File size: 23,947 Bytes
6f4a369 |
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 |
2023-10-17 10:30:46,480 ----------------------------------------------------------------------------------------------------
2023-10-17 10:30:46,481 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 10:30:46,481 ----------------------------------------------------------------------------------------------------
2023-10-17 10:30:46,481 MultiCorpus: 966 train + 219 dev + 204 test sentences
- NER_HIPE_2022 Corpus: 966 train + 219 dev + 204 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/ajmc/fr/with_doc_seperator
2023-10-17 10:30:46,481 ----------------------------------------------------------------------------------------------------
2023-10-17 10:30:46,481 Train: 966 sentences
2023-10-17 10:30:46,481 (train_with_dev=False, train_with_test=False)
2023-10-17 10:30:46,481 ----------------------------------------------------------------------------------------------------
2023-10-17 10:30:46,481 Training Params:
2023-10-17 10:30:46,481 - learning_rate: "3e-05"
2023-10-17 10:30:46,481 - mini_batch_size: "8"
2023-10-17 10:30:46,482 - max_epochs: "10"
2023-10-17 10:30:46,482 - shuffle: "True"
2023-10-17 10:30:46,482 ----------------------------------------------------------------------------------------------------
2023-10-17 10:30:46,482 Plugins:
2023-10-17 10:30:46,482 - TensorboardLogger
2023-10-17 10:30:46,482 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 10:30:46,482 ----------------------------------------------------------------------------------------------------
2023-10-17 10:30:46,482 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 10:30:46,482 - metric: "('micro avg', 'f1-score')"
2023-10-17 10:30:46,482 ----------------------------------------------------------------------------------------------------
2023-10-17 10:30:46,482 Computation:
2023-10-17 10:30:46,482 - compute on device: cuda:0
2023-10-17 10:30:46,482 - embedding storage: none
2023-10-17 10:30:46,482 ----------------------------------------------------------------------------------------------------
2023-10-17 10:30:46,482 Model training base path: "hmbench-ajmc/fr-hmteams/teams-base-historic-multilingual-discriminator-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-1"
2023-10-17 10:30:46,482 ----------------------------------------------------------------------------------------------------
2023-10-17 10:30:46,482 ----------------------------------------------------------------------------------------------------
2023-10-17 10:30:46,482 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 10:30:47,180 epoch 1 - iter 12/121 - loss 3.32014436 - time (sec): 0.70 - samples/sec: 3341.33 - lr: 0.000003 - momentum: 0.000000
2023-10-17 10:30:47,895 epoch 1 - iter 24/121 - loss 3.03698276 - time (sec): 1.41 - samples/sec: 3164.77 - lr: 0.000006 - momentum: 0.000000
2023-10-17 10:30:48,614 epoch 1 - iter 36/121 - loss 2.60289546 - time (sec): 2.13 - samples/sec: 3210.20 - lr: 0.000009 - momentum: 0.000000
2023-10-17 10:30:49,403 epoch 1 - iter 48/121 - loss 2.12575385 - time (sec): 2.92 - samples/sec: 3218.41 - lr: 0.000012 - momentum: 0.000000
2023-10-17 10:30:50,203 epoch 1 - iter 60/121 - loss 1.79218853 - time (sec): 3.72 - samples/sec: 3229.86 - lr: 0.000015 - momentum: 0.000000
2023-10-17 10:30:50,952 epoch 1 - iter 72/121 - loss 1.58367041 - time (sec): 4.47 - samples/sec: 3227.29 - lr: 0.000018 - momentum: 0.000000
2023-10-17 10:30:51,711 epoch 1 - iter 84/121 - loss 1.41113578 - time (sec): 5.23 - samples/sec: 3230.06 - lr: 0.000021 - momentum: 0.000000
2023-10-17 10:30:52,492 epoch 1 - iter 96/121 - loss 1.26321860 - time (sec): 6.01 - samples/sec: 3242.76 - lr: 0.000024 - momentum: 0.000000
2023-10-17 10:30:53,219 epoch 1 - iter 108/121 - loss 1.16515662 - time (sec): 6.74 - samples/sec: 3258.23 - lr: 0.000027 - momentum: 0.000000
2023-10-17 10:30:54,001 epoch 1 - iter 120/121 - loss 1.07050963 - time (sec): 7.52 - samples/sec: 3269.52 - lr: 0.000030 - momentum: 0.000000
2023-10-17 10:30:54,065 ----------------------------------------------------------------------------------------------------
2023-10-17 10:30:54,065 EPOCH 1 done: loss 1.0649 - lr: 0.000030
2023-10-17 10:30:54,663 DEV : loss 0.2681349813938141 - f1-score (micro avg) 0.4935
2023-10-17 10:30:54,684 saving best model
2023-10-17 10:30:55,080 ----------------------------------------------------------------------------------------------------
2023-10-17 10:30:55,820 epoch 2 - iter 12/121 - loss 0.25245481 - time (sec): 0.74 - samples/sec: 3337.64 - lr: 0.000030 - momentum: 0.000000
2023-10-17 10:30:56,535 epoch 2 - iter 24/121 - loss 0.26119176 - time (sec): 1.45 - samples/sec: 3133.98 - lr: 0.000029 - momentum: 0.000000
2023-10-17 10:30:57,237 epoch 2 - iter 36/121 - loss 0.24934766 - time (sec): 2.16 - samples/sec: 3221.62 - lr: 0.000029 - momentum: 0.000000
2023-10-17 10:30:58,019 epoch 2 - iter 48/121 - loss 0.23822482 - time (sec): 2.94 - samples/sec: 3309.63 - lr: 0.000029 - momentum: 0.000000
2023-10-17 10:30:58,819 epoch 2 - iter 60/121 - loss 0.22963477 - time (sec): 3.74 - samples/sec: 3281.18 - lr: 0.000028 - momentum: 0.000000
2023-10-17 10:30:59,537 epoch 2 - iter 72/121 - loss 0.22640817 - time (sec): 4.46 - samples/sec: 3325.09 - lr: 0.000028 - momentum: 0.000000
2023-10-17 10:31:00,284 epoch 2 - iter 84/121 - loss 0.21803280 - time (sec): 5.20 - samples/sec: 3297.21 - lr: 0.000028 - momentum: 0.000000
2023-10-17 10:31:01,073 epoch 2 - iter 96/121 - loss 0.21012624 - time (sec): 5.99 - samples/sec: 3292.82 - lr: 0.000027 - momentum: 0.000000
2023-10-17 10:31:01,889 epoch 2 - iter 108/121 - loss 0.20287535 - time (sec): 6.81 - samples/sec: 3265.45 - lr: 0.000027 - momentum: 0.000000
2023-10-17 10:31:02,631 epoch 2 - iter 120/121 - loss 0.19816081 - time (sec): 7.55 - samples/sec: 3253.85 - lr: 0.000027 - momentum: 0.000000
2023-10-17 10:31:02,683 ----------------------------------------------------------------------------------------------------
2023-10-17 10:31:02,684 EPOCH 2 done: loss 0.1970 - lr: 0.000027
2023-10-17 10:31:03,620 DEV : loss 0.14390070736408234 - f1-score (micro avg) 0.7865
2023-10-17 10:31:03,626 saving best model
2023-10-17 10:31:04,142 ----------------------------------------------------------------------------------------------------
2023-10-17 10:31:04,908 epoch 3 - iter 12/121 - loss 0.12851785 - time (sec): 0.76 - samples/sec: 3266.94 - lr: 0.000026 - momentum: 0.000000
2023-10-17 10:31:05,676 epoch 3 - iter 24/121 - loss 0.11801002 - time (sec): 1.53 - samples/sec: 3172.63 - lr: 0.000026 - momentum: 0.000000
2023-10-17 10:31:06,544 epoch 3 - iter 36/121 - loss 0.11883823 - time (sec): 2.40 - samples/sec: 3152.48 - lr: 0.000026 - momentum: 0.000000
2023-10-17 10:31:07,247 epoch 3 - iter 48/121 - loss 0.11584504 - time (sec): 3.10 - samples/sec: 3236.72 - lr: 0.000025 - momentum: 0.000000
2023-10-17 10:31:07,977 epoch 3 - iter 60/121 - loss 0.11598349 - time (sec): 3.83 - samples/sec: 3231.54 - lr: 0.000025 - momentum: 0.000000
2023-10-17 10:31:08,729 epoch 3 - iter 72/121 - loss 0.11195592 - time (sec): 4.59 - samples/sec: 3307.42 - lr: 0.000025 - momentum: 0.000000
2023-10-17 10:31:09,437 epoch 3 - iter 84/121 - loss 0.11244840 - time (sec): 5.29 - samples/sec: 3263.56 - lr: 0.000024 - momentum: 0.000000
2023-10-17 10:31:10,139 epoch 3 - iter 96/121 - loss 0.11154599 - time (sec): 6.00 - samples/sec: 3284.81 - lr: 0.000024 - momentum: 0.000000
2023-10-17 10:31:10,902 epoch 3 - iter 108/121 - loss 0.11047003 - time (sec): 6.76 - samples/sec: 3316.08 - lr: 0.000024 - momentum: 0.000000
2023-10-17 10:31:11,650 epoch 3 - iter 120/121 - loss 0.11039208 - time (sec): 7.51 - samples/sec: 3286.62 - lr: 0.000023 - momentum: 0.000000
2023-10-17 10:31:11,702 ----------------------------------------------------------------------------------------------------
2023-10-17 10:31:11,702 EPOCH 3 done: loss 0.1103 - lr: 0.000023
2023-10-17 10:31:12,489 DEV : loss 0.12160782516002655 - f1-score (micro avg) 0.8291
2023-10-17 10:31:12,496 saving best model
2023-10-17 10:31:13,042 ----------------------------------------------------------------------------------------------------
2023-10-17 10:31:13,802 epoch 4 - iter 12/121 - loss 0.11078327 - time (sec): 0.76 - samples/sec: 3345.15 - lr: 0.000023 - momentum: 0.000000
2023-10-17 10:31:14,541 epoch 4 - iter 24/121 - loss 0.08753515 - time (sec): 1.49 - samples/sec: 3287.89 - lr: 0.000023 - momentum: 0.000000
2023-10-17 10:31:15,310 epoch 4 - iter 36/121 - loss 0.07917905 - time (sec): 2.26 - samples/sec: 3283.48 - lr: 0.000022 - momentum: 0.000000
2023-10-17 10:31:16,076 epoch 4 - iter 48/121 - loss 0.08826692 - time (sec): 3.03 - samples/sec: 3202.60 - lr: 0.000022 - momentum: 0.000000
2023-10-17 10:31:16,818 epoch 4 - iter 60/121 - loss 0.08527749 - time (sec): 3.77 - samples/sec: 3296.24 - lr: 0.000022 - momentum: 0.000000
2023-10-17 10:31:17,630 epoch 4 - iter 72/121 - loss 0.07703033 - time (sec): 4.58 - samples/sec: 3254.92 - lr: 0.000021 - momentum: 0.000000
2023-10-17 10:31:18,439 epoch 4 - iter 84/121 - loss 0.07353993 - time (sec): 5.39 - samples/sec: 3231.75 - lr: 0.000021 - momentum: 0.000000
2023-10-17 10:31:19,176 epoch 4 - iter 96/121 - loss 0.08026535 - time (sec): 6.13 - samples/sec: 3235.43 - lr: 0.000021 - momentum: 0.000000
2023-10-17 10:31:19,945 epoch 4 - iter 108/121 - loss 0.07860544 - time (sec): 6.90 - samples/sec: 3210.15 - lr: 0.000020 - momentum: 0.000000
2023-10-17 10:31:20,747 epoch 4 - iter 120/121 - loss 0.07637764 - time (sec): 7.70 - samples/sec: 3188.34 - lr: 0.000020 - momentum: 0.000000
2023-10-17 10:31:20,801 ----------------------------------------------------------------------------------------------------
2023-10-17 10:31:20,801 EPOCH 4 done: loss 0.0759 - lr: 0.000020
2023-10-17 10:31:21,561 DEV : loss 0.15159755945205688 - f1-score (micro avg) 0.8256
2023-10-17 10:31:21,567 ----------------------------------------------------------------------------------------------------
2023-10-17 10:31:22,406 epoch 5 - iter 12/121 - loss 0.04424314 - time (sec): 0.84 - samples/sec: 3332.19 - lr: 0.000020 - momentum: 0.000000
2023-10-17 10:31:23,111 epoch 5 - iter 24/121 - loss 0.04608236 - time (sec): 1.54 - samples/sec: 3209.53 - lr: 0.000019 - momentum: 0.000000
2023-10-17 10:31:23,879 epoch 5 - iter 36/121 - loss 0.05514976 - time (sec): 2.31 - samples/sec: 3254.88 - lr: 0.000019 - momentum: 0.000000
2023-10-17 10:31:24,663 epoch 5 - iter 48/121 - loss 0.05155823 - time (sec): 3.10 - samples/sec: 3216.27 - lr: 0.000019 - momentum: 0.000000
2023-10-17 10:31:25,390 epoch 5 - iter 60/121 - loss 0.04998764 - time (sec): 3.82 - samples/sec: 3231.78 - lr: 0.000018 - momentum: 0.000000
2023-10-17 10:31:26,176 epoch 5 - iter 72/121 - loss 0.04908976 - time (sec): 4.61 - samples/sec: 3215.16 - lr: 0.000018 - momentum: 0.000000
2023-10-17 10:31:26,888 epoch 5 - iter 84/121 - loss 0.05123180 - time (sec): 5.32 - samples/sec: 3284.23 - lr: 0.000018 - momentum: 0.000000
2023-10-17 10:31:27,640 epoch 5 - iter 96/121 - loss 0.05213783 - time (sec): 6.07 - samples/sec: 3239.37 - lr: 0.000017 - momentum: 0.000000
2023-10-17 10:31:28,342 epoch 5 - iter 108/121 - loss 0.05143224 - time (sec): 6.77 - samples/sec: 3261.78 - lr: 0.000017 - momentum: 0.000000
2023-10-17 10:31:29,078 epoch 5 - iter 120/121 - loss 0.05380168 - time (sec): 7.51 - samples/sec: 3269.60 - lr: 0.000017 - momentum: 0.000000
2023-10-17 10:31:29,148 ----------------------------------------------------------------------------------------------------
2023-10-17 10:31:29,148 EPOCH 5 done: loss 0.0535 - lr: 0.000017
2023-10-17 10:31:29,934 DEV : loss 0.1807931661605835 - f1-score (micro avg) 0.8184
2023-10-17 10:31:29,939 ----------------------------------------------------------------------------------------------------
2023-10-17 10:31:30,664 epoch 6 - iter 12/121 - loss 0.02907978 - time (sec): 0.72 - samples/sec: 3666.62 - lr: 0.000016 - momentum: 0.000000
2023-10-17 10:31:31,448 epoch 6 - iter 24/121 - loss 0.03751196 - time (sec): 1.51 - samples/sec: 3403.32 - lr: 0.000016 - momentum: 0.000000
2023-10-17 10:31:32,212 epoch 6 - iter 36/121 - loss 0.03357465 - time (sec): 2.27 - samples/sec: 3379.66 - lr: 0.000016 - momentum: 0.000000
2023-10-17 10:31:32,943 epoch 6 - iter 48/121 - loss 0.03308335 - time (sec): 3.00 - samples/sec: 3306.25 - lr: 0.000015 - momentum: 0.000000
2023-10-17 10:31:33,743 epoch 6 - iter 60/121 - loss 0.03111144 - time (sec): 3.80 - samples/sec: 3262.25 - lr: 0.000015 - momentum: 0.000000
2023-10-17 10:31:34,478 epoch 6 - iter 72/121 - loss 0.03577265 - time (sec): 4.54 - samples/sec: 3274.54 - lr: 0.000015 - momentum: 0.000000
2023-10-17 10:31:35,208 epoch 6 - iter 84/121 - loss 0.03581138 - time (sec): 5.27 - samples/sec: 3276.39 - lr: 0.000014 - momentum: 0.000000
2023-10-17 10:31:35,984 epoch 6 - iter 96/121 - loss 0.03882524 - time (sec): 6.04 - samples/sec: 3271.90 - lr: 0.000014 - momentum: 0.000000
2023-10-17 10:31:36,683 epoch 6 - iter 108/121 - loss 0.04161598 - time (sec): 6.74 - samples/sec: 3285.34 - lr: 0.000014 - momentum: 0.000000
2023-10-17 10:31:37,436 epoch 6 - iter 120/121 - loss 0.04249901 - time (sec): 7.50 - samples/sec: 3284.29 - lr: 0.000013 - momentum: 0.000000
2023-10-17 10:31:37,483 ----------------------------------------------------------------------------------------------------
2023-10-17 10:31:37,483 EPOCH 6 done: loss 0.0428 - lr: 0.000013
2023-10-17 10:31:38,252 DEV : loss 0.1748889535665512 - f1-score (micro avg) 0.8253
2023-10-17 10:31:38,257 ----------------------------------------------------------------------------------------------------
2023-10-17 10:31:39,002 epoch 7 - iter 12/121 - loss 0.03329241 - time (sec): 0.74 - samples/sec: 3548.32 - lr: 0.000013 - momentum: 0.000000
2023-10-17 10:31:39,762 epoch 7 - iter 24/121 - loss 0.03701914 - time (sec): 1.50 - samples/sec: 3354.87 - lr: 0.000013 - momentum: 0.000000
2023-10-17 10:31:40,548 epoch 7 - iter 36/121 - loss 0.03906978 - time (sec): 2.29 - samples/sec: 3286.41 - lr: 0.000012 - momentum: 0.000000
2023-10-17 10:31:41,283 epoch 7 - iter 48/121 - loss 0.03915569 - time (sec): 3.02 - samples/sec: 3288.91 - lr: 0.000012 - momentum: 0.000000
2023-10-17 10:31:42,096 epoch 7 - iter 60/121 - loss 0.03641657 - time (sec): 3.84 - samples/sec: 3299.32 - lr: 0.000012 - momentum: 0.000000
2023-10-17 10:31:42,798 epoch 7 - iter 72/121 - loss 0.03452311 - time (sec): 4.54 - samples/sec: 3329.70 - lr: 0.000011 - momentum: 0.000000
2023-10-17 10:31:43,580 epoch 7 - iter 84/121 - loss 0.03223815 - time (sec): 5.32 - samples/sec: 3301.68 - lr: 0.000011 - momentum: 0.000000
2023-10-17 10:31:44,284 epoch 7 - iter 96/121 - loss 0.03201468 - time (sec): 6.03 - samples/sec: 3266.72 - lr: 0.000011 - momentum: 0.000000
2023-10-17 10:31:45,041 epoch 7 - iter 108/121 - loss 0.03162600 - time (sec): 6.78 - samples/sec: 3269.40 - lr: 0.000010 - momentum: 0.000000
2023-10-17 10:31:45,765 epoch 7 - iter 120/121 - loss 0.03052726 - time (sec): 7.51 - samples/sec: 3270.34 - lr: 0.000010 - momentum: 0.000000
2023-10-17 10:31:45,821 ----------------------------------------------------------------------------------------------------
2023-10-17 10:31:45,821 EPOCH 7 done: loss 0.0307 - lr: 0.000010
2023-10-17 10:31:46,595 DEV : loss 0.19148583710193634 - f1-score (micro avg) 0.8285
2023-10-17 10:31:46,602 ----------------------------------------------------------------------------------------------------
2023-10-17 10:31:47,350 epoch 8 - iter 12/121 - loss 0.02759122 - time (sec): 0.75 - samples/sec: 3163.70 - lr: 0.000010 - momentum: 0.000000
2023-10-17 10:31:48,073 epoch 8 - iter 24/121 - loss 0.01779458 - time (sec): 1.47 - samples/sec: 3283.46 - lr: 0.000009 - momentum: 0.000000
2023-10-17 10:31:48,882 epoch 8 - iter 36/121 - loss 0.01975104 - time (sec): 2.28 - samples/sec: 3250.84 - lr: 0.000009 - momentum: 0.000000
2023-10-17 10:31:49,674 epoch 8 - iter 48/121 - loss 0.02004782 - time (sec): 3.07 - samples/sec: 3265.70 - lr: 0.000009 - momentum: 0.000000
2023-10-17 10:31:50,435 epoch 8 - iter 60/121 - loss 0.02196078 - time (sec): 3.83 - samples/sec: 3274.88 - lr: 0.000008 - momentum: 0.000000
2023-10-17 10:31:51,149 epoch 8 - iter 72/121 - loss 0.02076302 - time (sec): 4.55 - samples/sec: 3326.58 - lr: 0.000008 - momentum: 0.000000
2023-10-17 10:31:51,896 epoch 8 - iter 84/121 - loss 0.02046546 - time (sec): 5.29 - samples/sec: 3305.81 - lr: 0.000008 - momentum: 0.000000
2023-10-17 10:31:52,592 epoch 8 - iter 96/121 - loss 0.02295288 - time (sec): 5.99 - samples/sec: 3273.69 - lr: 0.000008 - momentum: 0.000000
2023-10-17 10:31:53,434 epoch 8 - iter 108/121 - loss 0.02253970 - time (sec): 6.83 - samples/sec: 3288.26 - lr: 0.000007 - momentum: 0.000000
2023-10-17 10:31:54,157 epoch 8 - iter 120/121 - loss 0.02232507 - time (sec): 7.55 - samples/sec: 3256.21 - lr: 0.000007 - momentum: 0.000000
2023-10-17 10:31:54,209 ----------------------------------------------------------------------------------------------------
2023-10-17 10:31:54,209 EPOCH 8 done: loss 0.0222 - lr: 0.000007
2023-10-17 10:31:54,991 DEV : loss 0.20282834768295288 - f1-score (micro avg) 0.8335
2023-10-17 10:31:54,998 saving best model
2023-10-17 10:31:55,515 ----------------------------------------------------------------------------------------------------
2023-10-17 10:31:56,245 epoch 9 - iter 12/121 - loss 0.02835444 - time (sec): 0.73 - samples/sec: 3199.43 - lr: 0.000006 - momentum: 0.000000
2023-10-17 10:31:57,028 epoch 9 - iter 24/121 - loss 0.02810493 - time (sec): 1.51 - samples/sec: 3014.12 - lr: 0.000006 - momentum: 0.000000
2023-10-17 10:31:57,834 epoch 9 - iter 36/121 - loss 0.02511085 - time (sec): 2.32 - samples/sec: 3074.24 - lr: 0.000006 - momentum: 0.000000
2023-10-17 10:31:58,596 epoch 9 - iter 48/121 - loss 0.01989365 - time (sec): 3.08 - samples/sec: 3162.59 - lr: 0.000006 - momentum: 0.000000
2023-10-17 10:31:59,361 epoch 9 - iter 60/121 - loss 0.02370575 - time (sec): 3.84 - samples/sec: 3136.55 - lr: 0.000005 - momentum: 0.000000
2023-10-17 10:32:00,170 epoch 9 - iter 72/121 - loss 0.02151435 - time (sec): 4.65 - samples/sec: 3130.56 - lr: 0.000005 - momentum: 0.000000
2023-10-17 10:32:00,899 epoch 9 - iter 84/121 - loss 0.02058558 - time (sec): 5.38 - samples/sec: 3137.75 - lr: 0.000005 - momentum: 0.000000
2023-10-17 10:32:01,638 epoch 9 - iter 96/121 - loss 0.01989155 - time (sec): 6.12 - samples/sec: 3183.90 - lr: 0.000004 - momentum: 0.000000
2023-10-17 10:32:02,392 epoch 9 - iter 108/121 - loss 0.01893313 - time (sec): 6.88 - samples/sec: 3207.02 - lr: 0.000004 - momentum: 0.000000
2023-10-17 10:32:03,161 epoch 9 - iter 120/121 - loss 0.01778422 - time (sec): 7.64 - samples/sec: 3210.96 - lr: 0.000004 - momentum: 0.000000
2023-10-17 10:32:03,225 ----------------------------------------------------------------------------------------------------
2023-10-17 10:32:03,226 EPOCH 9 done: loss 0.0176 - lr: 0.000004
2023-10-17 10:32:04,012 DEV : loss 0.21387448906898499 - f1-score (micro avg) 0.8369
2023-10-17 10:32:04,018 saving best model
2023-10-17 10:32:04,521 ----------------------------------------------------------------------------------------------------
2023-10-17 10:32:05,226 epoch 10 - iter 12/121 - loss 0.00693211 - time (sec): 0.70 - samples/sec: 3366.89 - lr: 0.000003 - momentum: 0.000000
2023-10-17 10:32:05,952 epoch 10 - iter 24/121 - loss 0.00979826 - time (sec): 1.43 - samples/sec: 3436.56 - lr: 0.000003 - momentum: 0.000000
2023-10-17 10:32:06,680 epoch 10 - iter 36/121 - loss 0.00987826 - time (sec): 2.16 - samples/sec: 3403.89 - lr: 0.000003 - momentum: 0.000000
2023-10-17 10:32:07,502 epoch 10 - iter 48/121 - loss 0.00849950 - time (sec): 2.98 - samples/sec: 3324.42 - lr: 0.000002 - momentum: 0.000000
2023-10-17 10:32:08,354 epoch 10 - iter 60/121 - loss 0.00902092 - time (sec): 3.83 - samples/sec: 3253.92 - lr: 0.000002 - momentum: 0.000000
2023-10-17 10:32:09,127 epoch 10 - iter 72/121 - loss 0.00916563 - time (sec): 4.60 - samples/sec: 3237.50 - lr: 0.000002 - momentum: 0.000000
2023-10-17 10:32:09,874 epoch 10 - iter 84/121 - loss 0.00985326 - time (sec): 5.35 - samples/sec: 3243.53 - lr: 0.000001 - momentum: 0.000000
2023-10-17 10:32:10,679 epoch 10 - iter 96/121 - loss 0.01078985 - time (sec): 6.16 - samples/sec: 3250.04 - lr: 0.000001 - momentum: 0.000000
2023-10-17 10:32:11,422 epoch 10 - iter 108/121 - loss 0.01288405 - time (sec): 6.90 - samples/sec: 3244.72 - lr: 0.000001 - momentum: 0.000000
2023-10-17 10:32:12,146 epoch 10 - iter 120/121 - loss 0.01283307 - time (sec): 7.62 - samples/sec: 3234.37 - lr: 0.000000 - momentum: 0.000000
2023-10-17 10:32:12,195 ----------------------------------------------------------------------------------------------------
2023-10-17 10:32:12,195 EPOCH 10 done: loss 0.0128 - lr: 0.000000
2023-10-17 10:32:12,981 DEV : loss 0.21676737070083618 - f1-score (micro avg) 0.8396
2023-10-17 10:32:12,986 saving best model
2023-10-17 10:32:13,945 ----------------------------------------------------------------------------------------------------
2023-10-17 10:32:13,947 Loading model from best epoch ...
2023-10-17 10:32:15,322 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 10:32:16,199
Results:
- F-score (micro) 0.8093
- F-score (macro) 0.5585
- Accuracy 0.7005
By class:
precision recall f1-score support
pers 0.8207 0.8561 0.8380 139
scope 0.8248 0.8760 0.8496 129
work 0.7000 0.7875 0.7412 80
loc 1.0000 0.2222 0.3636 9
date 0.0000 0.0000 0.0000 3
micro avg 0.7941 0.8250 0.8093 360
macro avg 0.6691 0.5484 0.5585 360
weighted avg 0.7930 0.8250 0.8018 360
2023-10-17 10:32:16,199 ----------------------------------------------------------------------------------------------------
|