000
2023-10-08 20:35:23,549 ----------------------------------------------------------------------------------------------------
2023-10-08 20:35:23,549 EPOCH 9 done: loss 0.0920 - lr: 0.000019
2023-10-08 20:35:30,153 DEV : loss 0.14555731415748596 - f1-score (micro avg) 0.8155
2023-10-08 20:35:30,159 saving best model
2023-10-08 20:35:34,523 ----------------------------------------------------------------------------------------------------
2023-10-08 20:35:44,235 epoch 10 - iter 12/121 - loss 0.08735455 - time (sec): 9.71 - samples/sec: 273.33 - lr: 0.000017 - momentum: 0.000000
2023-10-08 20:35:53,165 epoch 10 - iter 24/121 - loss 0.07783909 - time (sec): 18.64 - samples/sec: 268.29 - lr: 0.000015 - momentum: 0.000000
2023-10-08 20:36:01,952 epoch 10 - iter 36/121 - loss 0.07594591 - time (sec): 27.43 - samples/sec: 266.49 - lr: 0.000013 - momentum: 0.000000
2023-10-08 20:36:11,081 epoch 10 - iter 48/121 - loss 0.08000383 - time (sec): 36.56 - samples/sec: 259.32 - lr: 0.000012 - momentum: 0.000000
2023-10-08 20:36:20,688 epoch 10 - iter 60/121 - loss 0.07915962 - time (sec): 46.16 - samples/sec: 260.58 - lr: 0.000010 - momentum: 0.000000
2023-10-08 20:36:29,307 epoch 10 - iter 72/121 - loss 0.07897937 - time (sec): 54.78 - samples/sec: 260.01 - lr: 0.000008 - momentum: 0.000000
2023-10-08 20:36:38,576 epoch 10 - iter 84/121 - loss 0.07708388 - time (sec): 64.05 - samples/sec: 259.85 - lr: 0.000006 - momentum: 0.000000
2023-10-08 20:36:48,548 epoch 10 - iter 96/121 - loss 0.07938083 - time (sec): 74.02 - samples/sec: 260.65 - lr: 0.000005 - momentum: 0.000000
2023-10-08 20:36:58,737 epoch 10 - iter 108/121 - loss 0.08403503 - time (sec): 84.21 - samples/sec: 262.05 - lr: 0.000003 - momentum: 0.000000
2023-10-08 20:37:08,377 epoch 10 - iter 120/121 - loss 0.08467019 - time (sec): 93.85 - samples/sec: 261.21 - lr: 0.000001 - momentum: 0.000000
2023-10-08 20:37:09,144 ----------------------------------------------------------------------------------------------------
2023-10-08 20:37:09,144 EPOCH 10 done: loss 0.0845 - lr: 0.000001
2023-10-08 20:37:15,607 DEV : loss 0.1435980498790741 - f1-score (micro avg) 0.8144
2023-10-08 20:37:16,487 ----------------------------------------------------------------------------------------------------
2023-10-08 20:37:16,489 Loading model from best epoch ...
2023-10-08 20:37:19,070 SequenceTagger predicts: Dictionary with 25 tags: O, S-scope, B-scope, E-scope, I-scope, S-pers, B-pers, E-pers, I-pers, S-work, B-work, E-work, I-work, S-loc, B-loc, E-loc, I-loc, S-object, B-object, E-object, I-object, S-date, B-date, E-date, I-date
2023-10-08 20:37:25,596
Results:
- F-score (micro) 0.8022
- F-score (macro) 0.4828
- Accuracy 0.6995
By class:
precision recall f1-score support
pers 0.8252 0.8489 0.8369 139
scope 0.7905 0.9070 0.8448 129
work 0.6848 0.7875 0.7326 80
loc 0.0000 0.0000 0.0000 9
date 0.0000 0.0000 0.0000 3
micro avg 0.7781 0.8278 0.8022 360
macro avg 0.4601 0.5087 0.4828 360
weighted avg 0.7541 0.8278 0.7886 360
2023-10-08 20:37:25,597 ----------------------------------------------------------------------------------------------------
|