File size: 25,114 Bytes
5d676fc |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 |
2023-10-06 15:59:08,898 ----------------------------------------------------------------------------------------------------
2023-10-06 15:59:08,899 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): T5LayerNorm()
(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): T5LayerNorm()
(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): T5LayerNorm()
(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): T5LayerNorm()
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
(final_layer_norm): T5LayerNorm()
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=1472, out_features=25, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-06 15:59:08,899 ----------------------------------------------------------------------------------------------------
2023-10-06 15:59:08,899 MultiCorpus: 1214 train + 266 dev + 251 test sentences
- NER_HIPE_2022 Corpus: 1214 train + 266 dev + 251 test sentences - /app/.flair/datasets/ner_hipe_2022/v2.1/ajmc/en/with_doc_seperator
2023-10-06 15:59:08,899 ----------------------------------------------------------------------------------------------------
2023-10-06 15:59:08,899 Train: 1214 sentences
2023-10-06 15:59:08,899 (train_with_dev=False, train_with_test=False)
2023-10-06 15:59:08,899 ----------------------------------------------------------------------------------------------------
2023-10-06 15:59:08,900 Training Params:
2023-10-06 15:59:08,900 - learning_rate: "0.00015"
2023-10-06 15:59:08,900 - mini_batch_size: "8"
2023-10-06 15:59:08,900 - max_epochs: "10"
2023-10-06 15:59:08,900 - shuffle: "True"
2023-10-06 15:59:08,900 ----------------------------------------------------------------------------------------------------
2023-10-06 15:59:08,900 Plugins:
2023-10-06 15:59:08,900 - TensorboardLogger
2023-10-06 15:59:08,900 - LinearScheduler | warmup_fraction: '0.1'
2023-10-06 15:59:08,900 ----------------------------------------------------------------------------------------------------
2023-10-06 15:59:08,900 Final evaluation on model from best epoch (best-model.pt)
2023-10-06 15:59:08,900 - metric: "('micro avg', 'f1-score')"
2023-10-06 15:59:08,900 ----------------------------------------------------------------------------------------------------
2023-10-06 15:59:08,900 Computation:
2023-10-06 15:59:08,900 - compute on device: cuda:0
2023-10-06 15:59:08,900 - embedding storage: none
2023-10-06 15:59:08,901 ----------------------------------------------------------------------------------------------------
2023-10-06 15:59:08,901 Model training base path: "hmbench-ajmc/en-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs8-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-5"
2023-10-06 15:59:08,901 ----------------------------------------------------------------------------------------------------
2023-10-06 15:59:08,901 ----------------------------------------------------------------------------------------------------
2023-10-06 15:59:08,901 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-06 15:59:19,279 epoch 1 - iter 15/152 - loss 3.23553399 - time (sec): 10.38 - samples/sec: 308.10 - lr: 0.000014 - momentum: 0.000000
2023-10-06 15:59:29,286 epoch 1 - iter 30/152 - loss 3.22939813 - time (sec): 20.38 - samples/sec: 302.35 - lr: 0.000029 - momentum: 0.000000
2023-10-06 15:59:39,222 epoch 1 - iter 45/152 - loss 3.21751683 - time (sec): 30.32 - samples/sec: 295.52 - lr: 0.000043 - momentum: 0.000000
2023-10-06 15:59:49,336 epoch 1 - iter 60/152 - loss 3.19471768 - time (sec): 40.43 - samples/sec: 295.15 - lr: 0.000058 - momentum: 0.000000
2023-10-06 15:59:59,415 epoch 1 - iter 75/152 - loss 3.14939917 - time (sec): 50.51 - samples/sec: 294.80 - lr: 0.000073 - momentum: 0.000000
2023-10-06 16:00:10,582 epoch 1 - iter 90/152 - loss 3.07470578 - time (sec): 61.68 - samples/sec: 296.90 - lr: 0.000088 - momentum: 0.000000
2023-10-06 16:00:19,963 epoch 1 - iter 105/152 - loss 3.00882280 - time (sec): 71.06 - samples/sec: 293.71 - lr: 0.000103 - momentum: 0.000000
2023-10-06 16:00:30,508 epoch 1 - iter 120/152 - loss 2.91433039 - time (sec): 81.61 - samples/sec: 294.71 - lr: 0.000117 - momentum: 0.000000
2023-10-06 16:00:41,106 epoch 1 - iter 135/152 - loss 2.81522670 - time (sec): 92.20 - samples/sec: 295.26 - lr: 0.000132 - momentum: 0.000000
2023-10-06 16:00:52,076 epoch 1 - iter 150/152 - loss 2.71049372 - time (sec): 103.17 - samples/sec: 296.29 - lr: 0.000147 - momentum: 0.000000
2023-10-06 16:00:53,455 ----------------------------------------------------------------------------------------------------
2023-10-06 16:00:53,455 EPOCH 1 done: loss 2.6965 - lr: 0.000147
2023-10-06 16:01:00,771 DEV : loss 1.6262575387954712 - f1-score (micro avg) 0.0
2023-10-06 16:01:00,779 ----------------------------------------------------------------------------------------------------
2023-10-06 16:01:11,250 epoch 2 - iter 15/152 - loss 1.54970409 - time (sec): 10.47 - samples/sec: 292.66 - lr: 0.000148 - momentum: 0.000000
2023-10-06 16:01:21,614 epoch 2 - iter 30/152 - loss 1.43184541 - time (sec): 20.83 - samples/sec: 292.65 - lr: 0.000147 - momentum: 0.000000
2023-10-06 16:01:32,678 epoch 2 - iter 45/152 - loss 1.33590739 - time (sec): 31.90 - samples/sec: 292.03 - lr: 0.000145 - momentum: 0.000000
2023-10-06 16:01:43,716 epoch 2 - iter 60/152 - loss 1.22960476 - time (sec): 42.93 - samples/sec: 291.58 - lr: 0.000144 - momentum: 0.000000
2023-10-06 16:01:53,250 epoch 2 - iter 75/152 - loss 1.15504462 - time (sec): 52.47 - samples/sec: 287.33 - lr: 0.000142 - momentum: 0.000000
2023-10-06 16:02:03,759 epoch 2 - iter 90/152 - loss 1.07433988 - time (sec): 62.98 - samples/sec: 287.51 - lr: 0.000140 - momentum: 0.000000
2023-10-06 16:02:15,118 epoch 2 - iter 105/152 - loss 1.03695154 - time (sec): 74.34 - samples/sec: 284.18 - lr: 0.000139 - momentum: 0.000000
2023-10-06 16:02:26,195 epoch 2 - iter 120/152 - loss 0.99220094 - time (sec): 85.41 - samples/sec: 284.13 - lr: 0.000137 - momentum: 0.000000
2023-10-06 16:02:37,420 epoch 2 - iter 135/152 - loss 0.94715002 - time (sec): 96.64 - samples/sec: 284.38 - lr: 0.000135 - momentum: 0.000000
2023-10-06 16:02:48,400 epoch 2 - iter 150/152 - loss 0.90263114 - time (sec): 107.62 - samples/sec: 284.49 - lr: 0.000134 - momentum: 0.000000
2023-10-06 16:02:49,774 ----------------------------------------------------------------------------------------------------
2023-10-06 16:02:49,774 EPOCH 2 done: loss 0.9008 - lr: 0.000134
2023-10-06 16:02:57,798 DEV : loss 0.5528021454811096 - f1-score (micro avg) 0.0
2023-10-06 16:02:57,805 ----------------------------------------------------------------------------------------------------
2023-10-06 16:03:08,641 epoch 3 - iter 15/152 - loss 0.57289098 - time (sec): 10.83 - samples/sec: 266.65 - lr: 0.000132 - momentum: 0.000000
2023-10-06 16:03:19,591 epoch 3 - iter 30/152 - loss 0.49562157 - time (sec): 21.78 - samples/sec: 269.00 - lr: 0.000130 - momentum: 0.000000
2023-10-06 16:03:29,996 epoch 3 - iter 45/152 - loss 0.45703984 - time (sec): 32.19 - samples/sec: 267.32 - lr: 0.000129 - momentum: 0.000000
2023-10-06 16:03:41,830 epoch 3 - iter 60/152 - loss 0.44664831 - time (sec): 44.02 - samples/sec: 271.95 - lr: 0.000127 - momentum: 0.000000
2023-10-06 16:03:52,649 epoch 3 - iter 75/152 - loss 0.42891430 - time (sec): 54.84 - samples/sec: 271.54 - lr: 0.000125 - momentum: 0.000000
2023-10-06 16:04:03,823 epoch 3 - iter 90/152 - loss 0.42302343 - time (sec): 66.02 - samples/sec: 271.99 - lr: 0.000124 - momentum: 0.000000
2023-10-06 16:04:15,076 epoch 3 - iter 105/152 - loss 0.41046195 - time (sec): 77.27 - samples/sec: 273.02 - lr: 0.000122 - momentum: 0.000000
2023-10-06 16:04:26,291 epoch 3 - iter 120/152 - loss 0.39898527 - time (sec): 88.48 - samples/sec: 274.32 - lr: 0.000120 - momentum: 0.000000
2023-10-06 16:04:37,183 epoch 3 - iter 135/152 - loss 0.38534366 - time (sec): 99.38 - samples/sec: 274.77 - lr: 0.000119 - momentum: 0.000000
2023-10-06 16:04:48,644 epoch 3 - iter 150/152 - loss 0.37413664 - time (sec): 110.84 - samples/sec: 275.55 - lr: 0.000117 - momentum: 0.000000
2023-10-06 16:04:50,208 ----------------------------------------------------------------------------------------------------
2023-10-06 16:04:50,208 EPOCH 3 done: loss 0.3737 - lr: 0.000117
2023-10-06 16:04:58,144 DEV : loss 0.3319585919380188 - f1-score (micro avg) 0.507
2023-10-06 16:04:58,151 saving best model
2023-10-06 16:04:59,194 ----------------------------------------------------------------------------------------------------
2023-10-06 16:05:10,813 epoch 4 - iter 15/152 - loss 0.29376549 - time (sec): 11.62 - samples/sec: 272.54 - lr: 0.000115 - momentum: 0.000000
2023-10-06 16:05:21,224 epoch 4 - iter 30/152 - loss 0.27340139 - time (sec): 22.03 - samples/sec: 268.74 - lr: 0.000114 - momentum: 0.000000
2023-10-06 16:05:32,266 epoch 4 - iter 45/152 - loss 0.26581077 - time (sec): 33.07 - samples/sec: 270.88 - lr: 0.000112 - momentum: 0.000000
2023-10-06 16:05:43,392 epoch 4 - iter 60/152 - loss 0.25503646 - time (sec): 44.20 - samples/sec: 271.95 - lr: 0.000110 - momentum: 0.000000
2023-10-06 16:05:54,574 epoch 4 - iter 75/152 - loss 0.25391004 - time (sec): 55.38 - samples/sec: 272.89 - lr: 0.000109 - momentum: 0.000000
2023-10-06 16:06:05,579 epoch 4 - iter 90/152 - loss 0.24599003 - time (sec): 66.38 - samples/sec: 273.65 - lr: 0.000107 - momentum: 0.000000
2023-10-06 16:06:16,309 epoch 4 - iter 105/152 - loss 0.24005813 - time (sec): 77.11 - samples/sec: 273.35 - lr: 0.000105 - momentum: 0.000000
2023-10-06 16:06:28,262 epoch 4 - iter 120/152 - loss 0.23682524 - time (sec): 89.07 - samples/sec: 276.48 - lr: 0.000104 - momentum: 0.000000
2023-10-06 16:06:39,989 epoch 4 - iter 135/152 - loss 0.22950166 - time (sec): 100.79 - samples/sec: 276.87 - lr: 0.000102 - momentum: 0.000000
2023-10-06 16:06:50,184 epoch 4 - iter 150/152 - loss 0.22413553 - time (sec): 110.99 - samples/sec: 276.17 - lr: 0.000101 - momentum: 0.000000
2023-10-06 16:06:51,440 ----------------------------------------------------------------------------------------------------
2023-10-06 16:06:51,440 EPOCH 4 done: loss 0.2235 - lr: 0.000101
2023-10-06 16:06:59,149 DEV : loss 0.22811780869960785 - f1-score (micro avg) 0.6674
2023-10-06 16:06:59,155 saving best model
2023-10-06 16:07:04,230 ----------------------------------------------------------------------------------------------------
2023-10-06 16:07:15,407 epoch 5 - iter 15/152 - loss 0.13651398 - time (sec): 11.18 - samples/sec: 275.15 - lr: 0.000099 - momentum: 0.000000
2023-10-06 16:07:26,193 epoch 5 - iter 30/152 - loss 0.16481221 - time (sec): 21.96 - samples/sec: 276.25 - lr: 0.000097 - momentum: 0.000000
2023-10-06 16:07:37,585 epoch 5 - iter 45/152 - loss 0.17014447 - time (sec): 33.35 - samples/sec: 277.42 - lr: 0.000095 - momentum: 0.000000
2023-10-06 16:07:48,939 epoch 5 - iter 60/152 - loss 0.16586242 - time (sec): 44.71 - samples/sec: 277.72 - lr: 0.000094 - momentum: 0.000000
2023-10-06 16:08:00,314 epoch 5 - iter 75/152 - loss 0.16777109 - time (sec): 56.08 - samples/sec: 278.40 - lr: 0.000092 - momentum: 0.000000
2023-10-06 16:08:10,997 epoch 5 - iter 90/152 - loss 0.15877418 - time (sec): 66.77 - samples/sec: 277.78 - lr: 0.000091 - momentum: 0.000000
2023-10-06 16:08:21,997 epoch 5 - iter 105/152 - loss 0.15242023 - time (sec): 77.77 - samples/sec: 279.60 - lr: 0.000089 - momentum: 0.000000
2023-10-06 16:08:32,507 epoch 5 - iter 120/152 - loss 0.15341168 - time (sec): 88.28 - samples/sec: 282.14 - lr: 0.000087 - momentum: 0.000000
2023-10-06 16:08:42,899 epoch 5 - iter 135/152 - loss 0.15092835 - time (sec): 98.67 - samples/sec: 282.27 - lr: 0.000086 - momentum: 0.000000
2023-10-06 16:08:52,951 epoch 5 - iter 150/152 - loss 0.15021431 - time (sec): 108.72 - samples/sec: 282.95 - lr: 0.000084 - momentum: 0.000000
2023-10-06 16:08:53,974 ----------------------------------------------------------------------------------------------------
2023-10-06 16:08:53,975 EPOCH 5 done: loss 0.1494 - lr: 0.000084
2023-10-06 16:09:01,309 DEV : loss 0.17536617815494537 - f1-score (micro avg) 0.6989
2023-10-06 16:09:01,317 saving best model
2023-10-06 16:09:05,624 ----------------------------------------------------------------------------------------------------
2023-10-06 16:09:15,972 epoch 6 - iter 15/152 - loss 0.14040234 - time (sec): 10.35 - samples/sec: 307.74 - lr: 0.000082 - momentum: 0.000000
2023-10-06 16:09:26,692 epoch 6 - iter 30/152 - loss 0.12875256 - time (sec): 21.07 - samples/sec: 305.60 - lr: 0.000080 - momentum: 0.000000
2023-10-06 16:09:37,492 epoch 6 - iter 45/152 - loss 0.12395352 - time (sec): 31.87 - samples/sec: 305.49 - lr: 0.000079 - momentum: 0.000000
2023-10-06 16:09:48,334 epoch 6 - iter 60/152 - loss 0.11736335 - time (sec): 42.71 - samples/sec: 300.74 - lr: 0.000077 - momentum: 0.000000
2023-10-06 16:09:58,601 epoch 6 - iter 75/152 - loss 0.11162072 - time (sec): 52.98 - samples/sec: 300.53 - lr: 0.000076 - momentum: 0.000000
2023-10-06 16:10:08,041 epoch 6 - iter 90/152 - loss 0.10797765 - time (sec): 62.42 - samples/sec: 297.76 - lr: 0.000074 - momentum: 0.000000
2023-10-06 16:10:18,903 epoch 6 - iter 105/152 - loss 0.10655617 - time (sec): 73.28 - samples/sec: 297.96 - lr: 0.000072 - momentum: 0.000000
2023-10-06 16:10:28,818 epoch 6 - iter 120/152 - loss 0.10611263 - time (sec): 83.19 - samples/sec: 296.81 - lr: 0.000071 - momentum: 0.000000
2023-10-06 16:10:39,027 epoch 6 - iter 135/152 - loss 0.10372822 - time (sec): 93.40 - samples/sec: 295.53 - lr: 0.000069 - momentum: 0.000000
2023-10-06 16:10:49,505 epoch 6 - iter 150/152 - loss 0.10463535 - time (sec): 103.88 - samples/sec: 295.31 - lr: 0.000067 - momentum: 0.000000
2023-10-06 16:10:50,622 ----------------------------------------------------------------------------------------------------
2023-10-06 16:10:50,623 EPOCH 6 done: loss 0.1058 - lr: 0.000067
2023-10-06 16:10:57,871 DEV : loss 0.15567055344581604 - f1-score (micro avg) 0.8195
2023-10-06 16:10:57,879 saving best model
2023-10-06 16:11:02,242 ----------------------------------------------------------------------------------------------------
2023-10-06 16:11:12,218 epoch 7 - iter 15/152 - loss 0.07733067 - time (sec): 9.97 - samples/sec: 287.52 - lr: 0.000066 - momentum: 0.000000
2023-10-06 16:11:22,182 epoch 7 - iter 30/152 - loss 0.08154846 - time (sec): 19.94 - samples/sec: 288.58 - lr: 0.000064 - momentum: 0.000000
2023-10-06 16:11:32,564 epoch 7 - iter 45/152 - loss 0.07207774 - time (sec): 30.32 - samples/sec: 288.91 - lr: 0.000062 - momentum: 0.000000
2023-10-06 16:11:43,021 epoch 7 - iter 60/152 - loss 0.06814567 - time (sec): 40.78 - samples/sec: 290.09 - lr: 0.000061 - momentum: 0.000000
2023-10-06 16:11:53,448 epoch 7 - iter 75/152 - loss 0.06892973 - time (sec): 51.20 - samples/sec: 289.35 - lr: 0.000059 - momentum: 0.000000
2023-10-06 16:12:04,408 epoch 7 - iter 90/152 - loss 0.07718801 - time (sec): 62.17 - samples/sec: 291.18 - lr: 0.000057 - momentum: 0.000000
2023-10-06 16:12:15,055 epoch 7 - iter 105/152 - loss 0.07474215 - time (sec): 72.81 - samples/sec: 291.20 - lr: 0.000056 - momentum: 0.000000
2023-10-06 16:12:25,644 epoch 7 - iter 120/152 - loss 0.07507597 - time (sec): 83.40 - samples/sec: 289.76 - lr: 0.000054 - momentum: 0.000000
2023-10-06 16:12:36,817 epoch 7 - iter 135/152 - loss 0.07591231 - time (sec): 94.57 - samples/sec: 290.83 - lr: 0.000052 - momentum: 0.000000
2023-10-06 16:12:47,737 epoch 7 - iter 150/152 - loss 0.08084531 - time (sec): 105.49 - samples/sec: 290.18 - lr: 0.000051 - momentum: 0.000000
2023-10-06 16:12:49,027 ----------------------------------------------------------------------------------------------------
2023-10-06 16:12:49,028 EPOCH 7 done: loss 0.0802 - lr: 0.000051
2023-10-06 16:12:56,660 DEV : loss 0.14975833892822266 - f1-score (micro avg) 0.8255
2023-10-06 16:12:56,667 saving best model
2023-10-06 16:13:01,000 ----------------------------------------------------------------------------------------------------
2023-10-06 16:13:11,751 epoch 8 - iter 15/152 - loss 0.08031097 - time (sec): 10.75 - samples/sec: 281.79 - lr: 0.000049 - momentum: 0.000000
2023-10-06 16:13:22,627 epoch 8 - iter 30/152 - loss 0.08197433 - time (sec): 21.62 - samples/sec: 276.26 - lr: 0.000047 - momentum: 0.000000
2023-10-06 16:13:32,872 epoch 8 - iter 45/152 - loss 0.07403169 - time (sec): 31.87 - samples/sec: 272.89 - lr: 0.000046 - momentum: 0.000000
2023-10-06 16:13:43,876 epoch 8 - iter 60/152 - loss 0.07606961 - time (sec): 42.87 - samples/sec: 274.59 - lr: 0.000044 - momentum: 0.000000
2023-10-06 16:13:55,223 epoch 8 - iter 75/152 - loss 0.07054570 - time (sec): 54.22 - samples/sec: 276.43 - lr: 0.000042 - momentum: 0.000000
2023-10-06 16:14:06,219 epoch 8 - iter 90/152 - loss 0.06846871 - time (sec): 65.22 - samples/sec: 275.27 - lr: 0.000041 - momentum: 0.000000
2023-10-06 16:14:17,729 epoch 8 - iter 105/152 - loss 0.06593960 - time (sec): 76.73 - samples/sec: 276.19 - lr: 0.000039 - momentum: 0.000000
2023-10-06 16:14:29,435 epoch 8 - iter 120/152 - loss 0.06549689 - time (sec): 88.43 - samples/sec: 278.16 - lr: 0.000037 - momentum: 0.000000
2023-10-06 16:14:40,711 epoch 8 - iter 135/152 - loss 0.06601897 - time (sec): 99.71 - samples/sec: 278.56 - lr: 0.000036 - momentum: 0.000000
2023-10-06 16:14:51,103 epoch 8 - iter 150/152 - loss 0.06589859 - time (sec): 110.10 - samples/sec: 277.69 - lr: 0.000034 - momentum: 0.000000
2023-10-06 16:14:52,590 ----------------------------------------------------------------------------------------------------
2023-10-06 16:14:52,590 EPOCH 8 done: loss 0.0657 - lr: 0.000034
2023-10-06 16:15:00,397 DEV : loss 0.14192526042461395 - f1-score (micro avg) 0.8298
2023-10-06 16:15:00,404 saving best model
2023-10-06 16:15:04,738 ----------------------------------------------------------------------------------------------------
2023-10-06 16:15:16,420 epoch 9 - iter 15/152 - loss 0.08055117 - time (sec): 11.68 - samples/sec: 289.30 - lr: 0.000032 - momentum: 0.000000
2023-10-06 16:15:27,670 epoch 9 - iter 30/152 - loss 0.07059158 - time (sec): 22.93 - samples/sec: 286.22 - lr: 0.000031 - momentum: 0.000000
2023-10-06 16:15:38,460 epoch 9 - iter 45/152 - loss 0.06701449 - time (sec): 33.72 - samples/sec: 282.38 - lr: 0.000029 - momentum: 0.000000
2023-10-06 16:15:49,085 epoch 9 - iter 60/152 - loss 0.06306260 - time (sec): 44.35 - samples/sec: 279.62 - lr: 0.000027 - momentum: 0.000000
2023-10-06 16:16:00,117 epoch 9 - iter 75/152 - loss 0.05895212 - time (sec): 55.38 - samples/sec: 276.85 - lr: 0.000026 - momentum: 0.000000
2023-10-06 16:16:11,413 epoch 9 - iter 90/152 - loss 0.05798036 - time (sec): 66.67 - samples/sec: 276.98 - lr: 0.000024 - momentum: 0.000000
2023-10-06 16:16:22,971 epoch 9 - iter 105/152 - loss 0.05724903 - time (sec): 78.23 - samples/sec: 277.15 - lr: 0.000022 - momentum: 0.000000
2023-10-06 16:16:33,774 epoch 9 - iter 120/152 - loss 0.05970107 - time (sec): 89.03 - samples/sec: 276.13 - lr: 0.000021 - momentum: 0.000000
2023-10-06 16:16:44,573 epoch 9 - iter 135/152 - loss 0.05612059 - time (sec): 99.83 - samples/sec: 276.36 - lr: 0.000019 - momentum: 0.000000
2023-10-06 16:16:55,558 epoch 9 - iter 150/152 - loss 0.05649651 - time (sec): 110.82 - samples/sec: 276.24 - lr: 0.000018 - momentum: 0.000000
2023-10-06 16:16:56,798 ----------------------------------------------------------------------------------------------------
2023-10-06 16:16:56,798 EPOCH 9 done: loss 0.0569 - lr: 0.000018
2023-10-06 16:17:04,762 DEV : loss 0.14784981310367584 - f1-score (micro avg) 0.826
2023-10-06 16:17:04,771 ----------------------------------------------------------------------------------------------------
2023-10-06 16:17:15,545 epoch 10 - iter 15/152 - loss 0.06202699 - time (sec): 10.77 - samples/sec: 266.79 - lr: 0.000016 - momentum: 0.000000
2023-10-06 16:17:27,709 epoch 10 - iter 30/152 - loss 0.05314995 - time (sec): 22.94 - samples/sec: 277.98 - lr: 0.000014 - momentum: 0.000000
2023-10-06 16:17:38,589 epoch 10 - iter 45/152 - loss 0.05329503 - time (sec): 33.82 - samples/sec: 276.99 - lr: 0.000012 - momentum: 0.000000
2023-10-06 16:17:49,528 epoch 10 - iter 60/152 - loss 0.05040656 - time (sec): 44.76 - samples/sec: 276.12 - lr: 0.000011 - momentum: 0.000000
2023-10-06 16:18:00,129 epoch 10 - iter 75/152 - loss 0.05349118 - time (sec): 55.36 - samples/sec: 273.26 - lr: 0.000009 - momentum: 0.000000
2023-10-06 16:18:11,286 epoch 10 - iter 90/152 - loss 0.05167885 - time (sec): 66.51 - samples/sec: 274.66 - lr: 0.000008 - momentum: 0.000000
2023-10-06 16:18:22,040 epoch 10 - iter 105/152 - loss 0.04974350 - time (sec): 77.27 - samples/sec: 272.88 - lr: 0.000006 - momentum: 0.000000
2023-10-06 16:18:33,569 epoch 10 - iter 120/152 - loss 0.05016859 - time (sec): 88.80 - samples/sec: 273.80 - lr: 0.000004 - momentum: 0.000000
2023-10-06 16:18:44,895 epoch 10 - iter 135/152 - loss 0.04954922 - time (sec): 100.12 - samples/sec: 274.62 - lr: 0.000003 - momentum: 0.000000
2023-10-06 16:18:56,220 epoch 10 - iter 150/152 - loss 0.05098316 - time (sec): 111.45 - samples/sec: 275.71 - lr: 0.000001 - momentum: 0.000000
2023-10-06 16:18:57,336 ----------------------------------------------------------------------------------------------------
2023-10-06 16:18:57,336 EPOCH 10 done: loss 0.0514 - lr: 0.000001
2023-10-06 16:19:05,303 DEV : loss 0.14607474207878113 - f1-score (micro avg) 0.8323
2023-10-06 16:19:05,311 saving best model
2023-10-06 16:19:10,518 ----------------------------------------------------------------------------------------------------
2023-10-06 16:19:10,519 Loading model from best epoch ...
2023-10-06 16:19:13,086 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-date, B-date, E-date, I-date, S-object, B-object, E-object, I-object
2023-10-06 16:19:20,434
Results:
- F-score (micro) 0.794
- F-score (macro) 0.4833
- Accuracy 0.6659
By class:
precision recall f1-score support
scope 0.7500 0.7947 0.7717 151
work 0.6949 0.8632 0.7700 95
pers 0.8125 0.9479 0.8750 96
loc 0.0000 0.0000 0.0000 3
date 0.0000 0.0000 0.0000 3
micro avg 0.7513 0.8420 0.7940 348
macro avg 0.4515 0.5212 0.4833 348
weighted avg 0.7393 0.8420 0.7864 348
2023-10-06 16:19:20,434 ----------------------------------------------------------------------------------------------------
|