File size: 25,618 Bytes
57b1f09 |
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 |
2023-10-12 02:19:10,693 ----------------------------------------------------------------------------------------------------
2023-10-12 02:19:10,695 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()
)"
2023-10-12 02:19:10,696 ----------------------------------------------------------------------------------------------------
2023-10-12 02:19:10,696 MultiCorpus: 20847 train + 1123 dev + 3350 test sentences
- NER_HIPE_2022 Corpus: 20847 train + 1123 dev + 3350 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/de/with_doc_seperator
2023-10-12 02:19:10,696 ----------------------------------------------------------------------------------------------------
2023-10-12 02:19:10,696 Train: 20847 sentences
2023-10-12 02:19:10,696 (train_with_dev=False, train_with_test=False)
2023-10-12 02:19:10,696 ----------------------------------------------------------------------------------------------------
2023-10-12 02:19:10,696 Training Params:
2023-10-12 02:19:10,696 - learning_rate: "0.00015"
2023-10-12 02:19:10,696 - mini_batch_size: "4"
2023-10-12 02:19:10,696 - max_epochs: "10"
2023-10-12 02:19:10,696 - shuffle: "True"
2023-10-12 02:19:10,696 ----------------------------------------------------------------------------------------------------
2023-10-12 02:19:10,696 Plugins:
2023-10-12 02:19:10,697 - TensorboardLogger
2023-10-12 02:19:10,697 - LinearScheduler | warmup_fraction: '0.1'
2023-10-12 02:19:10,697 ----------------------------------------------------------------------------------------------------
2023-10-12 02:19:10,697 Final evaluation on model from best epoch (best-model.pt)
2023-10-12 02:19:10,697 - metric: "('micro avg', 'f1-score')"
2023-10-12 02:19:10,697 ----------------------------------------------------------------------------------------------------
2023-10-12 02:19:10,697 Computation:
2023-10-12 02:19:10,697 - compute on device: cuda:0
2023-10-12 02:19:10,697 - embedding storage: none
2023-10-12 02:19:10,697 ----------------------------------------------------------------------------------------------------
2023-10-12 02:19:10,697 Model training base path: "hmbench-newseye/de-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs4-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-4"
2023-10-12 02:19:10,697 ----------------------------------------------------------------------------------------------------
2023-10-12 02:19:10,697 ----------------------------------------------------------------------------------------------------
2023-10-12 02:19:10,697 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-12 02:21:27,177 epoch 1 - iter 521/5212 - loss 2.79668095 - time (sec): 136.48 - samples/sec: 242.74 - lr: 0.000015 - momentum: 0.000000
2023-10-12 02:23:45,273 epoch 1 - iter 1042/5212 - loss 2.35843978 - time (sec): 274.57 - samples/sec: 247.36 - lr: 0.000030 - momentum: 0.000000
2023-10-12 02:26:04,245 epoch 1 - iter 1563/5212 - loss 1.80796410 - time (sec): 413.55 - samples/sec: 254.16 - lr: 0.000045 - momentum: 0.000000
2023-10-12 02:28:22,402 epoch 1 - iter 2084/5212 - loss 1.45533862 - time (sec): 551.70 - samples/sec: 257.38 - lr: 0.000060 - momentum: 0.000000
2023-10-12 02:30:41,547 epoch 1 - iter 2605/5212 - loss 1.24266578 - time (sec): 690.85 - samples/sec: 260.63 - lr: 0.000075 - momentum: 0.000000
2023-10-12 02:32:58,209 epoch 1 - iter 3126/5212 - loss 1.09834367 - time (sec): 827.51 - samples/sec: 259.96 - lr: 0.000090 - momentum: 0.000000
2023-10-12 02:35:17,942 epoch 1 - iter 3647/5212 - loss 0.98084623 - time (sec): 967.24 - samples/sec: 261.96 - lr: 0.000105 - momentum: 0.000000
2023-10-12 02:37:35,810 epoch 1 - iter 4168/5212 - loss 0.88950826 - time (sec): 1105.11 - samples/sec: 262.06 - lr: 0.000120 - momentum: 0.000000
2023-10-12 02:39:57,111 epoch 1 - iter 4689/5212 - loss 0.80746283 - time (sec): 1246.41 - samples/sec: 264.35 - lr: 0.000135 - momentum: 0.000000
2023-10-12 02:42:16,837 epoch 1 - iter 5210/5212 - loss 0.74371560 - time (sec): 1386.14 - samples/sec: 264.92 - lr: 0.000150 - momentum: 0.000000
2023-10-12 02:42:17,401 ----------------------------------------------------------------------------------------------------
2023-10-12 02:42:17,402 EPOCH 1 done: loss 0.7433 - lr: 0.000150
2023-10-12 02:42:52,120 DEV : loss 0.12636248767375946 - f1-score (micro avg) 0.2776
2023-10-12 02:42:52,174 saving best model
2023-10-12 02:42:53,044 ----------------------------------------------------------------------------------------------------
2023-10-12 02:45:11,088 epoch 2 - iter 521/5212 - loss 0.17686924 - time (sec): 138.04 - samples/sec: 262.94 - lr: 0.000148 - momentum: 0.000000
2023-10-12 02:47:30,960 epoch 2 - iter 1042/5212 - loss 0.15464838 - time (sec): 277.91 - samples/sec: 267.07 - lr: 0.000147 - momentum: 0.000000
2023-10-12 02:49:50,501 epoch 2 - iter 1563/5212 - loss 0.15677141 - time (sec): 417.45 - samples/sec: 261.48 - lr: 0.000145 - momentum: 0.000000
2023-10-12 02:52:16,593 epoch 2 - iter 2084/5212 - loss 0.15541495 - time (sec): 563.55 - samples/sec: 263.09 - lr: 0.000143 - momentum: 0.000000
2023-10-12 02:54:38,805 epoch 2 - iter 2605/5212 - loss 0.15288785 - time (sec): 705.76 - samples/sec: 261.65 - lr: 0.000142 - momentum: 0.000000
2023-10-12 02:56:58,256 epoch 2 - iter 3126/5212 - loss 0.15178570 - time (sec): 845.21 - samples/sec: 258.20 - lr: 0.000140 - momentum: 0.000000
2023-10-12 02:59:16,224 epoch 2 - iter 3647/5212 - loss 0.15320470 - time (sec): 983.18 - samples/sec: 254.99 - lr: 0.000138 - momentum: 0.000000
2023-10-12 03:01:39,472 epoch 2 - iter 4168/5212 - loss 0.14998413 - time (sec): 1126.43 - samples/sec: 256.63 - lr: 0.000137 - momentum: 0.000000
2023-10-12 03:04:04,945 epoch 2 - iter 4689/5212 - loss 0.14664370 - time (sec): 1271.90 - samples/sec: 259.73 - lr: 0.000135 - momentum: 0.000000
2023-10-12 03:06:26,106 epoch 2 - iter 5210/5212 - loss 0.14460575 - time (sec): 1413.06 - samples/sec: 259.97 - lr: 0.000133 - momentum: 0.000000
2023-10-12 03:06:26,551 ----------------------------------------------------------------------------------------------------
2023-10-12 03:06:26,552 EPOCH 2 done: loss 0.1446 - lr: 0.000133
2023-10-12 03:07:05,601 DEV : loss 0.14167223870754242 - f1-score (micro avg) 0.3339
2023-10-12 03:07:05,653 saving best model
2023-10-12 03:07:08,265 ----------------------------------------------------------------------------------------------------
2023-10-12 03:09:23,808 epoch 3 - iter 521/5212 - loss 0.10082822 - time (sec): 135.54 - samples/sec: 254.81 - lr: 0.000132 - momentum: 0.000000
2023-10-12 03:11:37,979 epoch 3 - iter 1042/5212 - loss 0.09748085 - time (sec): 269.71 - samples/sec: 250.08 - lr: 0.000130 - momentum: 0.000000
2023-10-12 03:13:58,974 epoch 3 - iter 1563/5212 - loss 0.09985524 - time (sec): 410.70 - samples/sec: 264.24 - lr: 0.000128 - momentum: 0.000000
2023-10-12 03:16:15,894 epoch 3 - iter 2084/5212 - loss 0.09997962 - time (sec): 547.62 - samples/sec: 263.03 - lr: 0.000127 - momentum: 0.000000
2023-10-12 03:18:33,733 epoch 3 - iter 2605/5212 - loss 0.09885177 - time (sec): 685.46 - samples/sec: 261.01 - lr: 0.000125 - momentum: 0.000000
2023-10-12 03:20:55,408 epoch 3 - iter 3126/5212 - loss 0.09501689 - time (sec): 827.14 - samples/sec: 264.99 - lr: 0.000123 - momentum: 0.000000
2023-10-12 03:23:18,463 epoch 3 - iter 3647/5212 - loss 0.09597740 - time (sec): 970.19 - samples/sec: 267.70 - lr: 0.000122 - momentum: 0.000000
2023-10-12 03:25:40,852 epoch 3 - iter 4168/5212 - loss 0.09794887 - time (sec): 1112.58 - samples/sec: 263.01 - lr: 0.000120 - momentum: 0.000000
2023-10-12 03:28:07,225 epoch 3 - iter 4689/5212 - loss 0.09904610 - time (sec): 1258.96 - samples/sec: 261.73 - lr: 0.000118 - momentum: 0.000000
2023-10-12 03:30:34,184 epoch 3 - iter 5210/5212 - loss 0.09816484 - time (sec): 1405.91 - samples/sec: 261.22 - lr: 0.000117 - momentum: 0.000000
2023-10-12 03:30:34,730 ----------------------------------------------------------------------------------------------------
2023-10-12 03:30:34,731 EPOCH 3 done: loss 0.0981 - lr: 0.000117
2023-10-12 03:31:15,205 DEV : loss 0.2600547671318054 - f1-score (micro avg) 0.3618
2023-10-12 03:31:15,262 saving best model
2023-10-12 03:31:17,833 ----------------------------------------------------------------------------------------------------
2023-10-12 03:33:41,338 epoch 4 - iter 521/5212 - loss 0.06934331 - time (sec): 143.50 - samples/sec: 252.51 - lr: 0.000115 - momentum: 0.000000
2023-10-12 03:36:05,531 epoch 4 - iter 1042/5212 - loss 0.07084255 - time (sec): 287.69 - samples/sec: 257.98 - lr: 0.000113 - momentum: 0.000000
2023-10-12 03:38:29,695 epoch 4 - iter 1563/5212 - loss 0.06755543 - time (sec): 431.86 - samples/sec: 261.22 - lr: 0.000112 - momentum: 0.000000
2023-10-12 03:40:54,693 epoch 4 - iter 2084/5212 - loss 0.06616838 - time (sec): 576.86 - samples/sec: 261.23 - lr: 0.000110 - momentum: 0.000000
2023-10-12 03:43:17,865 epoch 4 - iter 2605/5212 - loss 0.06558266 - time (sec): 720.03 - samples/sec: 258.99 - lr: 0.000108 - momentum: 0.000000
2023-10-12 03:45:41,115 epoch 4 - iter 3126/5212 - loss 0.06435344 - time (sec): 863.28 - samples/sec: 259.71 - lr: 0.000107 - momentum: 0.000000
2023-10-12 03:48:03,559 epoch 4 - iter 3647/5212 - loss 0.06413682 - time (sec): 1005.72 - samples/sec: 259.21 - lr: 0.000105 - momentum: 0.000000
2023-10-12 03:50:23,570 epoch 4 - iter 4168/5212 - loss 0.06596094 - time (sec): 1145.73 - samples/sec: 257.63 - lr: 0.000103 - momentum: 0.000000
2023-10-12 03:52:45,111 epoch 4 - iter 4689/5212 - loss 0.06614742 - time (sec): 1287.27 - samples/sec: 258.43 - lr: 0.000102 - momentum: 0.000000
2023-10-12 03:55:04,718 epoch 4 - iter 5210/5212 - loss 0.06648116 - time (sec): 1426.88 - samples/sec: 257.45 - lr: 0.000100 - momentum: 0.000000
2023-10-12 03:55:05,152 ----------------------------------------------------------------------------------------------------
2023-10-12 03:55:05,153 EPOCH 4 done: loss 0.0665 - lr: 0.000100
2023-10-12 03:55:45,279 DEV : loss 0.3101561367511749 - f1-score (micro avg) 0.3555
2023-10-12 03:55:45,331 ----------------------------------------------------------------------------------------------------
2023-10-12 03:58:04,744 epoch 5 - iter 521/5212 - loss 0.04426358 - time (sec): 139.41 - samples/sec: 259.84 - lr: 0.000098 - momentum: 0.000000
2023-10-12 04:00:24,001 epoch 5 - iter 1042/5212 - loss 0.04273940 - time (sec): 278.67 - samples/sec: 260.65 - lr: 0.000097 - momentum: 0.000000
2023-10-12 04:02:41,900 epoch 5 - iter 1563/5212 - loss 0.04286202 - time (sec): 416.57 - samples/sec: 256.36 - lr: 0.000095 - momentum: 0.000000
2023-10-12 04:05:03,372 epoch 5 - iter 2084/5212 - loss 0.04497980 - time (sec): 558.04 - samples/sec: 260.25 - lr: 0.000093 - momentum: 0.000000
2023-10-12 04:07:19,613 epoch 5 - iter 2605/5212 - loss 0.04479044 - time (sec): 694.28 - samples/sec: 258.94 - lr: 0.000092 - momentum: 0.000000
2023-10-12 04:09:43,942 epoch 5 - iter 3126/5212 - loss 0.04436294 - time (sec): 838.61 - samples/sec: 258.68 - lr: 0.000090 - momentum: 0.000000
2023-10-12 04:12:11,821 epoch 5 - iter 3647/5212 - loss 0.04380965 - time (sec): 986.49 - samples/sec: 258.99 - lr: 0.000088 - momentum: 0.000000
2023-10-12 04:14:37,186 epoch 5 - iter 4168/5212 - loss 0.04530386 - time (sec): 1131.85 - samples/sec: 258.58 - lr: 0.000087 - momentum: 0.000000
2023-10-12 04:17:02,247 epoch 5 - iter 4689/5212 - loss 0.04665775 - time (sec): 1276.91 - samples/sec: 257.28 - lr: 0.000085 - momentum: 0.000000
2023-10-12 04:19:31,906 epoch 5 - iter 5210/5212 - loss 0.04638068 - time (sec): 1426.57 - samples/sec: 257.51 - lr: 0.000083 - momentum: 0.000000
2023-10-12 04:19:32,352 ----------------------------------------------------------------------------------------------------
2023-10-12 04:19:32,353 EPOCH 5 done: loss 0.0464 - lr: 0.000083
2023-10-12 04:20:13,213 DEV : loss 0.3113304078578949 - f1-score (micro avg) 0.4003
2023-10-12 04:20:13,271 saving best model
2023-10-12 04:20:14,217 ----------------------------------------------------------------------------------------------------
2023-10-12 04:22:41,995 epoch 6 - iter 521/5212 - loss 0.02551644 - time (sec): 147.78 - samples/sec: 258.80 - lr: 0.000082 - momentum: 0.000000
2023-10-12 04:25:08,081 epoch 6 - iter 1042/5212 - loss 0.02937621 - time (sec): 293.86 - samples/sec: 259.22 - lr: 0.000080 - momentum: 0.000000
2023-10-12 04:27:33,632 epoch 6 - iter 1563/5212 - loss 0.03011832 - time (sec): 439.41 - samples/sec: 254.55 - lr: 0.000078 - momentum: 0.000000
2023-10-12 04:29:57,124 epoch 6 - iter 2084/5212 - loss 0.03067784 - time (sec): 582.90 - samples/sec: 252.43 - lr: 0.000077 - momentum: 0.000000
2023-10-12 04:32:22,709 epoch 6 - iter 2605/5212 - loss 0.03033894 - time (sec): 728.49 - samples/sec: 256.01 - lr: 0.000075 - momentum: 0.000000
2023-10-12 04:34:46,393 epoch 6 - iter 3126/5212 - loss 0.03057800 - time (sec): 872.17 - samples/sec: 256.98 - lr: 0.000073 - momentum: 0.000000
2023-10-12 04:37:07,410 epoch 6 - iter 3647/5212 - loss 0.03169587 - time (sec): 1013.19 - samples/sec: 256.43 - lr: 0.000072 - momentum: 0.000000
2023-10-12 04:39:27,981 epoch 6 - iter 4168/5212 - loss 0.03188952 - time (sec): 1153.76 - samples/sec: 256.11 - lr: 0.000070 - momentum: 0.000000
2023-10-12 04:41:50,066 epoch 6 - iter 4689/5212 - loss 0.03265092 - time (sec): 1295.85 - samples/sec: 256.18 - lr: 0.000068 - momentum: 0.000000
2023-10-12 04:44:09,929 epoch 6 - iter 5210/5212 - loss 0.03297566 - time (sec): 1435.71 - samples/sec: 255.87 - lr: 0.000067 - momentum: 0.000000
2023-10-12 04:44:10,371 ----------------------------------------------------------------------------------------------------
2023-10-12 04:44:10,371 EPOCH 6 done: loss 0.0330 - lr: 0.000067
2023-10-12 04:44:50,633 DEV : loss 0.40718281269073486 - f1-score (micro avg) 0.3971
2023-10-12 04:44:50,685 ----------------------------------------------------------------------------------------------------
2023-10-12 04:47:11,977 epoch 7 - iter 521/5212 - loss 0.02552297 - time (sec): 141.29 - samples/sec: 259.67 - lr: 0.000065 - momentum: 0.000000
2023-10-12 04:49:32,146 epoch 7 - iter 1042/5212 - loss 0.02437180 - time (sec): 281.46 - samples/sec: 271.70 - lr: 0.000063 - momentum: 0.000000
2023-10-12 04:51:48,960 epoch 7 - iter 1563/5212 - loss 0.02514062 - time (sec): 418.27 - samples/sec: 267.26 - lr: 0.000062 - momentum: 0.000000
2023-10-12 04:54:07,753 epoch 7 - iter 2084/5212 - loss 0.02516668 - time (sec): 557.07 - samples/sec: 267.87 - lr: 0.000060 - momentum: 0.000000
2023-10-12 04:56:27,890 epoch 7 - iter 2605/5212 - loss 0.02469412 - time (sec): 697.20 - samples/sec: 265.74 - lr: 0.000058 - momentum: 0.000000
2023-10-12 04:58:49,869 epoch 7 - iter 3126/5212 - loss 0.02456715 - time (sec): 839.18 - samples/sec: 265.93 - lr: 0.000057 - momentum: 0.000000
2023-10-12 05:01:09,634 epoch 7 - iter 3647/5212 - loss 0.02477855 - time (sec): 978.95 - samples/sec: 263.65 - lr: 0.000055 - momentum: 0.000000
2023-10-12 05:03:36,700 epoch 7 - iter 4168/5212 - loss 0.02394687 - time (sec): 1126.01 - samples/sec: 262.72 - lr: 0.000053 - momentum: 0.000000
2023-10-12 05:06:00,466 epoch 7 - iter 4689/5212 - loss 0.02356060 - time (sec): 1269.78 - samples/sec: 260.95 - lr: 0.000052 - momentum: 0.000000
2023-10-12 05:08:24,608 epoch 7 - iter 5210/5212 - loss 0.02323141 - time (sec): 1413.92 - samples/sec: 259.78 - lr: 0.000050 - momentum: 0.000000
2023-10-12 05:08:25,097 ----------------------------------------------------------------------------------------------------
2023-10-12 05:08:25,097 EPOCH 7 done: loss 0.0232 - lr: 0.000050
2023-10-12 05:09:04,885 DEV : loss 0.408222496509552 - f1-score (micro avg) 0.4035
2023-10-12 05:09:04,936 saving best model
2023-10-12 05:09:07,501 ----------------------------------------------------------------------------------------------------
2023-10-12 05:11:27,797 epoch 8 - iter 521/5212 - loss 0.01537607 - time (sec): 140.29 - samples/sec: 265.21 - lr: 0.000048 - momentum: 0.000000
2023-10-12 05:13:47,952 epoch 8 - iter 1042/5212 - loss 0.01753064 - time (sec): 280.45 - samples/sec: 271.51 - lr: 0.000047 - momentum: 0.000000
2023-10-12 05:16:11,225 epoch 8 - iter 1563/5212 - loss 0.01696724 - time (sec): 423.72 - samples/sec: 277.40 - lr: 0.000045 - momentum: 0.000000
2023-10-12 05:18:28,174 epoch 8 - iter 2084/5212 - loss 0.01669412 - time (sec): 560.67 - samples/sec: 273.78 - lr: 0.000043 - momentum: 0.000000
2023-10-12 05:20:45,546 epoch 8 - iter 2605/5212 - loss 0.01686027 - time (sec): 698.04 - samples/sec: 269.95 - lr: 0.000042 - momentum: 0.000000
2023-10-12 05:23:00,618 epoch 8 - iter 3126/5212 - loss 0.01671656 - time (sec): 833.11 - samples/sec: 267.12 - lr: 0.000040 - momentum: 0.000000
2023-10-12 05:25:16,222 epoch 8 - iter 3647/5212 - loss 0.01600572 - time (sec): 968.72 - samples/sec: 265.31 - lr: 0.000038 - momentum: 0.000000
2023-10-12 05:27:35,013 epoch 8 - iter 4168/5212 - loss 0.01597679 - time (sec): 1107.51 - samples/sec: 264.67 - lr: 0.000037 - momentum: 0.000000
2023-10-12 05:29:56,551 epoch 8 - iter 4689/5212 - loss 0.01553053 - time (sec): 1249.05 - samples/sec: 263.33 - lr: 0.000035 - momentum: 0.000000
2023-10-12 05:32:19,571 epoch 8 - iter 5210/5212 - loss 0.01639257 - time (sec): 1392.06 - samples/sec: 263.90 - lr: 0.000033 - momentum: 0.000000
2023-10-12 05:32:19,996 ----------------------------------------------------------------------------------------------------
2023-10-12 05:32:19,997 EPOCH 8 done: loss 0.0164 - lr: 0.000033
2023-10-12 05:32:58,072 DEV : loss 0.4335840940475464 - f1-score (micro avg) 0.404
2023-10-12 05:32:58,123 saving best model
2023-10-12 05:33:00,797 ----------------------------------------------------------------------------------------------------
2023-10-12 05:35:21,905 epoch 9 - iter 521/5212 - loss 0.01030793 - time (sec): 141.10 - samples/sec: 275.22 - lr: 0.000032 - momentum: 0.000000
2023-10-12 05:37:41,298 epoch 9 - iter 1042/5212 - loss 0.01148381 - time (sec): 280.50 - samples/sec: 274.31 - lr: 0.000030 - momentum: 0.000000
2023-10-12 05:40:02,195 epoch 9 - iter 1563/5212 - loss 0.01090402 - time (sec): 421.39 - samples/sec: 263.64 - lr: 0.000028 - momentum: 0.000000
2023-10-12 05:42:24,062 epoch 9 - iter 2084/5212 - loss 0.01205054 - time (sec): 563.26 - samples/sec: 259.21 - lr: 0.000027 - momentum: 0.000000
2023-10-12 05:44:49,921 epoch 9 - iter 2605/5212 - loss 0.01249863 - time (sec): 709.12 - samples/sec: 258.65 - lr: 0.000025 - momentum: 0.000000
2023-10-12 05:47:12,873 epoch 9 - iter 3126/5212 - loss 0.01215066 - time (sec): 852.07 - samples/sec: 258.10 - lr: 0.000023 - momentum: 0.000000
2023-10-12 05:49:39,769 epoch 9 - iter 3647/5212 - loss 0.01116151 - time (sec): 998.97 - samples/sec: 259.45 - lr: 0.000022 - momentum: 0.000000
2023-10-12 05:52:01,961 epoch 9 - iter 4168/5212 - loss 0.01081521 - time (sec): 1141.16 - samples/sec: 257.32 - lr: 0.000020 - momentum: 0.000000
2023-10-12 05:54:25,864 epoch 9 - iter 4689/5212 - loss 0.01082631 - time (sec): 1285.06 - samples/sec: 256.89 - lr: 0.000018 - momentum: 0.000000
2023-10-12 05:56:50,574 epoch 9 - iter 5210/5212 - loss 0.01087435 - time (sec): 1429.77 - samples/sec: 256.94 - lr: 0.000017 - momentum: 0.000000
2023-10-12 05:56:51,009 ----------------------------------------------------------------------------------------------------
2023-10-12 05:56:51,009 EPOCH 9 done: loss 0.0109 - lr: 0.000017
2023-10-12 05:57:30,552 DEV : loss 0.47575101256370544 - f1-score (micro avg) 0.3948
2023-10-12 05:57:30,605 ----------------------------------------------------------------------------------------------------
2023-10-12 05:59:52,416 epoch 10 - iter 521/5212 - loss 0.00498377 - time (sec): 141.81 - samples/sec: 252.02 - lr: 0.000015 - momentum: 0.000000
2023-10-12 06:02:13,588 epoch 10 - iter 1042/5212 - loss 0.00681335 - time (sec): 282.98 - samples/sec: 255.31 - lr: 0.000013 - momentum: 0.000000
2023-10-12 06:04:37,692 epoch 10 - iter 1563/5212 - loss 0.00622093 - time (sec): 427.08 - samples/sec: 259.59 - lr: 0.000012 - momentum: 0.000000
2023-10-12 06:06:59,831 epoch 10 - iter 2084/5212 - loss 0.00629281 - time (sec): 569.22 - samples/sec: 258.24 - lr: 0.000010 - momentum: 0.000000
2023-10-12 06:09:21,704 epoch 10 - iter 2605/5212 - loss 0.00707237 - time (sec): 711.10 - samples/sec: 257.28 - lr: 0.000008 - momentum: 0.000000
2023-10-12 06:11:42,090 epoch 10 - iter 3126/5212 - loss 0.00730698 - time (sec): 851.48 - samples/sec: 256.91 - lr: 0.000007 - momentum: 0.000000
2023-10-12 06:14:01,717 epoch 10 - iter 3647/5212 - loss 0.00730729 - time (sec): 991.11 - samples/sec: 258.63 - lr: 0.000005 - momentum: 0.000000
2023-10-12 06:16:22,285 epoch 10 - iter 4168/5212 - loss 0.00690845 - time (sec): 1131.68 - samples/sec: 260.93 - lr: 0.000003 - momentum: 0.000000
2023-10-12 06:18:43,018 epoch 10 - iter 4689/5212 - loss 0.00696219 - time (sec): 1272.41 - samples/sec: 260.74 - lr: 0.000002 - momentum: 0.000000
2023-10-12 06:21:02,617 epoch 10 - iter 5210/5212 - loss 0.00715270 - time (sec): 1412.01 - samples/sec: 260.15 - lr: 0.000000 - momentum: 0.000000
2023-10-12 06:21:03,064 ----------------------------------------------------------------------------------------------------
2023-10-12 06:21:03,065 EPOCH 10 done: loss 0.0072 - lr: 0.000000
2023-10-12 06:21:42,939 DEV : loss 0.4921533763408661 - f1-score (micro avg) 0.3907
2023-10-12 06:21:43,893 ----------------------------------------------------------------------------------------------------
2023-10-12 06:21:43,895 Loading model from best epoch ...
2023-10-12 06:21:47,649 SequenceTagger predicts: Dictionary with 17 tags: O, S-LOC, B-LOC, E-LOC, I-LOC, S-PER, B-PER, E-PER, I-PER, S-ORG, B-ORG, E-ORG, I-ORG, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd
2023-10-12 06:23:28,188
Results:
- F-score (micro) 0.4681
- F-score (macro) 0.3274
- Accuracy 0.3108
By class:
precision recall f1-score support
LOC 0.5033 0.5610 0.5306 1214
PER 0.4123 0.4567 0.4334 808
ORG 0.3282 0.3654 0.3458 353
HumanProd 0.0000 0.0000 0.0000 15
micro avg 0.4454 0.4933 0.4681 2390
macro avg 0.3110 0.3458 0.3274 2390
weighted avg 0.4435 0.4933 0.4671 2390
2023-10-12 06:23:28,188 ----------------------------------------------------------------------------------------------------
|