File size: 25,531 Bytes
01f75c1 |
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 05:01:34,727 ----------------------------------------------------------------------------------------------------
2023-10-11 05:01:34,729 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 05:01:34,729 ----------------------------------------------------------------------------------------------------
2023-10-11 05:01:34,729 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 05:01:34,729 ----------------------------------------------------------------------------------------------------
2023-10-11 05:01:34,729 Train: 7142 sentences
2023-10-11 05:01:34,729 (train_with_dev=False, train_with_test=False)
2023-10-11 05:01:34,729 ----------------------------------------------------------------------------------------------------
2023-10-11 05:01:34,730 Training Params:
2023-10-11 05:01:34,730 - learning_rate: "0.00015"
2023-10-11 05:01:34,730 - mini_batch_size: "4"
2023-10-11 05:01:34,730 - max_epochs: "10"
2023-10-11 05:01:34,730 - shuffle: "True"
2023-10-11 05:01:34,730 ----------------------------------------------------------------------------------------------------
2023-10-11 05:01:34,730 Plugins:
2023-10-11 05:01:34,730 - TensorboardLogger
2023-10-11 05:01:34,730 - LinearScheduler | warmup_fraction: '0.1'
2023-10-11 05:01:34,730 ----------------------------------------------------------------------------------------------------
2023-10-11 05:01:34,730 Final evaluation on model from best epoch (best-model.pt)
2023-10-11 05:01:34,730 - metric: "('micro avg', 'f1-score')"
2023-10-11 05:01:34,730 ----------------------------------------------------------------------------------------------------
2023-10-11 05:01:34,730 Computation:
2023-10-11 05:01:34,730 - compute on device: cuda:0
2023-10-11 05:01:34,730 - embedding storage: none
2023-10-11 05:01:34,731 ----------------------------------------------------------------------------------------------------
2023-10-11 05:01:34,731 Model training base path: "hmbench-newseye/fr-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs4-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-2"
2023-10-11 05:01:34,731 ----------------------------------------------------------------------------------------------------
2023-10-11 05:01:34,731 ----------------------------------------------------------------------------------------------------
2023-10-11 05:01:34,731 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-11 05:02:27,827 epoch 1 - iter 178/1786 - loss 2.83041151 - time (sec): 53.09 - samples/sec: 500.15 - lr: 0.000015 - momentum: 0.000000
2023-10-11 05:03:20,133 epoch 1 - iter 356/1786 - loss 2.69429508 - time (sec): 105.40 - samples/sec: 491.24 - lr: 0.000030 - momentum: 0.000000
2023-10-11 05:04:14,338 epoch 1 - iter 534/1786 - loss 2.41681822 - time (sec): 159.61 - samples/sec: 486.67 - lr: 0.000045 - momentum: 0.000000
2023-10-11 05:05:10,271 epoch 1 - iter 712/1786 - loss 2.10924834 - time (sec): 215.54 - samples/sec: 477.63 - lr: 0.000060 - momentum: 0.000000
2023-10-11 05:06:05,087 epoch 1 - iter 890/1786 - loss 1.84410867 - time (sec): 270.35 - samples/sec: 470.72 - lr: 0.000075 - momentum: 0.000000
2023-10-11 05:07:05,666 epoch 1 - iter 1068/1786 - loss 1.63794432 - time (sec): 330.93 - samples/sec: 462.45 - lr: 0.000090 - momentum: 0.000000
2023-10-11 05:08:04,516 epoch 1 - iter 1246/1786 - loss 1.47552627 - time (sec): 389.78 - samples/sec: 456.44 - lr: 0.000105 - momentum: 0.000000
2023-10-11 05:09:00,303 epoch 1 - iter 1424/1786 - loss 1.35086011 - time (sec): 445.57 - samples/sec: 451.33 - lr: 0.000120 - momentum: 0.000000
2023-10-11 05:09:52,316 epoch 1 - iter 1602/1786 - loss 1.24502390 - time (sec): 497.58 - samples/sec: 451.20 - lr: 0.000134 - momentum: 0.000000
2023-10-11 05:10:43,774 epoch 1 - iter 1780/1786 - loss 1.15419317 - time (sec): 549.04 - samples/sec: 452.16 - lr: 0.000149 - momentum: 0.000000
2023-10-11 05:10:45,213 ----------------------------------------------------------------------------------------------------
2023-10-11 05:10:45,213 EPOCH 1 done: loss 1.1525 - lr: 0.000149
2023-10-11 05:11:06,239 DEV : loss 0.20120850205421448 - f1-score (micro avg) 0.5061
2023-10-11 05:11:06,271 saving best model
2023-10-11 05:11:07,380 ----------------------------------------------------------------------------------------------------
2023-10-11 05:12:00,838 epoch 2 - iter 178/1786 - loss 0.20647638 - time (sec): 53.46 - samples/sec: 470.14 - lr: 0.000148 - momentum: 0.000000
2023-10-11 05:12:55,186 epoch 2 - iter 356/1786 - loss 0.19196816 - time (sec): 107.80 - samples/sec: 477.92 - lr: 0.000147 - momentum: 0.000000
2023-10-11 05:13:49,188 epoch 2 - iter 534/1786 - loss 0.17945236 - time (sec): 161.81 - samples/sec: 471.84 - lr: 0.000145 - momentum: 0.000000
2023-10-11 05:14:41,335 epoch 2 - iter 712/1786 - loss 0.17099654 - time (sec): 213.95 - samples/sec: 469.79 - lr: 0.000143 - momentum: 0.000000
2023-10-11 05:15:33,535 epoch 2 - iter 890/1786 - loss 0.16314737 - time (sec): 266.15 - samples/sec: 468.04 - lr: 0.000142 - momentum: 0.000000
2023-10-11 05:16:27,194 epoch 2 - iter 1068/1786 - loss 0.15692538 - time (sec): 319.81 - samples/sec: 464.19 - lr: 0.000140 - momentum: 0.000000
2023-10-11 05:17:17,836 epoch 2 - iter 1246/1786 - loss 0.15088690 - time (sec): 370.45 - samples/sec: 464.60 - lr: 0.000138 - momentum: 0.000000
2023-10-11 05:18:09,337 epoch 2 - iter 1424/1786 - loss 0.14597758 - time (sec): 421.96 - samples/sec: 467.64 - lr: 0.000137 - momentum: 0.000000
2023-10-11 05:19:01,609 epoch 2 - iter 1602/1786 - loss 0.14238346 - time (sec): 474.23 - samples/sec: 471.09 - lr: 0.000135 - momentum: 0.000000
2023-10-11 05:19:52,592 epoch 2 - iter 1780/1786 - loss 0.13731741 - time (sec): 525.21 - samples/sec: 472.20 - lr: 0.000133 - momentum: 0.000000
2023-10-11 05:19:54,174 ----------------------------------------------------------------------------------------------------
2023-10-11 05:19:54,174 EPOCH 2 done: loss 0.1372 - lr: 0.000133
2023-10-11 05:20:14,463 DEV : loss 0.11661199480295181 - f1-score (micro avg) 0.7552
2023-10-11 05:20:14,492 saving best model
2023-10-11 05:20:17,182 ----------------------------------------------------------------------------------------------------
2023-10-11 05:21:15,759 epoch 3 - iter 178/1786 - loss 0.07763298 - time (sec): 58.57 - samples/sec: 441.25 - lr: 0.000132 - momentum: 0.000000
2023-10-11 05:22:11,182 epoch 3 - iter 356/1786 - loss 0.07495553 - time (sec): 113.99 - samples/sec: 426.96 - lr: 0.000130 - momentum: 0.000000
2023-10-11 05:23:09,155 epoch 3 - iter 534/1786 - loss 0.07669306 - time (sec): 171.97 - samples/sec: 430.39 - lr: 0.000128 - momentum: 0.000000
2023-10-11 05:24:05,593 epoch 3 - iter 712/1786 - loss 0.07775822 - time (sec): 228.41 - samples/sec: 430.05 - lr: 0.000127 - momentum: 0.000000
2023-10-11 05:25:03,308 epoch 3 - iter 890/1786 - loss 0.07467629 - time (sec): 286.12 - samples/sec: 429.61 - lr: 0.000125 - momentum: 0.000000
2023-10-11 05:26:00,547 epoch 3 - iter 1068/1786 - loss 0.07681723 - time (sec): 343.36 - samples/sec: 431.47 - lr: 0.000123 - momentum: 0.000000
2023-10-11 05:26:58,181 epoch 3 - iter 1246/1786 - loss 0.07505527 - time (sec): 400.99 - samples/sec: 435.90 - lr: 0.000122 - momentum: 0.000000
2023-10-11 05:27:52,072 epoch 3 - iter 1424/1786 - loss 0.07490696 - time (sec): 454.88 - samples/sec: 438.92 - lr: 0.000120 - momentum: 0.000000
2023-10-11 05:28:46,503 epoch 3 - iter 1602/1786 - loss 0.07479295 - time (sec): 509.32 - samples/sec: 439.46 - lr: 0.000118 - momentum: 0.000000
2023-10-11 05:29:46,396 epoch 3 - iter 1780/1786 - loss 0.07517529 - time (sec): 569.21 - samples/sec: 436.00 - lr: 0.000117 - momentum: 0.000000
2023-10-11 05:29:48,337 ----------------------------------------------------------------------------------------------------
2023-10-11 05:29:48,338 EPOCH 3 done: loss 0.0754 - lr: 0.000117
2023-10-11 05:30:11,692 DEV : loss 0.1217304989695549 - f1-score (micro avg) 0.7823
2023-10-11 05:30:11,727 saving best model
2023-10-11 05:30:14,519 ----------------------------------------------------------------------------------------------------
2023-10-11 05:31:12,835 epoch 4 - iter 178/1786 - loss 0.06283829 - time (sec): 58.31 - samples/sec: 424.49 - lr: 0.000115 - momentum: 0.000000
2023-10-11 05:32:06,563 epoch 4 - iter 356/1786 - loss 0.05265726 - time (sec): 112.04 - samples/sec: 426.97 - lr: 0.000113 - momentum: 0.000000
2023-10-11 05:33:01,812 epoch 4 - iter 534/1786 - loss 0.05130324 - time (sec): 167.29 - samples/sec: 441.84 - lr: 0.000112 - momentum: 0.000000
2023-10-11 05:33:57,758 epoch 4 - iter 712/1786 - loss 0.05156142 - time (sec): 223.24 - samples/sec: 450.76 - lr: 0.000110 - momentum: 0.000000
2023-10-11 05:34:50,238 epoch 4 - iter 890/1786 - loss 0.05132046 - time (sec): 275.72 - samples/sec: 448.93 - lr: 0.000108 - momentum: 0.000000
2023-10-11 05:35:46,545 epoch 4 - iter 1068/1786 - loss 0.05071305 - time (sec): 332.02 - samples/sec: 447.62 - lr: 0.000107 - momentum: 0.000000
2023-10-11 05:36:42,170 epoch 4 - iter 1246/1786 - loss 0.05180539 - time (sec): 387.65 - samples/sec: 450.20 - lr: 0.000105 - momentum: 0.000000
2023-10-11 05:37:38,584 epoch 4 - iter 1424/1786 - loss 0.05196249 - time (sec): 444.06 - samples/sec: 447.63 - lr: 0.000103 - momentum: 0.000000
2023-10-11 05:38:33,429 epoch 4 - iter 1602/1786 - loss 0.05160455 - time (sec): 498.91 - samples/sec: 447.19 - lr: 0.000102 - momentum: 0.000000
2023-10-11 05:39:31,504 epoch 4 - iter 1780/1786 - loss 0.05213671 - time (sec): 556.98 - samples/sec: 445.74 - lr: 0.000100 - momentum: 0.000000
2023-10-11 05:39:33,161 ----------------------------------------------------------------------------------------------------
2023-10-11 05:39:33,161 EPOCH 4 done: loss 0.0520 - lr: 0.000100
2023-10-11 05:39:55,740 DEV : loss 0.15838001668453217 - f1-score (micro avg) 0.7935
2023-10-11 05:39:55,771 saving best model
2023-10-11 05:39:58,431 ----------------------------------------------------------------------------------------------------
2023-10-11 05:40:58,919 epoch 5 - iter 178/1786 - loss 0.04547964 - time (sec): 60.48 - samples/sec: 411.44 - lr: 0.000098 - momentum: 0.000000
2023-10-11 05:41:53,865 epoch 5 - iter 356/1786 - loss 0.04560730 - time (sec): 115.43 - samples/sec: 411.80 - lr: 0.000097 - momentum: 0.000000
2023-10-11 05:42:52,796 epoch 5 - iter 534/1786 - loss 0.04303715 - time (sec): 174.36 - samples/sec: 415.11 - lr: 0.000095 - momentum: 0.000000
2023-10-11 05:43:49,710 epoch 5 - iter 712/1786 - loss 0.04135883 - time (sec): 231.27 - samples/sec: 420.84 - lr: 0.000093 - momentum: 0.000000
2023-10-11 05:44:48,917 epoch 5 - iter 890/1786 - loss 0.04142354 - time (sec): 290.48 - samples/sec: 417.48 - lr: 0.000092 - momentum: 0.000000
2023-10-11 05:45:52,793 epoch 5 - iter 1068/1786 - loss 0.03982523 - time (sec): 354.36 - samples/sec: 412.77 - lr: 0.000090 - momentum: 0.000000
2023-10-11 05:47:01,510 epoch 5 - iter 1246/1786 - loss 0.04033156 - time (sec): 423.07 - samples/sec: 408.31 - lr: 0.000088 - momentum: 0.000000
2023-10-11 05:48:01,676 epoch 5 - iter 1424/1786 - loss 0.04015505 - time (sec): 483.24 - samples/sec: 408.92 - lr: 0.000087 - momentum: 0.000000
2023-10-11 05:48:58,632 epoch 5 - iter 1602/1786 - loss 0.03983915 - time (sec): 540.20 - samples/sec: 411.43 - lr: 0.000085 - momentum: 0.000000
2023-10-11 05:49:52,916 epoch 5 - iter 1780/1786 - loss 0.03991586 - time (sec): 594.48 - samples/sec: 417.29 - lr: 0.000083 - momentum: 0.000000
2023-10-11 05:49:54,495 ----------------------------------------------------------------------------------------------------
2023-10-11 05:49:54,496 EPOCH 5 done: loss 0.0398 - lr: 0.000083
2023-10-11 05:50:15,738 DEV : loss 0.1628066748380661 - f1-score (micro avg) 0.8089
2023-10-11 05:50:15,773 saving best model
2023-10-11 05:50:18,533 ----------------------------------------------------------------------------------------------------
2023-10-11 05:51:12,723 epoch 6 - iter 178/1786 - loss 0.02789412 - time (sec): 54.19 - samples/sec: 456.91 - lr: 0.000082 - momentum: 0.000000
2023-10-11 05:52:07,670 epoch 6 - iter 356/1786 - loss 0.02958477 - time (sec): 109.13 - samples/sec: 456.59 - lr: 0.000080 - momentum: 0.000000
2023-10-11 05:52:59,150 epoch 6 - iter 534/1786 - loss 0.02757124 - time (sec): 160.61 - samples/sec: 463.60 - lr: 0.000078 - momentum: 0.000000
2023-10-11 05:53:50,942 epoch 6 - iter 712/1786 - loss 0.02729736 - time (sec): 212.41 - samples/sec: 467.18 - lr: 0.000077 - momentum: 0.000000
2023-10-11 05:54:43,400 epoch 6 - iter 890/1786 - loss 0.02840525 - time (sec): 264.86 - samples/sec: 468.78 - lr: 0.000075 - momentum: 0.000000
2023-10-11 05:55:37,498 epoch 6 - iter 1068/1786 - loss 0.02774450 - time (sec): 318.96 - samples/sec: 470.91 - lr: 0.000073 - momentum: 0.000000
2023-10-11 05:56:32,273 epoch 6 - iter 1246/1786 - loss 0.02655476 - time (sec): 373.74 - samples/sec: 466.63 - lr: 0.000072 - momentum: 0.000000
2023-10-11 05:57:26,142 epoch 6 - iter 1424/1786 - loss 0.02755150 - time (sec): 427.61 - samples/sec: 466.59 - lr: 0.000070 - momentum: 0.000000
2023-10-11 05:58:19,208 epoch 6 - iter 1602/1786 - loss 0.02776326 - time (sec): 480.67 - samples/sec: 468.70 - lr: 0.000068 - momentum: 0.000000
2023-10-11 05:59:09,786 epoch 6 - iter 1780/1786 - loss 0.02837332 - time (sec): 531.25 - samples/sec: 467.37 - lr: 0.000067 - momentum: 0.000000
2023-10-11 05:59:11,193 ----------------------------------------------------------------------------------------------------
2023-10-11 05:59:11,193 EPOCH 6 done: loss 0.0283 - lr: 0.000067
2023-10-11 05:59:32,445 DEV : loss 0.17363940179347992 - f1-score (micro avg) 0.8075
2023-10-11 05:59:32,474 ----------------------------------------------------------------------------------------------------
2023-10-11 06:00:27,457 epoch 7 - iter 178/1786 - loss 0.02062730 - time (sec): 54.98 - samples/sec: 497.62 - lr: 0.000065 - momentum: 0.000000
2023-10-11 06:01:26,025 epoch 7 - iter 356/1786 - loss 0.02080307 - time (sec): 113.55 - samples/sec: 457.95 - lr: 0.000063 - momentum: 0.000000
2023-10-11 06:02:18,555 epoch 7 - iter 534/1786 - loss 0.02116990 - time (sec): 166.08 - samples/sec: 455.85 - lr: 0.000062 - momentum: 0.000000
2023-10-11 06:03:10,462 epoch 7 - iter 712/1786 - loss 0.02046165 - time (sec): 217.99 - samples/sec: 458.64 - lr: 0.000060 - momentum: 0.000000
2023-10-11 06:04:02,581 epoch 7 - iter 890/1786 - loss 0.02124109 - time (sec): 270.10 - samples/sec: 460.37 - lr: 0.000058 - momentum: 0.000000
2023-10-11 06:04:55,342 epoch 7 - iter 1068/1786 - loss 0.01994612 - time (sec): 322.86 - samples/sec: 464.15 - lr: 0.000057 - momentum: 0.000000
2023-10-11 06:05:46,465 epoch 7 - iter 1246/1786 - loss 0.02017532 - time (sec): 373.99 - samples/sec: 464.41 - lr: 0.000055 - momentum: 0.000000
2023-10-11 06:06:39,403 epoch 7 - iter 1424/1786 - loss 0.01973361 - time (sec): 426.93 - samples/sec: 468.63 - lr: 0.000053 - momentum: 0.000000
2023-10-11 06:07:31,780 epoch 7 - iter 1602/1786 - loss 0.01962053 - time (sec): 479.30 - samples/sec: 468.69 - lr: 0.000052 - momentum: 0.000000
2023-10-11 06:08:23,880 epoch 7 - iter 1780/1786 - loss 0.02041587 - time (sec): 531.40 - samples/sec: 466.78 - lr: 0.000050 - momentum: 0.000000
2023-10-11 06:08:25,507 ----------------------------------------------------------------------------------------------------
2023-10-11 06:08:25,507 EPOCH 7 done: loss 0.0204 - lr: 0.000050
2023-10-11 06:08:46,785 DEV : loss 0.1936260312795639 - f1-score (micro avg) 0.8109
2023-10-11 06:08:46,815 saving best model
2023-10-11 06:08:49,446 ----------------------------------------------------------------------------------------------------
2023-10-11 06:09:41,617 epoch 8 - iter 178/1786 - loss 0.01944064 - time (sec): 52.17 - samples/sec: 461.54 - lr: 0.000048 - momentum: 0.000000
2023-10-11 06:10:33,775 epoch 8 - iter 356/1786 - loss 0.01633750 - time (sec): 104.32 - samples/sec: 465.74 - lr: 0.000047 - momentum: 0.000000
2023-10-11 06:11:26,536 epoch 8 - iter 534/1786 - loss 0.01426110 - time (sec): 157.09 - samples/sec: 468.29 - lr: 0.000045 - momentum: 0.000000
2023-10-11 06:12:20,764 epoch 8 - iter 712/1786 - loss 0.01661153 - time (sec): 211.31 - samples/sec: 470.04 - lr: 0.000043 - momentum: 0.000000
2023-10-11 06:13:16,553 epoch 8 - iter 890/1786 - loss 0.01686518 - time (sec): 267.10 - samples/sec: 462.43 - lr: 0.000042 - momentum: 0.000000
2023-10-11 06:14:17,499 epoch 8 - iter 1068/1786 - loss 0.01792696 - time (sec): 328.05 - samples/sec: 452.14 - lr: 0.000040 - momentum: 0.000000
2023-10-11 06:15:22,194 epoch 8 - iter 1246/1786 - loss 0.01811955 - time (sec): 392.74 - samples/sec: 439.83 - lr: 0.000038 - momentum: 0.000000
2023-10-11 06:16:23,362 epoch 8 - iter 1424/1786 - loss 0.01715705 - time (sec): 453.91 - samples/sec: 437.14 - lr: 0.000037 - momentum: 0.000000
2023-10-11 06:17:23,666 epoch 8 - iter 1602/1786 - loss 0.01642572 - time (sec): 514.22 - samples/sec: 433.14 - lr: 0.000035 - momentum: 0.000000
2023-10-11 06:18:18,761 epoch 8 - iter 1780/1786 - loss 0.01609207 - time (sec): 569.31 - samples/sec: 435.25 - lr: 0.000033 - momentum: 0.000000
2023-10-11 06:18:20,641 ----------------------------------------------------------------------------------------------------
2023-10-11 06:18:20,641 EPOCH 8 done: loss 0.0161 - lr: 0.000033
2023-10-11 06:18:42,738 DEV : loss 0.21119572222232819 - f1-score (micro avg) 0.8059
2023-10-11 06:18:42,770 ----------------------------------------------------------------------------------------------------
2023-10-11 06:19:41,501 epoch 9 - iter 178/1786 - loss 0.01445335 - time (sec): 58.73 - samples/sec: 439.60 - lr: 0.000032 - momentum: 0.000000
2023-10-11 06:20:35,546 epoch 9 - iter 356/1786 - loss 0.01611300 - time (sec): 112.77 - samples/sec: 445.46 - lr: 0.000030 - momentum: 0.000000
2023-10-11 06:21:27,480 epoch 9 - iter 534/1786 - loss 0.01242337 - time (sec): 164.71 - samples/sec: 454.46 - lr: 0.000028 - momentum: 0.000000
2023-10-11 06:22:19,171 epoch 9 - iter 712/1786 - loss 0.01188148 - time (sec): 216.40 - samples/sec: 458.47 - lr: 0.000027 - momentum: 0.000000
2023-10-11 06:23:11,965 epoch 9 - iter 890/1786 - loss 0.01133948 - time (sec): 269.19 - samples/sec: 460.33 - lr: 0.000025 - momentum: 0.000000
2023-10-11 06:24:06,959 epoch 9 - iter 1068/1786 - loss 0.01089479 - time (sec): 324.19 - samples/sec: 458.23 - lr: 0.000023 - momentum: 0.000000
2023-10-11 06:24:58,784 epoch 9 - iter 1246/1786 - loss 0.01082033 - time (sec): 376.01 - samples/sec: 456.11 - lr: 0.000022 - momentum: 0.000000
2023-10-11 06:25:52,869 epoch 9 - iter 1424/1786 - loss 0.01140983 - time (sec): 430.10 - samples/sec: 458.01 - lr: 0.000020 - momentum: 0.000000
2023-10-11 06:26:48,015 epoch 9 - iter 1602/1786 - loss 0.01172453 - time (sec): 485.24 - samples/sec: 460.19 - lr: 0.000018 - momentum: 0.000000
2023-10-11 06:27:40,966 epoch 9 - iter 1780/1786 - loss 0.01123782 - time (sec): 538.19 - samples/sec: 460.59 - lr: 0.000017 - momentum: 0.000000
2023-10-11 06:27:42,742 ----------------------------------------------------------------------------------------------------
2023-10-11 06:27:42,743 EPOCH 9 done: loss 0.0112 - lr: 0.000017
2023-10-11 06:28:04,945 DEV : loss 0.222304567694664 - f1-score (micro avg) 0.8
2023-10-11 06:28:04,975 ----------------------------------------------------------------------------------------------------
2023-10-11 06:28:57,976 epoch 10 - iter 178/1786 - loss 0.00772526 - time (sec): 53.00 - samples/sec: 443.35 - lr: 0.000015 - momentum: 0.000000
2023-10-11 06:29:51,408 epoch 10 - iter 356/1786 - loss 0.00840680 - time (sec): 106.43 - samples/sec: 454.34 - lr: 0.000013 - momentum: 0.000000
2023-10-11 06:30:46,308 epoch 10 - iter 534/1786 - loss 0.00752016 - time (sec): 161.33 - samples/sec: 458.27 - lr: 0.000012 - momentum: 0.000000
2023-10-11 06:31:38,377 epoch 10 - iter 712/1786 - loss 0.00697489 - time (sec): 213.40 - samples/sec: 458.38 - lr: 0.000010 - momentum: 0.000000
2023-10-11 06:32:31,964 epoch 10 - iter 890/1786 - loss 0.00802755 - time (sec): 266.99 - samples/sec: 464.30 - lr: 0.000008 - momentum: 0.000000
2023-10-11 06:33:27,144 epoch 10 - iter 1068/1786 - loss 0.00896322 - time (sec): 322.17 - samples/sec: 467.72 - lr: 0.000007 - momentum: 0.000000
2023-10-11 06:34:18,958 epoch 10 - iter 1246/1786 - loss 0.00910142 - time (sec): 373.98 - samples/sec: 464.87 - lr: 0.000005 - momentum: 0.000000
2023-10-11 06:35:13,297 epoch 10 - iter 1424/1786 - loss 0.00868295 - time (sec): 428.32 - samples/sec: 465.41 - lr: 0.000003 - momentum: 0.000000
2023-10-11 06:36:07,043 epoch 10 - iter 1602/1786 - loss 0.00832301 - time (sec): 482.07 - samples/sec: 464.24 - lr: 0.000002 - momentum: 0.000000
2023-10-11 06:37:00,746 epoch 10 - iter 1780/1786 - loss 0.00823723 - time (sec): 535.77 - samples/sec: 463.07 - lr: 0.000000 - momentum: 0.000000
2023-10-11 06:37:02,365 ----------------------------------------------------------------------------------------------------
2023-10-11 06:37:02,365 EPOCH 10 done: loss 0.0082 - lr: 0.000000
2023-10-11 06:37:23,464 DEV : loss 0.22344990074634552 - f1-score (micro avg) 0.7963
2023-10-11 06:37:24,411 ----------------------------------------------------------------------------------------------------
2023-10-11 06:37:24,413 Loading model from best epoch ...
2023-10-11 06:37:28,377 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 06:38:40,166
Results:
- F-score (micro) 0.6861
- F-score (macro) 0.5975
- Accuracy 0.5426
By class:
precision recall f1-score support
LOC 0.7197 0.6986 0.7090 1095
PER 0.7824 0.7638 0.7730 1012
ORG 0.3843 0.5770 0.4614 357
HumanProd 0.3443 0.6364 0.4468 33
micro avg 0.6665 0.7068 0.6861 2497
macro avg 0.5577 0.6690 0.5975 2497
weighted avg 0.6922 0.7068 0.6961 2497
2023-10-11 06:38:40,166 ----------------------------------------------------------------------------------------------------
|