File size: 24,948 Bytes
9ebd31b |
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 |
2023-10-08 21:31:27,703 ----------------------------------------------------------------------------------------------------
2023-10-08 21:31:27,704 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 21:31:27,704 ----------------------------------------------------------------------------------------------------
2023-10-08 21:31:27,704 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 21:31:27,704 ----------------------------------------------------------------------------------------------------
2023-10-08 21:31:27,704 Train: 966 sentences
2023-10-08 21:31:27,704 (train_with_dev=False, train_with_test=False)
2023-10-08 21:31:27,705 ----------------------------------------------------------------------------------------------------
2023-10-08 21:31:27,705 Training Params:
2023-10-08 21:31:27,705 - learning_rate: "0.00016"
2023-10-08 21:31:27,705 - mini_batch_size: "8"
2023-10-08 21:31:27,705 - max_epochs: "10"
2023-10-08 21:31:27,705 - shuffle: "True"
2023-10-08 21:31:27,705 ----------------------------------------------------------------------------------------------------
2023-10-08 21:31:27,705 Plugins:
2023-10-08 21:31:27,705 - TensorboardLogger
2023-10-08 21:31:27,705 - LinearScheduler | warmup_fraction: '0.1'
2023-10-08 21:31:27,705 ----------------------------------------------------------------------------------------------------
2023-10-08 21:31:27,705 Final evaluation on model from best epoch (best-model.pt)
2023-10-08 21:31:27,705 - metric: "('micro avg', 'f1-score')"
2023-10-08 21:31:27,705 ----------------------------------------------------------------------------------------------------
2023-10-08 21:31:27,705 Computation:
2023-10-08 21:31:27,705 - compute on device: cuda:0
2023-10-08 21:31:27,705 - embedding storage: none
2023-10-08 21:31:27,705 ----------------------------------------------------------------------------------------------------
2023-10-08 21:31:27,705 Model training base path: "hmbench-ajmc/fr-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs8-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-3"
2023-10-08 21:31:27,705 ----------------------------------------------------------------------------------------------------
2023-10-08 21:31:27,706 ----------------------------------------------------------------------------------------------------
2023-10-08 21:31:27,706 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-08 21:31:35,621 epoch 1 - iter 12/121 - loss 3.21166448 - time (sec): 7.91 - samples/sec: 280.39 - lr: 0.000015 - momentum: 0.000000
2023-10-08 21:31:44,349 epoch 1 - iter 24/121 - loss 3.20556874 - time (sec): 16.64 - samples/sec: 286.13 - lr: 0.000030 - momentum: 0.000000
2023-10-08 21:31:53,215 epoch 1 - iter 36/121 - loss 3.19507718 - time (sec): 25.51 - samples/sec: 287.67 - lr: 0.000046 - momentum: 0.000000
2023-10-08 21:32:01,324 epoch 1 - iter 48/121 - loss 3.18016809 - time (sec): 33.62 - samples/sec: 285.30 - lr: 0.000062 - momentum: 0.000000
2023-10-08 21:32:09,771 epoch 1 - iter 60/121 - loss 3.14881989 - time (sec): 42.06 - samples/sec: 287.18 - lr: 0.000078 - momentum: 0.000000
2023-10-08 21:32:18,794 epoch 1 - iter 72/121 - loss 3.09406347 - time (sec): 51.09 - samples/sec: 289.21 - lr: 0.000094 - momentum: 0.000000
2023-10-08 21:32:27,531 epoch 1 - iter 84/121 - loss 3.03193705 - time (sec): 59.82 - samples/sec: 287.18 - lr: 0.000110 - momentum: 0.000000
2023-10-08 21:32:36,394 epoch 1 - iter 96/121 - loss 2.95622833 - time (sec): 68.69 - samples/sec: 286.31 - lr: 0.000126 - momentum: 0.000000
2023-10-08 21:32:45,229 epoch 1 - iter 108/121 - loss 2.87141440 - time (sec): 77.52 - samples/sec: 286.45 - lr: 0.000141 - momentum: 0.000000
2023-10-08 21:32:53,831 epoch 1 - iter 120/121 - loss 2.79080138 - time (sec): 86.12 - samples/sec: 284.70 - lr: 0.000157 - momentum: 0.000000
2023-10-08 21:32:54,609 ----------------------------------------------------------------------------------------------------
2023-10-08 21:32:54,609 EPOCH 1 done: loss 2.7821 - lr: 0.000157
2023-10-08 21:33:00,387 DEV : loss 1.8131030797958374 - f1-score (micro avg) 0.0
2023-10-08 21:33:00,393 ----------------------------------------------------------------------------------------------------
2023-10-08 21:33:09,217 epoch 2 - iter 12/121 - loss 1.79636265 - time (sec): 8.82 - samples/sec: 293.44 - lr: 0.000158 - momentum: 0.000000
2023-10-08 21:33:18,393 epoch 2 - iter 24/121 - loss 1.68244545 - time (sec): 18.00 - samples/sec: 294.25 - lr: 0.000157 - momentum: 0.000000
2023-10-08 21:33:26,790 epoch 2 - iter 36/121 - loss 1.57537534 - time (sec): 26.40 - samples/sec: 288.80 - lr: 0.000155 - momentum: 0.000000
2023-10-08 21:33:35,523 epoch 2 - iter 48/121 - loss 1.47842825 - time (sec): 35.13 - samples/sec: 284.35 - lr: 0.000153 - momentum: 0.000000
2023-10-08 21:33:44,473 epoch 2 - iter 60/121 - loss 1.38341861 - time (sec): 44.08 - samples/sec: 281.58 - lr: 0.000151 - momentum: 0.000000
2023-10-08 21:33:53,695 epoch 2 - iter 72/121 - loss 1.30268725 - time (sec): 53.30 - samples/sec: 278.98 - lr: 0.000150 - momentum: 0.000000
2023-10-08 21:34:02,865 epoch 2 - iter 84/121 - loss 1.23521232 - time (sec): 62.47 - samples/sec: 279.01 - lr: 0.000148 - momentum: 0.000000
2023-10-08 21:34:11,289 epoch 2 - iter 96/121 - loss 1.17983078 - time (sec): 70.89 - samples/sec: 277.88 - lr: 0.000146 - momentum: 0.000000
2023-10-08 21:34:19,830 epoch 2 - iter 108/121 - loss 1.11490372 - time (sec): 79.44 - samples/sec: 277.56 - lr: 0.000144 - momentum: 0.000000
2023-10-08 21:34:28,449 epoch 2 - iter 120/121 - loss 1.06053368 - time (sec): 88.06 - samples/sec: 278.54 - lr: 0.000143 - momentum: 0.000000
2023-10-08 21:34:29,093 ----------------------------------------------------------------------------------------------------
2023-10-08 21:34:29,094 EPOCH 2 done: loss 1.0556 - lr: 0.000143
2023-10-08 21:34:34,906 DEV : loss 0.6444090008735657 - f1-score (micro avg) 0.0
2023-10-08 21:34:34,912 ----------------------------------------------------------------------------------------------------
2023-10-08 21:34:43,184 epoch 3 - iter 12/121 - loss 0.58722550 - time (sec): 8.27 - samples/sec: 275.54 - lr: 0.000141 - momentum: 0.000000
2023-10-08 21:34:52,072 epoch 3 - iter 24/121 - loss 0.59417629 - time (sec): 17.16 - samples/sec: 284.17 - lr: 0.000139 - momentum: 0.000000
2023-10-08 21:35:00,791 epoch 3 - iter 36/121 - loss 0.59259042 - time (sec): 25.88 - samples/sec: 282.60 - lr: 0.000137 - momentum: 0.000000
2023-10-08 21:35:09,468 epoch 3 - iter 48/121 - loss 0.58407643 - time (sec): 34.55 - samples/sec: 283.23 - lr: 0.000135 - momentum: 0.000000
2023-10-08 21:35:18,218 epoch 3 - iter 60/121 - loss 0.57152014 - time (sec): 43.30 - samples/sec: 285.37 - lr: 0.000134 - momentum: 0.000000
2023-10-08 21:35:26,404 epoch 3 - iter 72/121 - loss 0.54715064 - time (sec): 51.49 - samples/sec: 283.76 - lr: 0.000132 - momentum: 0.000000
2023-10-08 21:35:34,981 epoch 3 - iter 84/121 - loss 0.53051696 - time (sec): 60.07 - samples/sec: 284.28 - lr: 0.000130 - momentum: 0.000000
2023-10-08 21:35:43,807 epoch 3 - iter 96/121 - loss 0.51262726 - time (sec): 68.89 - samples/sec: 283.65 - lr: 0.000128 - momentum: 0.000000
2023-10-08 21:35:52,760 epoch 3 - iter 108/121 - loss 0.48852958 - time (sec): 77.85 - samples/sec: 285.76 - lr: 0.000127 - momentum: 0.000000
2023-10-08 21:36:01,236 epoch 3 - iter 120/121 - loss 0.47092846 - time (sec): 86.32 - samples/sec: 284.54 - lr: 0.000125 - momentum: 0.000000
2023-10-08 21:36:01,800 ----------------------------------------------------------------------------------------------------
2023-10-08 21:36:01,801 EPOCH 3 done: loss 0.4715 - lr: 0.000125
2023-10-08 21:36:07,771 DEV : loss 0.3705194294452667 - f1-score (micro avg) 0.4064
2023-10-08 21:36:07,777 saving best model
2023-10-08 21:36:08,922 ----------------------------------------------------------------------------------------------------
2023-10-08 21:36:17,557 epoch 4 - iter 12/121 - loss 0.30532656 - time (sec): 8.63 - samples/sec: 260.10 - lr: 0.000123 - momentum: 0.000000
2023-10-08 21:36:26,148 epoch 4 - iter 24/121 - loss 0.31458394 - time (sec): 17.22 - samples/sec: 273.89 - lr: 0.000121 - momentum: 0.000000
2023-10-08 21:36:34,016 epoch 4 - iter 36/121 - loss 0.31803950 - time (sec): 25.09 - samples/sec: 272.41 - lr: 0.000120 - momentum: 0.000000
2023-10-08 21:36:42,364 epoch 4 - iter 48/121 - loss 0.31800584 - time (sec): 33.44 - samples/sec: 276.21 - lr: 0.000118 - momentum: 0.000000
2023-10-08 21:36:51,364 epoch 4 - iter 60/121 - loss 0.29918326 - time (sec): 42.44 - samples/sec: 280.69 - lr: 0.000116 - momentum: 0.000000
2023-10-08 21:37:00,474 epoch 4 - iter 72/121 - loss 0.30078693 - time (sec): 51.55 - samples/sec: 282.22 - lr: 0.000114 - momentum: 0.000000
2023-10-08 21:37:08,744 epoch 4 - iter 84/121 - loss 0.30342155 - time (sec): 59.82 - samples/sec: 282.75 - lr: 0.000113 - momentum: 0.000000
2023-10-08 21:37:17,956 epoch 4 - iter 96/121 - loss 0.29833815 - time (sec): 69.03 - samples/sec: 285.24 - lr: 0.000111 - momentum: 0.000000
2023-10-08 21:37:26,364 epoch 4 - iter 108/121 - loss 0.29213228 - time (sec): 77.44 - samples/sec: 285.53 - lr: 0.000109 - momentum: 0.000000
2023-10-08 21:37:35,161 epoch 4 - iter 120/121 - loss 0.28600872 - time (sec): 86.24 - samples/sec: 285.21 - lr: 0.000107 - momentum: 0.000000
2023-10-08 21:37:35,702 ----------------------------------------------------------------------------------------------------
2023-10-08 21:37:35,703 EPOCH 4 done: loss 0.2864 - lr: 0.000107
2023-10-08 21:37:41,512 DEV : loss 0.26601657271385193 - f1-score (micro avg) 0.5204
2023-10-08 21:37:41,517 saving best model
2023-10-08 21:37:45,886 ----------------------------------------------------------------------------------------------------
2023-10-08 21:37:54,116 epoch 5 - iter 12/121 - loss 0.23432641 - time (sec): 8.23 - samples/sec: 277.22 - lr: 0.000105 - momentum: 0.000000
2023-10-08 21:38:03,245 epoch 5 - iter 24/121 - loss 0.23159006 - time (sec): 17.36 - samples/sec: 278.73 - lr: 0.000104 - momentum: 0.000000
2023-10-08 21:38:12,229 epoch 5 - iter 36/121 - loss 0.23704500 - time (sec): 26.34 - samples/sec: 285.67 - lr: 0.000102 - momentum: 0.000000
2023-10-08 21:38:20,653 epoch 5 - iter 48/121 - loss 0.24078930 - time (sec): 34.77 - samples/sec: 285.26 - lr: 0.000100 - momentum: 0.000000
2023-10-08 21:38:29,660 epoch 5 - iter 60/121 - loss 0.23998293 - time (sec): 43.77 - samples/sec: 284.75 - lr: 0.000098 - momentum: 0.000000
2023-10-08 21:38:38,636 epoch 5 - iter 72/121 - loss 0.23089916 - time (sec): 52.75 - samples/sec: 283.78 - lr: 0.000097 - momentum: 0.000000
2023-10-08 21:38:47,327 epoch 5 - iter 84/121 - loss 0.22345391 - time (sec): 61.44 - samples/sec: 281.73 - lr: 0.000095 - momentum: 0.000000
2023-10-08 21:38:56,444 epoch 5 - iter 96/121 - loss 0.21598786 - time (sec): 70.56 - samples/sec: 284.27 - lr: 0.000093 - momentum: 0.000000
2023-10-08 21:39:04,708 epoch 5 - iter 108/121 - loss 0.21161077 - time (sec): 78.82 - samples/sec: 282.72 - lr: 0.000091 - momentum: 0.000000
2023-10-08 21:39:12,919 epoch 5 - iter 120/121 - loss 0.20909319 - time (sec): 87.03 - samples/sec: 282.15 - lr: 0.000090 - momentum: 0.000000
2023-10-08 21:39:13,508 ----------------------------------------------------------------------------------------------------
2023-10-08 21:39:13,508 EPOCH 5 done: loss 0.2096 - lr: 0.000090
2023-10-08 21:39:19,503 DEV : loss 0.20941473543643951 - f1-score (micro avg) 0.6484
2023-10-08 21:39:19,508 saving best model
2023-10-08 21:39:23,895 ----------------------------------------------------------------------------------------------------
2023-10-08 21:39:33,041 epoch 6 - iter 12/121 - loss 0.18803803 - time (sec): 9.14 - samples/sec: 288.36 - lr: 0.000088 - momentum: 0.000000
2023-10-08 21:39:41,948 epoch 6 - iter 24/121 - loss 0.17673119 - time (sec): 18.05 - samples/sec: 283.58 - lr: 0.000086 - momentum: 0.000000
2023-10-08 21:39:50,405 epoch 6 - iter 36/121 - loss 0.17204922 - time (sec): 26.51 - samples/sec: 280.70 - lr: 0.000084 - momentum: 0.000000
2023-10-08 21:39:58,836 epoch 6 - iter 48/121 - loss 0.17883918 - time (sec): 34.94 - samples/sec: 274.47 - lr: 0.000082 - momentum: 0.000000
2023-10-08 21:40:07,420 epoch 6 - iter 60/121 - loss 0.17037530 - time (sec): 43.52 - samples/sec: 272.75 - lr: 0.000081 - momentum: 0.000000
2023-10-08 21:40:16,341 epoch 6 - iter 72/121 - loss 0.16322801 - time (sec): 52.44 - samples/sec: 273.36 - lr: 0.000079 - momentum: 0.000000
2023-10-08 21:40:25,403 epoch 6 - iter 84/121 - loss 0.16137249 - time (sec): 61.51 - samples/sec: 274.75 - lr: 0.000077 - momentum: 0.000000
2023-10-08 21:40:34,928 epoch 6 - iter 96/121 - loss 0.15730659 - time (sec): 71.03 - samples/sec: 275.47 - lr: 0.000075 - momentum: 0.000000
2023-10-08 21:40:44,096 epoch 6 - iter 108/121 - loss 0.15850412 - time (sec): 80.20 - samples/sec: 275.17 - lr: 0.000074 - momentum: 0.000000
2023-10-08 21:40:53,598 epoch 6 - iter 120/121 - loss 0.15656703 - time (sec): 89.70 - samples/sec: 273.95 - lr: 0.000072 - momentum: 0.000000
2023-10-08 21:40:54,256 ----------------------------------------------------------------------------------------------------
2023-10-08 21:40:54,257 EPOCH 6 done: loss 0.1561 - lr: 0.000072
2023-10-08 21:41:00,597 DEV : loss 0.17296475172042847 - f1-score (micro avg) 0.8248
2023-10-08 21:41:00,603 saving best model
2023-10-08 21:41:04,984 ----------------------------------------------------------------------------------------------------
2023-10-08 21:41:13,812 epoch 7 - iter 12/121 - loss 0.13301191 - time (sec): 8.83 - samples/sec: 263.65 - lr: 0.000070 - momentum: 0.000000
2023-10-08 21:41:22,884 epoch 7 - iter 24/121 - loss 0.13488940 - time (sec): 17.90 - samples/sec: 259.57 - lr: 0.000068 - momentum: 0.000000
2023-10-08 21:41:32,271 epoch 7 - iter 36/121 - loss 0.12322713 - time (sec): 27.29 - samples/sec: 264.32 - lr: 0.000066 - momentum: 0.000000
2023-10-08 21:41:41,269 epoch 7 - iter 48/121 - loss 0.12061266 - time (sec): 36.28 - samples/sec: 264.12 - lr: 0.000065 - momentum: 0.000000
2023-10-08 21:41:50,423 epoch 7 - iter 60/121 - loss 0.12133471 - time (sec): 45.44 - samples/sec: 263.50 - lr: 0.000063 - momentum: 0.000000
2023-10-08 21:41:59,980 epoch 7 - iter 72/121 - loss 0.12466745 - time (sec): 54.99 - samples/sec: 264.85 - lr: 0.000061 - momentum: 0.000000
2023-10-08 21:42:09,959 epoch 7 - iter 84/121 - loss 0.12157869 - time (sec): 64.97 - samples/sec: 263.94 - lr: 0.000059 - momentum: 0.000000
2023-10-08 21:42:19,799 epoch 7 - iter 96/121 - loss 0.12255567 - time (sec): 74.81 - samples/sec: 264.31 - lr: 0.000058 - momentum: 0.000000
2023-10-08 21:42:29,242 epoch 7 - iter 108/121 - loss 0.12010957 - time (sec): 84.26 - samples/sec: 264.62 - lr: 0.000056 - momentum: 0.000000
2023-10-08 21:42:38,404 epoch 7 - iter 120/121 - loss 0.11873726 - time (sec): 93.42 - samples/sec: 263.01 - lr: 0.000054 - momentum: 0.000000
2023-10-08 21:42:39,023 ----------------------------------------------------------------------------------------------------
2023-10-08 21:42:39,023 EPOCH 7 done: loss 0.1193 - lr: 0.000054
2023-10-08 21:42:45,499 DEV : loss 0.15980316698551178 - f1-score (micro avg) 0.8172
2023-10-08 21:42:45,505 ----------------------------------------------------------------------------------------------------
2023-10-08 21:42:54,521 epoch 8 - iter 12/121 - loss 0.09638295 - time (sec): 9.01 - samples/sec: 258.69 - lr: 0.000052 - momentum: 0.000000
2023-10-08 21:43:03,749 epoch 8 - iter 24/121 - loss 0.09546170 - time (sec): 18.24 - samples/sec: 267.40 - lr: 0.000051 - momentum: 0.000000
2023-10-08 21:43:13,070 epoch 8 - iter 36/121 - loss 0.08445985 - time (sec): 27.56 - samples/sec: 267.34 - lr: 0.000049 - momentum: 0.000000
2023-10-08 21:43:22,538 epoch 8 - iter 48/121 - loss 0.08900468 - time (sec): 37.03 - samples/sec: 268.25 - lr: 0.000047 - momentum: 0.000000
2023-10-08 21:43:32,545 epoch 8 - iter 60/121 - loss 0.08897714 - time (sec): 47.04 - samples/sec: 267.88 - lr: 0.000045 - momentum: 0.000000
2023-10-08 21:43:41,929 epoch 8 - iter 72/121 - loss 0.09257647 - time (sec): 56.42 - samples/sec: 266.20 - lr: 0.000044 - momentum: 0.000000
2023-10-08 21:43:50,894 epoch 8 - iter 84/121 - loss 0.09215473 - time (sec): 65.39 - samples/sec: 264.18 - lr: 0.000042 - momentum: 0.000000
2023-10-08 21:44:00,184 epoch 8 - iter 96/121 - loss 0.09869736 - time (sec): 74.68 - samples/sec: 263.56 - lr: 0.000040 - momentum: 0.000000
2023-10-08 21:44:10,029 epoch 8 - iter 108/121 - loss 0.09812610 - time (sec): 84.52 - samples/sec: 264.24 - lr: 0.000038 - momentum: 0.000000
2023-10-08 21:44:19,112 epoch 8 - iter 120/121 - loss 0.09960459 - time (sec): 93.61 - samples/sec: 263.20 - lr: 0.000037 - momentum: 0.000000
2023-10-08 21:44:19,597 ----------------------------------------------------------------------------------------------------
2023-10-08 21:44:19,597 EPOCH 8 done: loss 0.0993 - lr: 0.000037
2023-10-08 21:44:26,170 DEV : loss 0.15049293637275696 - f1-score (micro avg) 0.8193
2023-10-08 21:44:26,179 ----------------------------------------------------------------------------------------------------
2023-10-08 21:44:35,831 epoch 9 - iter 12/121 - loss 0.06910157 - time (sec): 9.65 - samples/sec: 250.64 - lr: 0.000035 - momentum: 0.000000
2023-10-08 21:44:45,166 epoch 9 - iter 24/121 - loss 0.07832480 - time (sec): 18.99 - samples/sec: 257.66 - lr: 0.000033 - momentum: 0.000000
2023-10-08 21:44:54,473 epoch 9 - iter 36/121 - loss 0.07871226 - time (sec): 28.29 - samples/sec: 258.33 - lr: 0.000031 - momentum: 0.000000
2023-10-08 21:45:04,266 epoch 9 - iter 48/121 - loss 0.07976700 - time (sec): 38.09 - samples/sec: 262.69 - lr: 0.000029 - momentum: 0.000000
2023-10-08 21:45:13,300 epoch 9 - iter 60/121 - loss 0.07905326 - time (sec): 47.12 - samples/sec: 262.01 - lr: 0.000028 - momentum: 0.000000
2023-10-08 21:45:22,870 epoch 9 - iter 72/121 - loss 0.08407537 - time (sec): 56.69 - samples/sec: 262.69 - lr: 0.000026 - momentum: 0.000000
2023-10-08 21:45:32,345 epoch 9 - iter 84/121 - loss 0.08388859 - time (sec): 66.16 - samples/sec: 263.71 - lr: 0.000024 - momentum: 0.000000
2023-10-08 21:45:41,281 epoch 9 - iter 96/121 - loss 0.08408272 - time (sec): 75.10 - samples/sec: 262.09 - lr: 0.000022 - momentum: 0.000000
2023-10-08 21:45:50,852 epoch 9 - iter 108/121 - loss 0.08356777 - time (sec): 84.67 - samples/sec: 261.19 - lr: 0.000021 - momentum: 0.000000
2023-10-08 21:46:00,175 epoch 9 - iter 120/121 - loss 0.08488223 - time (sec): 94.00 - samples/sec: 261.99 - lr: 0.000019 - momentum: 0.000000
2023-10-08 21:46:00,664 ----------------------------------------------------------------------------------------------------
2023-10-08 21:46:00,664 EPOCH 9 done: loss 0.0846 - lr: 0.000019
2023-10-08 21:46:07,146 DEV : loss 0.14527595043182373 - f1-score (micro avg) 0.8169
2023-10-08 21:46:07,152 ----------------------------------------------------------------------------------------------------
2023-10-08 21:46:17,560 epoch 10 - iter 12/121 - loss 0.08353176 - time (sec): 10.41 - samples/sec: 273.58 - lr: 0.000017 - momentum: 0.000000
2023-10-08 21:46:26,823 epoch 10 - iter 24/121 - loss 0.08497474 - time (sec): 19.67 - samples/sec: 270.56 - lr: 0.000015 - momentum: 0.000000
2023-10-08 21:46:36,078 epoch 10 - iter 36/121 - loss 0.08914676 - time (sec): 28.92 - samples/sec: 270.12 - lr: 0.000013 - momentum: 0.000000
2023-10-08 21:46:44,735 epoch 10 - iter 48/121 - loss 0.08317598 - time (sec): 37.58 - samples/sec: 266.33 - lr: 0.000012 - momentum: 0.000000
2023-10-08 21:46:54,605 epoch 10 - iter 60/121 - loss 0.08104612 - time (sec): 47.45 - samples/sec: 266.14 - lr: 0.000010 - momentum: 0.000000
2023-10-08 21:47:04,434 epoch 10 - iter 72/121 - loss 0.08313807 - time (sec): 57.28 - samples/sec: 265.80 - lr: 0.000008 - momentum: 0.000000
2023-10-08 21:47:13,165 epoch 10 - iter 84/121 - loss 0.08085529 - time (sec): 66.01 - samples/sec: 264.09 - lr: 0.000006 - momentum: 0.000000
2023-10-08 21:47:22,358 epoch 10 - iter 96/121 - loss 0.08208188 - time (sec): 75.20 - samples/sec: 264.56 - lr: 0.000005 - momentum: 0.000000
2023-10-08 21:47:31,868 epoch 10 - iter 108/121 - loss 0.08103821 - time (sec): 84.72 - samples/sec: 264.24 - lr: 0.000003 - momentum: 0.000000
2023-10-08 21:47:40,524 epoch 10 - iter 120/121 - loss 0.07968567 - time (sec): 93.37 - samples/sec: 262.35 - lr: 0.000001 - momentum: 0.000000
2023-10-08 21:47:41,290 ----------------------------------------------------------------------------------------------------
2023-10-08 21:47:41,290 EPOCH 10 done: loss 0.0793 - lr: 0.000001
2023-10-08 21:47:47,951 DEV : loss 0.14235417544841766 - f1-score (micro avg) 0.8225
2023-10-08 21:47:48,823 ----------------------------------------------------------------------------------------------------
2023-10-08 21:47:48,824 Loading model from best epoch ...
2023-10-08 21:47:51,886 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 21:47:58,362
Results:
- F-score (micro) 0.7703
- F-score (macro) 0.4636
- Accuracy 0.6582
By class:
precision recall f1-score support
pers 0.7468 0.8489 0.7946 139
scope 0.7655 0.8605 0.8102 129
work 0.7273 0.7000 0.7134 80
loc 0.0000 0.0000 0.0000 9
date 0.0000 0.0000 0.0000 3
micro avg 0.7500 0.7917 0.7703 360
macro avg 0.4479 0.4819 0.4636 360
weighted avg 0.7243 0.7917 0.7557 360
2023-10-08 21:47:58,362 ----------------------------------------------------------------------------------------------------
|