File size: 23,916 Bytes
7bcf3d2 |
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 |
2023-10-17 08:35:02,192 ----------------------------------------------------------------------------------------------------
2023-10-17 08:35:02,193 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): ElectraModel(
(embeddings): ElectraEmbeddings(
(word_embeddings): Embedding(32001, 768)
(position_embeddings): Embedding(512, 768)
(token_type_embeddings): Embedding(2, 768)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(encoder): ElectraEncoder(
(layer): ModuleList(
(0-11): 12 x ElectraLayer(
(attention): ElectraAttention(
(self): ElectraSelfAttention(
(query): Linear(in_features=768, out_features=768, bias=True)
(key): Linear(in_features=768, out_features=768, bias=True)
(value): Linear(in_features=768, out_features=768, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(output): ElectraSelfOutput(
(dense): Linear(in_features=768, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(intermediate): ElectraIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): ElectraOutput(
(dense): Linear(in_features=3072, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=768, out_features=25, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-17 08:35:02,193 ----------------------------------------------------------------------------------------------------
2023-10-17 08:35:02,193 MultiCorpus: 1100 train + 206 dev + 240 test sentences
- NER_HIPE_2022 Corpus: 1100 train + 206 dev + 240 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/ajmc/de/with_doc_seperator
2023-10-17 08:35:02,193 ----------------------------------------------------------------------------------------------------
2023-10-17 08:35:02,193 Train: 1100 sentences
2023-10-17 08:35:02,193 (train_with_dev=False, train_with_test=False)
2023-10-17 08:35:02,193 ----------------------------------------------------------------------------------------------------
2023-10-17 08:35:02,193 Training Params:
2023-10-17 08:35:02,193 - learning_rate: "5e-05"
2023-10-17 08:35:02,193 - mini_batch_size: "8"
2023-10-17 08:35:02,193 - max_epochs: "10"
2023-10-17 08:35:02,193 - shuffle: "True"
2023-10-17 08:35:02,193 ----------------------------------------------------------------------------------------------------
2023-10-17 08:35:02,193 Plugins:
2023-10-17 08:35:02,194 - TensorboardLogger
2023-10-17 08:35:02,194 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 08:35:02,194 ----------------------------------------------------------------------------------------------------
2023-10-17 08:35:02,194 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 08:35:02,194 - metric: "('micro avg', 'f1-score')"
2023-10-17 08:35:02,194 ----------------------------------------------------------------------------------------------------
2023-10-17 08:35:02,194 Computation:
2023-10-17 08:35:02,194 - compute on device: cuda:0
2023-10-17 08:35:02,194 - embedding storage: none
2023-10-17 08:35:02,194 ----------------------------------------------------------------------------------------------------
2023-10-17 08:35:02,194 Model training base path: "hmbench-ajmc/de-hmteams/teams-base-historic-multilingual-discriminator-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-2"
2023-10-17 08:35:02,194 ----------------------------------------------------------------------------------------------------
2023-10-17 08:35:02,194 ----------------------------------------------------------------------------------------------------
2023-10-17 08:35:02,194 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 08:35:02,926 epoch 1 - iter 13/138 - loss 4.22040289 - time (sec): 0.73 - samples/sec: 2911.78 - lr: 0.000004 - momentum: 0.000000
2023-10-17 08:35:03,651 epoch 1 - iter 26/138 - loss 3.70465772 - time (sec): 1.46 - samples/sec: 2775.13 - lr: 0.000009 - momentum: 0.000000
2023-10-17 08:35:04,368 epoch 1 - iter 39/138 - loss 3.00403495 - time (sec): 2.17 - samples/sec: 2774.98 - lr: 0.000014 - momentum: 0.000000
2023-10-17 08:35:05,127 epoch 1 - iter 52/138 - loss 2.50089604 - time (sec): 2.93 - samples/sec: 2741.49 - lr: 0.000018 - momentum: 0.000000
2023-10-17 08:35:05,943 epoch 1 - iter 65/138 - loss 2.07745839 - time (sec): 3.75 - samples/sec: 2764.64 - lr: 0.000023 - momentum: 0.000000
2023-10-17 08:35:06,711 epoch 1 - iter 78/138 - loss 1.81431920 - time (sec): 4.52 - samples/sec: 2781.72 - lr: 0.000028 - momentum: 0.000000
2023-10-17 08:35:07,462 epoch 1 - iter 91/138 - loss 1.59917697 - time (sec): 5.27 - samples/sec: 2803.24 - lr: 0.000033 - momentum: 0.000000
2023-10-17 08:35:08,228 epoch 1 - iter 104/138 - loss 1.43328790 - time (sec): 6.03 - samples/sec: 2814.17 - lr: 0.000037 - momentum: 0.000000
2023-10-17 08:35:08,990 epoch 1 - iter 117/138 - loss 1.30451502 - time (sec): 6.80 - samples/sec: 2833.88 - lr: 0.000042 - momentum: 0.000000
2023-10-17 08:35:09,780 epoch 1 - iter 130/138 - loss 1.20666017 - time (sec): 7.59 - samples/sec: 2828.47 - lr: 0.000047 - momentum: 0.000000
2023-10-17 08:35:10,283 ----------------------------------------------------------------------------------------------------
2023-10-17 08:35:10,283 EPOCH 1 done: loss 1.1565 - lr: 0.000047
2023-10-17 08:35:11,092 DEV : loss 0.20371423661708832 - f1-score (micro avg) 0.6853
2023-10-17 08:35:11,097 saving best model
2023-10-17 08:35:11,441 ----------------------------------------------------------------------------------------------------
2023-10-17 08:35:12,205 epoch 2 - iter 13/138 - loss 0.18510989 - time (sec): 0.76 - samples/sec: 2908.46 - lr: 0.000050 - momentum: 0.000000
2023-10-17 08:35:12,987 epoch 2 - iter 26/138 - loss 0.22286473 - time (sec): 1.54 - samples/sec: 3015.76 - lr: 0.000049 - momentum: 0.000000
2023-10-17 08:35:13,732 epoch 2 - iter 39/138 - loss 0.21354263 - time (sec): 2.29 - samples/sec: 2992.66 - lr: 0.000048 - momentum: 0.000000
2023-10-17 08:35:14,511 epoch 2 - iter 52/138 - loss 0.20370386 - time (sec): 3.07 - samples/sec: 2972.89 - lr: 0.000048 - momentum: 0.000000
2023-10-17 08:35:15,209 epoch 2 - iter 65/138 - loss 0.19557993 - time (sec): 3.77 - samples/sec: 2942.96 - lr: 0.000047 - momentum: 0.000000
2023-10-17 08:35:15,959 epoch 2 - iter 78/138 - loss 0.18670587 - time (sec): 4.52 - samples/sec: 2887.25 - lr: 0.000047 - momentum: 0.000000
2023-10-17 08:35:16,651 epoch 2 - iter 91/138 - loss 0.18106996 - time (sec): 5.21 - samples/sec: 2857.52 - lr: 0.000046 - momentum: 0.000000
2023-10-17 08:35:17,408 epoch 2 - iter 104/138 - loss 0.17656932 - time (sec): 5.97 - samples/sec: 2896.66 - lr: 0.000046 - momentum: 0.000000
2023-10-17 08:35:18,168 epoch 2 - iter 117/138 - loss 0.17142517 - time (sec): 6.73 - samples/sec: 2880.42 - lr: 0.000045 - momentum: 0.000000
2023-10-17 08:35:18,911 epoch 2 - iter 130/138 - loss 0.17318261 - time (sec): 7.47 - samples/sec: 2888.38 - lr: 0.000045 - momentum: 0.000000
2023-10-17 08:35:19,343 ----------------------------------------------------------------------------------------------------
2023-10-17 08:35:19,344 EPOCH 2 done: loss 0.1682 - lr: 0.000045
2023-10-17 08:35:19,974 DEV : loss 0.15141892433166504 - f1-score (micro avg) 0.8127
2023-10-17 08:35:19,979 saving best model
2023-10-17 08:35:20,422 ----------------------------------------------------------------------------------------------------
2023-10-17 08:35:21,203 epoch 3 - iter 13/138 - loss 0.10718037 - time (sec): 0.78 - samples/sec: 2872.78 - lr: 0.000044 - momentum: 0.000000
2023-10-17 08:35:21,944 epoch 3 - iter 26/138 - loss 0.09491367 - time (sec): 1.52 - samples/sec: 2859.72 - lr: 0.000043 - momentum: 0.000000
2023-10-17 08:35:22,667 epoch 3 - iter 39/138 - loss 0.10218904 - time (sec): 2.24 - samples/sec: 2925.36 - lr: 0.000043 - momentum: 0.000000
2023-10-17 08:35:23,382 epoch 3 - iter 52/138 - loss 0.11929519 - time (sec): 2.95 - samples/sec: 2918.24 - lr: 0.000042 - momentum: 0.000000
2023-10-17 08:35:24,120 epoch 3 - iter 65/138 - loss 0.10689042 - time (sec): 3.69 - samples/sec: 2929.70 - lr: 0.000042 - momentum: 0.000000
2023-10-17 08:35:24,820 epoch 3 - iter 78/138 - loss 0.10328288 - time (sec): 4.39 - samples/sec: 2895.41 - lr: 0.000041 - momentum: 0.000000
2023-10-17 08:35:25,584 epoch 3 - iter 91/138 - loss 0.10884179 - time (sec): 5.16 - samples/sec: 2943.16 - lr: 0.000041 - momentum: 0.000000
2023-10-17 08:35:26,296 epoch 3 - iter 104/138 - loss 0.10748961 - time (sec): 5.87 - samples/sec: 2897.24 - lr: 0.000040 - momentum: 0.000000
2023-10-17 08:35:27,061 epoch 3 - iter 117/138 - loss 0.10155459 - time (sec): 6.63 - samples/sec: 2914.05 - lr: 0.000040 - momentum: 0.000000
2023-10-17 08:35:27,789 epoch 3 - iter 130/138 - loss 0.10365295 - time (sec): 7.36 - samples/sec: 2909.58 - lr: 0.000039 - momentum: 0.000000
2023-10-17 08:35:28,255 ----------------------------------------------------------------------------------------------------
2023-10-17 08:35:28,255 EPOCH 3 done: loss 0.1017 - lr: 0.000039
2023-10-17 08:35:28,989 DEV : loss 0.1499420553445816 - f1-score (micro avg) 0.8242
2023-10-17 08:35:28,994 saving best model
2023-10-17 08:35:29,427 ----------------------------------------------------------------------------------------------------
2023-10-17 08:35:30,170 epoch 4 - iter 13/138 - loss 0.08215706 - time (sec): 0.74 - samples/sec: 2889.38 - lr: 0.000038 - momentum: 0.000000
2023-10-17 08:35:30,893 epoch 4 - iter 26/138 - loss 0.09423860 - time (sec): 1.46 - samples/sec: 2903.96 - lr: 0.000038 - momentum: 0.000000
2023-10-17 08:35:31,658 epoch 4 - iter 39/138 - loss 0.07897657 - time (sec): 2.23 - samples/sec: 3006.49 - lr: 0.000037 - momentum: 0.000000
2023-10-17 08:35:32,376 epoch 4 - iter 52/138 - loss 0.07584959 - time (sec): 2.95 - samples/sec: 2960.47 - lr: 0.000037 - momentum: 0.000000
2023-10-17 08:35:33,139 epoch 4 - iter 65/138 - loss 0.07486259 - time (sec): 3.71 - samples/sec: 2962.15 - lr: 0.000036 - momentum: 0.000000
2023-10-17 08:35:33,888 epoch 4 - iter 78/138 - loss 0.07396912 - time (sec): 4.46 - samples/sec: 2963.46 - lr: 0.000036 - momentum: 0.000000
2023-10-17 08:35:34,650 epoch 4 - iter 91/138 - loss 0.07816740 - time (sec): 5.22 - samples/sec: 2944.53 - lr: 0.000035 - momentum: 0.000000
2023-10-17 08:35:35,356 epoch 4 - iter 104/138 - loss 0.07634711 - time (sec): 5.93 - samples/sec: 2908.55 - lr: 0.000035 - momentum: 0.000000
2023-10-17 08:35:36,161 epoch 4 - iter 117/138 - loss 0.07445357 - time (sec): 6.73 - samples/sec: 2928.58 - lr: 0.000034 - momentum: 0.000000
2023-10-17 08:35:36,888 epoch 4 - iter 130/138 - loss 0.07280530 - time (sec): 7.46 - samples/sec: 2896.65 - lr: 0.000034 - momentum: 0.000000
2023-10-17 08:35:37,316 ----------------------------------------------------------------------------------------------------
2023-10-17 08:35:37,316 EPOCH 4 done: loss 0.0765 - lr: 0.000034
2023-10-17 08:35:37,955 DEV : loss 0.142630472779274 - f1-score (micro avg) 0.8541
2023-10-17 08:35:37,960 saving best model
2023-10-17 08:35:38,389 ----------------------------------------------------------------------------------------------------
2023-10-17 08:35:39,240 epoch 5 - iter 13/138 - loss 0.03121562 - time (sec): 0.85 - samples/sec: 2885.84 - lr: 0.000033 - momentum: 0.000000
2023-10-17 08:35:39,959 epoch 5 - iter 26/138 - loss 0.04200227 - time (sec): 1.57 - samples/sec: 2777.97 - lr: 0.000032 - momentum: 0.000000
2023-10-17 08:35:40,670 epoch 5 - iter 39/138 - loss 0.05399254 - time (sec): 2.28 - samples/sec: 2865.47 - lr: 0.000032 - momentum: 0.000000
2023-10-17 08:35:41,438 epoch 5 - iter 52/138 - loss 0.04714869 - time (sec): 3.05 - samples/sec: 2812.70 - lr: 0.000031 - momentum: 0.000000
2023-10-17 08:35:42,181 epoch 5 - iter 65/138 - loss 0.04556329 - time (sec): 3.79 - samples/sec: 2818.02 - lr: 0.000031 - momentum: 0.000000
2023-10-17 08:35:42,936 epoch 5 - iter 78/138 - loss 0.04860871 - time (sec): 4.54 - samples/sec: 2832.52 - lr: 0.000030 - momentum: 0.000000
2023-10-17 08:35:43,652 epoch 5 - iter 91/138 - loss 0.05073703 - time (sec): 5.26 - samples/sec: 2875.50 - lr: 0.000030 - momentum: 0.000000
2023-10-17 08:35:44,408 epoch 5 - iter 104/138 - loss 0.05067376 - time (sec): 6.02 - samples/sec: 2880.98 - lr: 0.000029 - momentum: 0.000000
2023-10-17 08:35:45,141 epoch 5 - iter 117/138 - loss 0.04980472 - time (sec): 6.75 - samples/sec: 2888.64 - lr: 0.000029 - momentum: 0.000000
2023-10-17 08:35:45,858 epoch 5 - iter 130/138 - loss 0.05156968 - time (sec): 7.47 - samples/sec: 2882.46 - lr: 0.000028 - momentum: 0.000000
2023-10-17 08:35:46,290 ----------------------------------------------------------------------------------------------------
2023-10-17 08:35:46,290 EPOCH 5 done: loss 0.0544 - lr: 0.000028
2023-10-17 08:35:46,941 DEV : loss 0.13915039598941803 - f1-score (micro avg) 0.8685
2023-10-17 08:35:46,946 saving best model
2023-10-17 08:35:47,394 ----------------------------------------------------------------------------------------------------
2023-10-17 08:35:48,156 epoch 6 - iter 13/138 - loss 0.05894751 - time (sec): 0.76 - samples/sec: 2914.36 - lr: 0.000027 - momentum: 0.000000
2023-10-17 08:35:48,874 epoch 6 - iter 26/138 - loss 0.04072450 - time (sec): 1.48 - samples/sec: 2788.36 - lr: 0.000027 - momentum: 0.000000
2023-10-17 08:35:49,620 epoch 6 - iter 39/138 - loss 0.03586809 - time (sec): 2.22 - samples/sec: 2841.81 - lr: 0.000026 - momentum: 0.000000
2023-10-17 08:35:50,366 epoch 6 - iter 52/138 - loss 0.03465352 - time (sec): 2.97 - samples/sec: 2859.07 - lr: 0.000026 - momentum: 0.000000
2023-10-17 08:35:51,104 epoch 6 - iter 65/138 - loss 0.03757808 - time (sec): 3.71 - samples/sec: 2857.92 - lr: 0.000025 - momentum: 0.000000
2023-10-17 08:35:51,890 epoch 6 - iter 78/138 - loss 0.04046940 - time (sec): 4.49 - samples/sec: 2856.53 - lr: 0.000025 - momentum: 0.000000
2023-10-17 08:35:52,701 epoch 6 - iter 91/138 - loss 0.04562090 - time (sec): 5.30 - samples/sec: 2894.15 - lr: 0.000024 - momentum: 0.000000
2023-10-17 08:35:53,438 epoch 6 - iter 104/138 - loss 0.04308195 - time (sec): 6.04 - samples/sec: 2893.39 - lr: 0.000024 - momentum: 0.000000
2023-10-17 08:35:54,134 epoch 6 - iter 117/138 - loss 0.04275599 - time (sec): 6.74 - samples/sec: 2889.34 - lr: 0.000023 - momentum: 0.000000
2023-10-17 08:35:54,891 epoch 6 - iter 130/138 - loss 0.04058690 - time (sec): 7.49 - samples/sec: 2882.52 - lr: 0.000023 - momentum: 0.000000
2023-10-17 08:35:55,346 ----------------------------------------------------------------------------------------------------
2023-10-17 08:35:55,347 EPOCH 6 done: loss 0.0403 - lr: 0.000023
2023-10-17 08:35:56,007 DEV : loss 0.17933738231658936 - f1-score (micro avg) 0.8634
2023-10-17 08:35:56,011 ----------------------------------------------------------------------------------------------------
2023-10-17 08:35:56,740 epoch 7 - iter 13/138 - loss 0.05558773 - time (sec): 0.73 - samples/sec: 3158.89 - lr: 0.000022 - momentum: 0.000000
2023-10-17 08:35:57,500 epoch 7 - iter 26/138 - loss 0.03701149 - time (sec): 1.49 - samples/sec: 3057.72 - lr: 0.000021 - momentum: 0.000000
2023-10-17 08:35:58,203 epoch 7 - iter 39/138 - loss 0.03583069 - time (sec): 2.19 - samples/sec: 3014.45 - lr: 0.000021 - momentum: 0.000000
2023-10-17 08:35:58,941 epoch 7 - iter 52/138 - loss 0.04594578 - time (sec): 2.93 - samples/sec: 2981.36 - lr: 0.000020 - momentum: 0.000000
2023-10-17 08:35:59,694 epoch 7 - iter 65/138 - loss 0.04030116 - time (sec): 3.68 - samples/sec: 2988.98 - lr: 0.000020 - momentum: 0.000000
2023-10-17 08:36:00,421 epoch 7 - iter 78/138 - loss 0.04106748 - time (sec): 4.41 - samples/sec: 2950.01 - lr: 0.000019 - momentum: 0.000000
2023-10-17 08:36:01,180 epoch 7 - iter 91/138 - loss 0.03909668 - time (sec): 5.17 - samples/sec: 2921.79 - lr: 0.000019 - momentum: 0.000000
2023-10-17 08:36:02,029 epoch 7 - iter 104/138 - loss 0.03806689 - time (sec): 6.02 - samples/sec: 2870.22 - lr: 0.000018 - momentum: 0.000000
2023-10-17 08:36:02,737 epoch 7 - iter 117/138 - loss 0.03591500 - time (sec): 6.72 - samples/sec: 2857.58 - lr: 0.000018 - momentum: 0.000000
2023-10-17 08:36:03,492 epoch 7 - iter 130/138 - loss 0.03447142 - time (sec): 7.48 - samples/sec: 2876.46 - lr: 0.000017 - momentum: 0.000000
2023-10-17 08:36:03,968 ----------------------------------------------------------------------------------------------------
2023-10-17 08:36:03,968 EPOCH 7 done: loss 0.0336 - lr: 0.000017
2023-10-17 08:36:04,614 DEV : loss 0.1858513355255127 - f1-score (micro avg) 0.8633
2023-10-17 08:36:04,619 ----------------------------------------------------------------------------------------------------
2023-10-17 08:36:05,347 epoch 8 - iter 13/138 - loss 0.01159785 - time (sec): 0.73 - samples/sec: 2930.99 - lr: 0.000016 - momentum: 0.000000
2023-10-17 08:36:06,107 epoch 8 - iter 26/138 - loss 0.00748875 - time (sec): 1.49 - samples/sec: 2819.78 - lr: 0.000016 - momentum: 0.000000
2023-10-17 08:36:06,828 epoch 8 - iter 39/138 - loss 0.00977630 - time (sec): 2.21 - samples/sec: 2846.32 - lr: 0.000015 - momentum: 0.000000
2023-10-17 08:36:07,619 epoch 8 - iter 52/138 - loss 0.01032406 - time (sec): 3.00 - samples/sec: 2884.06 - lr: 0.000015 - momentum: 0.000000
2023-10-17 08:36:08,355 epoch 8 - iter 65/138 - loss 0.01045180 - time (sec): 3.74 - samples/sec: 2905.72 - lr: 0.000014 - momentum: 0.000000
2023-10-17 08:36:09,149 epoch 8 - iter 78/138 - loss 0.01407818 - time (sec): 4.53 - samples/sec: 2886.56 - lr: 0.000014 - momentum: 0.000000
2023-10-17 08:36:09,892 epoch 8 - iter 91/138 - loss 0.01448040 - time (sec): 5.27 - samples/sec: 2862.11 - lr: 0.000013 - momentum: 0.000000
2023-10-17 08:36:10,630 epoch 8 - iter 104/138 - loss 0.01703694 - time (sec): 6.01 - samples/sec: 2872.26 - lr: 0.000013 - momentum: 0.000000
2023-10-17 08:36:11,389 epoch 8 - iter 117/138 - loss 0.02100446 - time (sec): 6.77 - samples/sec: 2862.47 - lr: 0.000012 - momentum: 0.000000
2023-10-17 08:36:12,097 epoch 8 - iter 130/138 - loss 0.02527052 - time (sec): 7.48 - samples/sec: 2866.84 - lr: 0.000012 - momentum: 0.000000
2023-10-17 08:36:12,579 ----------------------------------------------------------------------------------------------------
2023-10-17 08:36:12,579 EPOCH 8 done: loss 0.0251 - lr: 0.000012
2023-10-17 08:36:13,258 DEV : loss 0.18673621118068695 - f1-score (micro avg) 0.8619
2023-10-17 08:36:13,263 ----------------------------------------------------------------------------------------------------
2023-10-17 08:36:14,055 epoch 9 - iter 13/138 - loss 0.00286938 - time (sec): 0.79 - samples/sec: 2623.46 - lr: 0.000011 - momentum: 0.000000
2023-10-17 08:36:14,786 epoch 9 - iter 26/138 - loss 0.00516397 - time (sec): 1.52 - samples/sec: 2730.28 - lr: 0.000010 - momentum: 0.000000
2023-10-17 08:36:15,547 epoch 9 - iter 39/138 - loss 0.00459760 - time (sec): 2.28 - samples/sec: 2703.24 - lr: 0.000010 - momentum: 0.000000
2023-10-17 08:36:16,357 epoch 9 - iter 52/138 - loss 0.01948120 - time (sec): 3.09 - samples/sec: 2770.89 - lr: 0.000009 - momentum: 0.000000
2023-10-17 08:36:17,118 epoch 9 - iter 65/138 - loss 0.01938758 - time (sec): 3.85 - samples/sec: 2749.47 - lr: 0.000009 - momentum: 0.000000
2023-10-17 08:36:17,943 epoch 9 - iter 78/138 - loss 0.02191297 - time (sec): 4.68 - samples/sec: 2751.11 - lr: 0.000008 - momentum: 0.000000
2023-10-17 08:36:18,732 epoch 9 - iter 91/138 - loss 0.01910025 - time (sec): 5.47 - samples/sec: 2751.04 - lr: 0.000008 - momentum: 0.000000
2023-10-17 08:36:19,491 epoch 9 - iter 104/138 - loss 0.01905430 - time (sec): 6.23 - samples/sec: 2741.24 - lr: 0.000007 - momentum: 0.000000
2023-10-17 08:36:20,292 epoch 9 - iter 117/138 - loss 0.01906412 - time (sec): 7.03 - samples/sec: 2747.88 - lr: 0.000007 - momentum: 0.000000
2023-10-17 08:36:21,034 epoch 9 - iter 130/138 - loss 0.02012627 - time (sec): 7.77 - samples/sec: 2772.24 - lr: 0.000006 - momentum: 0.000000
2023-10-17 08:36:21,458 ----------------------------------------------------------------------------------------------------
2023-10-17 08:36:21,458 EPOCH 9 done: loss 0.0191 - lr: 0.000006
2023-10-17 08:36:22,181 DEV : loss 0.20274551212787628 - f1-score (micro avg) 0.8558
2023-10-17 08:36:22,186 ----------------------------------------------------------------------------------------------------
2023-10-17 08:36:22,923 epoch 10 - iter 13/138 - loss 0.00037874 - time (sec): 0.74 - samples/sec: 2816.38 - lr: 0.000005 - momentum: 0.000000
2023-10-17 08:36:23,701 epoch 10 - iter 26/138 - loss 0.00164556 - time (sec): 1.51 - samples/sec: 2860.71 - lr: 0.000005 - momentum: 0.000000
2023-10-17 08:36:24,464 epoch 10 - iter 39/138 - loss 0.00401456 - time (sec): 2.28 - samples/sec: 2919.05 - lr: 0.000004 - momentum: 0.000000
2023-10-17 08:36:25,258 epoch 10 - iter 52/138 - loss 0.01412709 - time (sec): 3.07 - samples/sec: 2880.68 - lr: 0.000004 - momentum: 0.000000
2023-10-17 08:36:25,995 epoch 10 - iter 65/138 - loss 0.01577883 - time (sec): 3.81 - samples/sec: 2837.15 - lr: 0.000003 - momentum: 0.000000
2023-10-17 08:36:26,765 epoch 10 - iter 78/138 - loss 0.01358062 - time (sec): 4.58 - samples/sec: 2842.94 - lr: 0.000003 - momentum: 0.000000
2023-10-17 08:36:27,536 epoch 10 - iter 91/138 - loss 0.01388950 - time (sec): 5.35 - samples/sec: 2848.01 - lr: 0.000002 - momentum: 0.000000
2023-10-17 08:36:28,287 epoch 10 - iter 104/138 - loss 0.01366988 - time (sec): 6.10 - samples/sec: 2867.03 - lr: 0.000002 - momentum: 0.000000
2023-10-17 08:36:29,019 epoch 10 - iter 117/138 - loss 0.01421039 - time (sec): 6.83 - samples/sec: 2841.28 - lr: 0.000001 - momentum: 0.000000
2023-10-17 08:36:29,736 epoch 10 - iter 130/138 - loss 0.01533784 - time (sec): 7.55 - samples/sec: 2839.22 - lr: 0.000000 - momentum: 0.000000
2023-10-17 08:36:30,198 ----------------------------------------------------------------------------------------------------
2023-10-17 08:36:30,198 EPOCH 10 done: loss 0.0146 - lr: 0.000000
2023-10-17 08:36:30,901 DEV : loss 0.20630405843257904 - f1-score (micro avg) 0.8599
2023-10-17 08:36:31,252 ----------------------------------------------------------------------------------------------------
2023-10-17 08:36:31,253 Loading model from best epoch ...
2023-10-17 08:36:32,566 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
2023-10-17 08:36:33,363
Results:
- F-score (micro) 0.886
- F-score (macro) 0.6312
- Accuracy 0.8067
By class:
precision recall f1-score support
scope 0.8851 0.8750 0.8800 176
pers 0.9677 0.9375 0.9524 128
work 0.7975 0.8514 0.8235 74
object 0.0000 0.0000 0.0000 2
loc 0.5000 0.5000 0.5000 2
micro avg 0.8871 0.8848 0.8860 382
macro avg 0.6301 0.6328 0.6312 382
weighted avg 0.8891 0.8848 0.8867 382
2023-10-17 08:36:33,363 ----------------------------------------------------------------------------------------------------
|