File size: 25,451 Bytes
1c3fbff |
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 |
2023-10-11 00:23:05,794 ----------------------------------------------------------------------------------------------------
2023-10-11 00:23:05,796 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 00:23:05,796 ----------------------------------------------------------------------------------------------------
2023-10-11 00:23:05,796 MultiCorpus: 7142 train + 698 dev + 2570 test sentences
- NER_HIPE_2022 Corpus: 7142 train + 698 dev + 2570 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/fr/with_doc_seperator
2023-10-11 00:23:05,796 ----------------------------------------------------------------------------------------------------
2023-10-11 00:23:05,796 Train: 7142 sentences
2023-10-11 00:23:05,797 (train_with_dev=False, train_with_test=False)
2023-10-11 00:23:05,797 ----------------------------------------------------------------------------------------------------
2023-10-11 00:23:05,797 Training Params:
2023-10-11 00:23:05,797 - learning_rate: "0.00016"
2023-10-11 00:23:05,797 - mini_batch_size: "4"
2023-10-11 00:23:05,797 - max_epochs: "10"
2023-10-11 00:23:05,797 - shuffle: "True"
2023-10-11 00:23:05,797 ----------------------------------------------------------------------------------------------------
2023-10-11 00:23:05,797 Plugins:
2023-10-11 00:23:05,797 - TensorboardLogger
2023-10-11 00:23:05,797 - LinearScheduler | warmup_fraction: '0.1'
2023-10-11 00:23:05,797 ----------------------------------------------------------------------------------------------------
2023-10-11 00:23:05,797 Final evaluation on model from best epoch (best-model.pt)
2023-10-11 00:23:05,797 - metric: "('micro avg', 'f1-score')"
2023-10-11 00:23:05,797 ----------------------------------------------------------------------------------------------------
2023-10-11 00:23:05,798 Computation:
2023-10-11 00:23:05,798 - compute on device: cuda:0
2023-10-11 00:23:05,798 - embedding storage: none
2023-10-11 00:23:05,798 ----------------------------------------------------------------------------------------------------
2023-10-11 00:23:05,798 Model training base path: "hmbench-newseye/fr-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs4-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-1"
2023-10-11 00:23:05,798 ----------------------------------------------------------------------------------------------------
2023-10-11 00:23:05,798 ----------------------------------------------------------------------------------------------------
2023-10-11 00:23:05,798 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-11 00:24:00,729 epoch 1 - iter 178/1786 - loss 2.82172262 - time (sec): 54.93 - samples/sec: 463.83 - lr: 0.000016 - momentum: 0.000000
2023-10-11 00:24:54,757 epoch 1 - iter 356/1786 - loss 2.68438929 - time (sec): 108.96 - samples/sec: 462.67 - lr: 0.000032 - momentum: 0.000000
2023-10-11 00:25:49,197 epoch 1 - iter 534/1786 - loss 2.40556334 - time (sec): 163.40 - samples/sec: 459.28 - lr: 0.000048 - momentum: 0.000000
2023-10-11 00:26:43,891 epoch 1 - iter 712/1786 - loss 2.09683094 - time (sec): 218.09 - samples/sec: 456.19 - lr: 0.000064 - momentum: 0.000000
2023-10-11 00:27:39,073 epoch 1 - iter 890/1786 - loss 1.79374454 - time (sec): 273.27 - samples/sec: 459.99 - lr: 0.000080 - momentum: 0.000000
2023-10-11 00:28:31,677 epoch 1 - iter 1068/1786 - loss 1.59687181 - time (sec): 325.88 - samples/sec: 456.68 - lr: 0.000096 - momentum: 0.000000
2023-10-11 00:29:23,556 epoch 1 - iter 1246/1786 - loss 1.43709470 - time (sec): 377.76 - samples/sec: 456.38 - lr: 0.000112 - momentum: 0.000000
2023-10-11 00:30:17,433 epoch 1 - iter 1424/1786 - loss 1.30135011 - time (sec): 431.63 - samples/sec: 457.91 - lr: 0.000127 - momentum: 0.000000
2023-10-11 00:31:10,923 epoch 1 - iter 1602/1786 - loss 1.19051970 - time (sec): 485.12 - samples/sec: 460.52 - lr: 0.000143 - momentum: 0.000000
2023-10-11 00:32:04,890 epoch 1 - iter 1780/1786 - loss 1.10265404 - time (sec): 539.09 - samples/sec: 459.99 - lr: 0.000159 - momentum: 0.000000
2023-10-11 00:32:06,577 ----------------------------------------------------------------------------------------------------
2023-10-11 00:32:06,577 EPOCH 1 done: loss 1.1001 - lr: 0.000159
2023-10-11 00:32:26,381 DEV : loss 0.21496298909187317 - f1-score (micro avg) 0.4516
2023-10-11 00:32:26,413 saving best model
2023-10-11 00:32:27,258 ----------------------------------------------------------------------------------------------------
2023-10-11 00:33:22,790 epoch 2 - iter 178/1786 - loss 0.22116459 - time (sec): 55.53 - samples/sec: 476.30 - lr: 0.000158 - momentum: 0.000000
2023-10-11 00:34:16,303 epoch 2 - iter 356/1786 - loss 0.21373852 - time (sec): 109.04 - samples/sec: 462.61 - lr: 0.000156 - momentum: 0.000000
2023-10-11 00:35:09,232 epoch 2 - iter 534/1786 - loss 0.19894140 - time (sec): 161.97 - samples/sec: 458.65 - lr: 0.000155 - momentum: 0.000000
2023-10-11 00:36:03,022 epoch 2 - iter 712/1786 - loss 0.18566527 - time (sec): 215.76 - samples/sec: 460.41 - lr: 0.000153 - momentum: 0.000000
2023-10-11 00:36:56,171 epoch 2 - iter 890/1786 - loss 0.17473385 - time (sec): 268.91 - samples/sec: 461.71 - lr: 0.000151 - momentum: 0.000000
2023-10-11 00:37:48,589 epoch 2 - iter 1068/1786 - loss 0.16918782 - time (sec): 321.33 - samples/sec: 461.01 - lr: 0.000149 - momentum: 0.000000
2023-10-11 00:38:43,484 epoch 2 - iter 1246/1786 - loss 0.16287825 - time (sec): 376.22 - samples/sec: 461.21 - lr: 0.000148 - momentum: 0.000000
2023-10-11 00:39:39,645 epoch 2 - iter 1424/1786 - loss 0.15775704 - time (sec): 432.38 - samples/sec: 461.75 - lr: 0.000146 - momentum: 0.000000
2023-10-11 00:40:34,151 epoch 2 - iter 1602/1786 - loss 0.15320043 - time (sec): 486.89 - samples/sec: 459.43 - lr: 0.000144 - momentum: 0.000000
2023-10-11 00:41:28,520 epoch 2 - iter 1780/1786 - loss 0.14884221 - time (sec): 541.26 - samples/sec: 458.41 - lr: 0.000142 - momentum: 0.000000
2023-10-11 00:41:30,090 ----------------------------------------------------------------------------------------------------
2023-10-11 00:41:30,091 EPOCH 2 done: loss 0.1488 - lr: 0.000142
2023-10-11 00:41:51,592 DEV : loss 0.10961832106113434 - f1-score (micro avg) 0.7418
2023-10-11 00:41:51,627 saving best model
2023-10-11 00:42:03,657 ----------------------------------------------------------------------------------------------------
2023-10-11 00:42:59,110 epoch 3 - iter 178/1786 - loss 0.07855565 - time (sec): 55.45 - samples/sec: 430.85 - lr: 0.000140 - momentum: 0.000000
2023-10-11 00:43:55,091 epoch 3 - iter 356/1786 - loss 0.07506546 - time (sec): 111.43 - samples/sec: 443.18 - lr: 0.000139 - momentum: 0.000000
2023-10-11 00:44:52,205 epoch 3 - iter 534/1786 - loss 0.07999800 - time (sec): 168.54 - samples/sec: 439.31 - lr: 0.000137 - momentum: 0.000000
2023-10-11 00:45:47,516 epoch 3 - iter 712/1786 - loss 0.08377628 - time (sec): 223.86 - samples/sec: 435.06 - lr: 0.000135 - momentum: 0.000000
2023-10-11 00:46:45,191 epoch 3 - iter 890/1786 - loss 0.08368157 - time (sec): 281.53 - samples/sec: 437.88 - lr: 0.000133 - momentum: 0.000000
2023-10-11 00:47:41,048 epoch 3 - iter 1068/1786 - loss 0.08196451 - time (sec): 337.39 - samples/sec: 439.01 - lr: 0.000132 - momentum: 0.000000
2023-10-11 00:48:37,387 epoch 3 - iter 1246/1786 - loss 0.07913681 - time (sec): 393.73 - samples/sec: 439.17 - lr: 0.000130 - momentum: 0.000000
2023-10-11 00:49:34,302 epoch 3 - iter 1424/1786 - loss 0.07927759 - time (sec): 450.64 - samples/sec: 440.23 - lr: 0.000128 - momentum: 0.000000
2023-10-11 00:50:31,342 epoch 3 - iter 1602/1786 - loss 0.07914526 - time (sec): 507.68 - samples/sec: 443.41 - lr: 0.000126 - momentum: 0.000000
2023-10-11 00:51:25,788 epoch 3 - iter 1780/1786 - loss 0.07955825 - time (sec): 562.13 - samples/sec: 441.21 - lr: 0.000125 - momentum: 0.000000
2023-10-11 00:51:27,478 ----------------------------------------------------------------------------------------------------
2023-10-11 00:51:27,479 EPOCH 3 done: loss 0.0795 - lr: 0.000125
2023-10-11 00:51:50,488 DEV : loss 0.13271120190620422 - f1-score (micro avg) 0.7694
2023-10-11 00:51:50,522 saving best model
2023-10-11 00:51:59,738 ----------------------------------------------------------------------------------------------------
2023-10-11 00:52:53,740 epoch 4 - iter 178/1786 - loss 0.05116751 - time (sec): 54.00 - samples/sec: 461.44 - lr: 0.000123 - momentum: 0.000000
2023-10-11 00:53:48,364 epoch 4 - iter 356/1786 - loss 0.05373327 - time (sec): 108.62 - samples/sec: 453.07 - lr: 0.000121 - momentum: 0.000000
2023-10-11 00:54:42,706 epoch 4 - iter 534/1786 - loss 0.05528519 - time (sec): 162.96 - samples/sec: 452.82 - lr: 0.000119 - momentum: 0.000000
2023-10-11 00:55:39,034 epoch 4 - iter 712/1786 - loss 0.05948955 - time (sec): 219.29 - samples/sec: 452.44 - lr: 0.000117 - momentum: 0.000000
2023-10-11 00:56:35,623 epoch 4 - iter 890/1786 - loss 0.05774224 - time (sec): 275.88 - samples/sec: 455.07 - lr: 0.000116 - momentum: 0.000000
2023-10-11 00:57:31,802 epoch 4 - iter 1068/1786 - loss 0.05684100 - time (sec): 332.06 - samples/sec: 453.93 - lr: 0.000114 - momentum: 0.000000
2023-10-11 00:58:28,188 epoch 4 - iter 1246/1786 - loss 0.05589756 - time (sec): 388.45 - samples/sec: 455.82 - lr: 0.000112 - momentum: 0.000000
2023-10-11 00:59:21,599 epoch 4 - iter 1424/1786 - loss 0.05592335 - time (sec): 441.86 - samples/sec: 456.10 - lr: 0.000110 - momentum: 0.000000
2023-10-11 01:00:15,044 epoch 4 - iter 1602/1786 - loss 0.05630542 - time (sec): 495.30 - samples/sec: 454.18 - lr: 0.000109 - momentum: 0.000000
2023-10-11 01:01:08,426 epoch 4 - iter 1780/1786 - loss 0.05643132 - time (sec): 548.68 - samples/sec: 452.21 - lr: 0.000107 - momentum: 0.000000
2023-10-11 01:01:09,965 ----------------------------------------------------------------------------------------------------
2023-10-11 01:01:09,966 EPOCH 4 done: loss 0.0565 - lr: 0.000107
2023-10-11 01:01:32,933 DEV : loss 0.1545310765504837 - f1-score (micro avg) 0.754
2023-10-11 01:01:32,964 ----------------------------------------------------------------------------------------------------
2023-10-11 01:02:28,970 epoch 5 - iter 178/1786 - loss 0.03780263 - time (sec): 56.00 - samples/sec: 454.04 - lr: 0.000105 - momentum: 0.000000
2023-10-11 01:03:22,450 epoch 5 - iter 356/1786 - loss 0.04158210 - time (sec): 109.48 - samples/sec: 444.41 - lr: 0.000103 - momentum: 0.000000
2023-10-11 01:04:19,870 epoch 5 - iter 534/1786 - loss 0.03924275 - time (sec): 166.90 - samples/sec: 446.20 - lr: 0.000101 - momentum: 0.000000
2023-10-11 01:05:16,274 epoch 5 - iter 712/1786 - loss 0.04250127 - time (sec): 223.31 - samples/sec: 450.83 - lr: 0.000100 - momentum: 0.000000
2023-10-11 01:06:10,674 epoch 5 - iter 890/1786 - loss 0.04170514 - time (sec): 277.71 - samples/sec: 444.62 - lr: 0.000098 - momentum: 0.000000
2023-10-11 01:07:04,253 epoch 5 - iter 1068/1786 - loss 0.04064150 - time (sec): 331.29 - samples/sec: 444.86 - lr: 0.000096 - momentum: 0.000000
2023-10-11 01:07:58,515 epoch 5 - iter 1246/1786 - loss 0.04058142 - time (sec): 385.55 - samples/sec: 447.34 - lr: 0.000094 - momentum: 0.000000
2023-10-11 01:08:52,190 epoch 5 - iter 1424/1786 - loss 0.04122103 - time (sec): 439.22 - samples/sec: 451.22 - lr: 0.000093 - momentum: 0.000000
2023-10-11 01:09:46,923 epoch 5 - iter 1602/1786 - loss 0.04009365 - time (sec): 493.96 - samples/sec: 451.63 - lr: 0.000091 - momentum: 0.000000
2023-10-11 01:10:40,618 epoch 5 - iter 1780/1786 - loss 0.04017012 - time (sec): 547.65 - samples/sec: 452.91 - lr: 0.000089 - momentum: 0.000000
2023-10-11 01:10:42,299 ----------------------------------------------------------------------------------------------------
2023-10-11 01:10:42,299 EPOCH 5 done: loss 0.0402 - lr: 0.000089
2023-10-11 01:11:05,576 DEV : loss 0.17520776391029358 - f1-score (micro avg) 0.793
2023-10-11 01:11:05,607 saving best model
2023-10-11 01:11:16,215 ----------------------------------------------------------------------------------------------------
2023-10-11 01:12:11,344 epoch 6 - iter 178/1786 - loss 0.03032200 - time (sec): 55.12 - samples/sec: 452.24 - lr: 0.000087 - momentum: 0.000000
2023-10-11 01:13:06,283 epoch 6 - iter 356/1786 - loss 0.03095283 - time (sec): 110.06 - samples/sec: 450.08 - lr: 0.000085 - momentum: 0.000000
2023-10-11 01:14:01,870 epoch 6 - iter 534/1786 - loss 0.03211693 - time (sec): 165.65 - samples/sec: 453.72 - lr: 0.000084 - momentum: 0.000000
2023-10-11 01:14:56,756 epoch 6 - iter 712/1786 - loss 0.03217743 - time (sec): 220.54 - samples/sec: 449.82 - lr: 0.000082 - momentum: 0.000000
2023-10-11 01:15:49,861 epoch 6 - iter 890/1786 - loss 0.03143026 - time (sec): 273.64 - samples/sec: 449.60 - lr: 0.000080 - momentum: 0.000000
2023-10-11 01:16:43,783 epoch 6 - iter 1068/1786 - loss 0.03104840 - time (sec): 327.56 - samples/sec: 451.43 - lr: 0.000078 - momentum: 0.000000
2023-10-11 01:17:38,716 epoch 6 - iter 1246/1786 - loss 0.03042104 - time (sec): 382.50 - samples/sec: 455.49 - lr: 0.000077 - momentum: 0.000000
2023-10-11 01:18:33,149 epoch 6 - iter 1424/1786 - loss 0.03050314 - time (sec): 436.93 - samples/sec: 455.70 - lr: 0.000075 - momentum: 0.000000
2023-10-11 01:19:26,347 epoch 6 - iter 1602/1786 - loss 0.03080150 - time (sec): 490.13 - samples/sec: 458.40 - lr: 0.000073 - momentum: 0.000000
2023-10-11 01:20:20,368 epoch 6 - iter 1780/1786 - loss 0.03128837 - time (sec): 544.15 - samples/sec: 455.81 - lr: 0.000071 - momentum: 0.000000
2023-10-11 01:20:22,030 ----------------------------------------------------------------------------------------------------
2023-10-11 01:20:22,030 EPOCH 6 done: loss 0.0312 - lr: 0.000071
2023-10-11 01:20:43,987 DEV : loss 0.1924738883972168 - f1-score (micro avg) 0.7722
2023-10-11 01:20:44,018 ----------------------------------------------------------------------------------------------------
2023-10-11 01:21:37,787 epoch 7 - iter 178/1786 - loss 0.01587708 - time (sec): 53.77 - samples/sec: 471.93 - lr: 0.000069 - momentum: 0.000000
2023-10-11 01:22:29,472 epoch 7 - iter 356/1786 - loss 0.01825781 - time (sec): 105.45 - samples/sec: 461.97 - lr: 0.000068 - momentum: 0.000000
2023-10-11 01:23:22,449 epoch 7 - iter 534/1786 - loss 0.01732356 - time (sec): 158.43 - samples/sec: 468.28 - lr: 0.000066 - momentum: 0.000000
2023-10-11 01:24:16,983 epoch 7 - iter 712/1786 - loss 0.01837650 - time (sec): 212.96 - samples/sec: 464.65 - lr: 0.000064 - momentum: 0.000000
2023-10-11 01:25:10,070 epoch 7 - iter 890/1786 - loss 0.01927023 - time (sec): 266.05 - samples/sec: 462.18 - lr: 0.000062 - momentum: 0.000000
2023-10-11 01:26:04,003 epoch 7 - iter 1068/1786 - loss 0.01865808 - time (sec): 319.98 - samples/sec: 463.41 - lr: 0.000061 - momentum: 0.000000
2023-10-11 01:26:59,688 epoch 7 - iter 1246/1786 - loss 0.01941833 - time (sec): 375.67 - samples/sec: 460.91 - lr: 0.000059 - momentum: 0.000000
2023-10-11 01:27:51,703 epoch 7 - iter 1424/1786 - loss 0.01972491 - time (sec): 427.68 - samples/sec: 459.36 - lr: 0.000057 - momentum: 0.000000
2023-10-11 01:28:45,857 epoch 7 - iter 1602/1786 - loss 0.02012094 - time (sec): 481.84 - samples/sec: 462.62 - lr: 0.000055 - momentum: 0.000000
2023-10-11 01:29:38,522 epoch 7 - iter 1780/1786 - loss 0.02068942 - time (sec): 534.50 - samples/sec: 464.17 - lr: 0.000053 - momentum: 0.000000
2023-10-11 01:29:40,057 ----------------------------------------------------------------------------------------------------
2023-10-11 01:29:40,057 EPOCH 7 done: loss 0.0206 - lr: 0.000053
2023-10-11 01:30:01,229 DEV : loss 0.21992838382720947 - f1-score (micro avg) 0.7884
2023-10-11 01:30:01,263 ----------------------------------------------------------------------------------------------------
2023-10-11 01:30:53,920 epoch 8 - iter 178/1786 - loss 0.01299467 - time (sec): 52.65 - samples/sec: 466.34 - lr: 0.000052 - momentum: 0.000000
2023-10-11 01:31:45,976 epoch 8 - iter 356/1786 - loss 0.01441319 - time (sec): 104.71 - samples/sec: 463.25 - lr: 0.000050 - momentum: 0.000000
2023-10-11 01:32:38,715 epoch 8 - iter 534/1786 - loss 0.01543810 - time (sec): 157.45 - samples/sec: 457.64 - lr: 0.000048 - momentum: 0.000000
2023-10-11 01:33:34,520 epoch 8 - iter 712/1786 - loss 0.01558449 - time (sec): 213.26 - samples/sec: 460.35 - lr: 0.000046 - momentum: 0.000000
2023-10-11 01:34:29,192 epoch 8 - iter 890/1786 - loss 0.01606204 - time (sec): 267.93 - samples/sec: 457.95 - lr: 0.000044 - momentum: 0.000000
2023-10-11 01:35:23,050 epoch 8 - iter 1068/1786 - loss 0.01737134 - time (sec): 321.78 - samples/sec: 453.47 - lr: 0.000043 - momentum: 0.000000
2023-10-11 01:36:18,022 epoch 8 - iter 1246/1786 - loss 0.01717115 - time (sec): 376.76 - samples/sec: 453.38 - lr: 0.000041 - momentum: 0.000000
2023-10-11 01:37:12,230 epoch 8 - iter 1424/1786 - loss 0.01719750 - time (sec): 430.96 - samples/sec: 452.83 - lr: 0.000039 - momentum: 0.000000
2023-10-11 01:38:07,659 epoch 8 - iter 1602/1786 - loss 0.01716691 - time (sec): 486.39 - samples/sec: 454.73 - lr: 0.000037 - momentum: 0.000000
2023-10-11 01:39:04,440 epoch 8 - iter 1780/1786 - loss 0.01671130 - time (sec): 543.18 - samples/sec: 456.11 - lr: 0.000036 - momentum: 0.000000
2023-10-11 01:39:06,365 ----------------------------------------------------------------------------------------------------
2023-10-11 01:39:06,365 EPOCH 8 done: loss 0.0169 - lr: 0.000036
2023-10-11 01:39:29,230 DEV : loss 0.22382444143295288 - f1-score (micro avg) 0.7783
2023-10-11 01:39:29,261 ----------------------------------------------------------------------------------------------------
2023-10-11 01:40:24,277 epoch 9 - iter 178/1786 - loss 0.01495584 - time (sec): 55.01 - samples/sec: 452.58 - lr: 0.000034 - momentum: 0.000000
2023-10-11 01:41:18,298 epoch 9 - iter 356/1786 - loss 0.01430322 - time (sec): 109.03 - samples/sec: 447.87 - lr: 0.000032 - momentum: 0.000000
2023-10-11 01:42:13,747 epoch 9 - iter 534/1786 - loss 0.01462408 - time (sec): 164.48 - samples/sec: 454.99 - lr: 0.000030 - momentum: 0.000000
2023-10-11 01:43:06,569 epoch 9 - iter 712/1786 - loss 0.01383529 - time (sec): 217.31 - samples/sec: 449.99 - lr: 0.000028 - momentum: 0.000000
2023-10-11 01:44:00,263 epoch 9 - iter 890/1786 - loss 0.01279666 - time (sec): 271.00 - samples/sec: 449.54 - lr: 0.000027 - momentum: 0.000000
2023-10-11 01:44:55,264 epoch 9 - iter 1068/1786 - loss 0.01221011 - time (sec): 326.00 - samples/sec: 448.49 - lr: 0.000025 - momentum: 0.000000
2023-10-11 01:45:49,115 epoch 9 - iter 1246/1786 - loss 0.01201829 - time (sec): 379.85 - samples/sec: 448.44 - lr: 0.000023 - momentum: 0.000000
2023-10-11 01:46:43,117 epoch 9 - iter 1424/1786 - loss 0.01119229 - time (sec): 433.85 - samples/sec: 451.22 - lr: 0.000021 - momentum: 0.000000
2023-10-11 01:47:37,095 epoch 9 - iter 1602/1786 - loss 0.01109461 - time (sec): 487.83 - samples/sec: 453.37 - lr: 0.000020 - momentum: 0.000000
2023-10-11 01:48:31,386 epoch 9 - iter 1780/1786 - loss 0.01142929 - time (sec): 542.12 - samples/sec: 457.32 - lr: 0.000018 - momentum: 0.000000
2023-10-11 01:48:33,130 ----------------------------------------------------------------------------------------------------
2023-10-11 01:48:33,131 EPOCH 9 done: loss 0.0115 - lr: 0.000018
2023-10-11 01:48:54,395 DEV : loss 0.23885728418827057 - f1-score (micro avg) 0.7909
2023-10-11 01:48:54,425 ----------------------------------------------------------------------------------------------------
2023-10-11 01:49:47,365 epoch 10 - iter 178/1786 - loss 0.00824939 - time (sec): 52.94 - samples/sec: 476.73 - lr: 0.000016 - momentum: 0.000000
2023-10-11 01:50:40,024 epoch 10 - iter 356/1786 - loss 0.00754479 - time (sec): 105.60 - samples/sec: 466.45 - lr: 0.000014 - momentum: 0.000000
2023-10-11 01:51:31,760 epoch 10 - iter 534/1786 - loss 0.00886912 - time (sec): 157.33 - samples/sec: 459.24 - lr: 0.000012 - momentum: 0.000000
2023-10-11 01:52:26,703 epoch 10 - iter 712/1786 - loss 0.00857059 - time (sec): 212.28 - samples/sec: 464.60 - lr: 0.000011 - momentum: 0.000000
2023-10-11 01:53:21,373 epoch 10 - iter 890/1786 - loss 0.00828606 - time (sec): 266.95 - samples/sec: 467.58 - lr: 0.000009 - momentum: 0.000000
2023-10-11 01:54:16,053 epoch 10 - iter 1068/1786 - loss 0.00847447 - time (sec): 321.63 - samples/sec: 462.32 - lr: 0.000007 - momentum: 0.000000
2023-10-11 01:55:13,521 epoch 10 - iter 1246/1786 - loss 0.00884080 - time (sec): 379.09 - samples/sec: 461.95 - lr: 0.000005 - momentum: 0.000000
2023-10-11 01:56:09,216 epoch 10 - iter 1424/1786 - loss 0.00980399 - time (sec): 434.79 - samples/sec: 456.91 - lr: 0.000004 - momentum: 0.000000
2023-10-11 01:57:05,107 epoch 10 - iter 1602/1786 - loss 0.00968986 - time (sec): 490.68 - samples/sec: 453.59 - lr: 0.000002 - momentum: 0.000000
2023-10-11 01:58:03,236 epoch 10 - iter 1780/1786 - loss 0.00921449 - time (sec): 548.81 - samples/sec: 451.98 - lr: 0.000000 - momentum: 0.000000
2023-10-11 01:58:05,045 ----------------------------------------------------------------------------------------------------
2023-10-11 01:58:05,046 EPOCH 10 done: loss 0.0092 - lr: 0.000000
2023-10-11 01:58:26,928 DEV : loss 0.24146509170532227 - f1-score (micro avg) 0.7818
2023-10-11 01:58:27,995 ----------------------------------------------------------------------------------------------------
2023-10-11 01:58:27,997 Loading model from best epoch ...
2023-10-11 01:58:32,001 SequenceTagger predicts: Dictionary with 17 tags: O, S-PER, B-PER, E-PER, I-PER, S-LOC, B-LOC, E-LOC, I-LOC, S-ORG, B-ORG, E-ORG, I-ORG, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd
2023-10-11 01:59:39,506
Results:
- F-score (micro) 0.6929
- F-score (macro) 0.6294
- Accuracy 0.5446
By class:
precision recall f1-score support
LOC 0.7365 0.6740 0.7039 1095
PER 0.7930 0.7569 0.7745 1012
ORG 0.4201 0.5742 0.4852 357
HumanProd 0.5625 0.5455 0.5538 33
micro avg 0.6941 0.6916 0.6929 2497
macro avg 0.6280 0.6376 0.6294 2497
weighted avg 0.7119 0.6916 0.6993 2497
2023-10-11 01:59:39,506 ----------------------------------------------------------------------------------------------------
|