File size: 25,034 Bytes
f0f7e81 |
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 |
2023-10-08 22:42:08,190 ----------------------------------------------------------------------------------------------------
2023-10-08 22:42:08,191 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()
)"
2023-10-08 22:42:08,192 ----------------------------------------------------------------------------------------------------
2023-10-08 22:42:08,192 MultiCorpus: 966 train + 219 dev + 204 test sentences
- NER_HIPE_2022 Corpus: 966 train + 219 dev + 204 test sentences - /app/.flair/datasets/ner_hipe_2022/v2.1/ajmc/fr/with_doc_seperator
2023-10-08 22:42:08,192 ----------------------------------------------------------------------------------------------------
2023-10-08 22:42:08,192 Train: 966 sentences
2023-10-08 22:42:08,192 (train_with_dev=False, train_with_test=False)
2023-10-08 22:42:08,192 ----------------------------------------------------------------------------------------------------
2023-10-08 22:42:08,192 Training Params:
2023-10-08 22:42:08,192 - learning_rate: "0.00016"
2023-10-08 22:42:08,192 - mini_batch_size: "8"
2023-10-08 22:42:08,192 - max_epochs: "10"
2023-10-08 22:42:08,192 - shuffle: "True"
2023-10-08 22:42:08,192 ----------------------------------------------------------------------------------------------------
2023-10-08 22:42:08,192 Plugins:
2023-10-08 22:42:08,192 - TensorboardLogger
2023-10-08 22:42:08,193 - LinearScheduler | warmup_fraction: '0.1'
2023-10-08 22:42:08,193 ----------------------------------------------------------------------------------------------------
2023-10-08 22:42:08,193 Final evaluation on model from best epoch (best-model.pt)
2023-10-08 22:42:08,193 - metric: "('micro avg', 'f1-score')"
2023-10-08 22:42:08,193 ----------------------------------------------------------------------------------------------------
2023-10-08 22:42:08,193 Computation:
2023-10-08 22:42:08,193 - compute on device: cuda:0
2023-10-08 22:42:08,193 - embedding storage: none
2023-10-08 22:42:08,193 ----------------------------------------------------------------------------------------------------
2023-10-08 22:42:08,193 Model training base path: "hmbench-ajmc/fr-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs8-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-4"
2023-10-08 22:42:08,193 ----------------------------------------------------------------------------------------------------
2023-10-08 22:42:08,193 ----------------------------------------------------------------------------------------------------
2023-10-08 22:42:08,193 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-08 22:42:17,054 epoch 1 - iter 12/121 - loss 3.24889415 - time (sec): 8.86 - samples/sec: 284.00 - lr: 0.000015 - momentum: 0.000000
2023-10-08 22:42:25,473 epoch 1 - iter 24/121 - loss 3.24375436 - time (sec): 17.28 - samples/sec: 278.54 - lr: 0.000030 - momentum: 0.000000
2023-10-08 22:42:34,654 epoch 1 - iter 36/121 - loss 3.23173654 - time (sec): 26.46 - samples/sec: 280.88 - lr: 0.000046 - momentum: 0.000000
2023-10-08 22:42:43,457 epoch 1 - iter 48/121 - loss 3.21282088 - time (sec): 35.26 - samples/sec: 279.22 - lr: 0.000062 - momentum: 0.000000
2023-10-08 22:42:52,690 epoch 1 - iter 60/121 - loss 3.17429011 - time (sec): 44.50 - samples/sec: 280.18 - lr: 0.000078 - momentum: 0.000000
2023-10-08 22:43:01,491 epoch 1 - iter 72/121 - loss 3.11654123 - time (sec): 53.30 - samples/sec: 279.55 - lr: 0.000094 - momentum: 0.000000
2023-10-08 22:43:10,364 epoch 1 - iter 84/121 - loss 3.04629267 - time (sec): 62.17 - samples/sec: 278.57 - lr: 0.000110 - momentum: 0.000000
2023-10-08 22:43:19,191 epoch 1 - iter 96/121 - loss 2.97098295 - time (sec): 71.00 - samples/sec: 276.60 - lr: 0.000126 - momentum: 0.000000
2023-10-08 22:43:28,545 epoch 1 - iter 108/121 - loss 2.88064158 - time (sec): 80.35 - samples/sec: 277.12 - lr: 0.000141 - momentum: 0.000000
2023-10-08 22:43:37,286 epoch 1 - iter 120/121 - loss 2.79298143 - time (sec): 89.09 - samples/sec: 276.61 - lr: 0.000157 - momentum: 0.000000
2023-10-08 22:43:37,764 ----------------------------------------------------------------------------------------------------
2023-10-08 22:43:37,765 EPOCH 1 done: loss 2.7892 - lr: 0.000157
2023-10-08 22:43:44,063 DEV : loss 1.7997684478759766 - f1-score (micro avg) 0.0
2023-10-08 22:43:44,068 ----------------------------------------------------------------------------------------------------
2023-10-08 22:43:53,168 epoch 2 - iter 12/121 - loss 1.76567829 - time (sec): 9.10 - samples/sec: 269.30 - lr: 0.000158 - momentum: 0.000000
2023-10-08 22:44:03,022 epoch 2 - iter 24/121 - loss 1.65613451 - time (sec): 18.95 - samples/sec: 271.63 - lr: 0.000157 - momentum: 0.000000
2023-10-08 22:44:11,666 epoch 2 - iter 36/121 - loss 1.56993594 - time (sec): 27.60 - samples/sec: 265.76 - lr: 0.000155 - momentum: 0.000000
2023-10-08 22:44:21,012 epoch 2 - iter 48/121 - loss 1.46749385 - time (sec): 36.94 - samples/sec: 268.58 - lr: 0.000153 - momentum: 0.000000
2023-10-08 22:44:30,207 epoch 2 - iter 60/121 - loss 1.38822825 - time (sec): 46.14 - samples/sec: 269.24 - lr: 0.000151 - momentum: 0.000000
2023-10-08 22:44:39,123 epoch 2 - iter 72/121 - loss 1.29536092 - time (sec): 55.05 - samples/sec: 268.01 - lr: 0.000150 - momentum: 0.000000
2023-10-08 22:44:48,775 epoch 2 - iter 84/121 - loss 1.21650287 - time (sec): 64.70 - samples/sec: 267.72 - lr: 0.000148 - momentum: 0.000000
2023-10-08 22:44:58,943 epoch 2 - iter 96/121 - loss 1.14579444 - time (sec): 74.87 - samples/sec: 267.43 - lr: 0.000146 - momentum: 0.000000
2023-10-08 22:45:07,881 epoch 2 - iter 108/121 - loss 1.10747235 - time (sec): 83.81 - samples/sec: 265.63 - lr: 0.000144 - momentum: 0.000000
2023-10-08 22:45:17,047 epoch 2 - iter 120/121 - loss 1.05793351 - time (sec): 92.98 - samples/sec: 264.65 - lr: 0.000143 - momentum: 0.000000
2023-10-08 22:45:17,580 ----------------------------------------------------------------------------------------------------
2023-10-08 22:45:17,580 EPOCH 2 done: loss 1.0580 - lr: 0.000143
2023-10-08 22:45:24,067 DEV : loss 0.6450902819633484 - f1-score (micro avg) 0.0
2023-10-08 22:45:24,073 ----------------------------------------------------------------------------------------------------
2023-10-08 22:45:32,969 epoch 3 - iter 12/121 - loss 0.63941489 - time (sec): 8.90 - samples/sec: 250.70 - lr: 0.000141 - momentum: 0.000000
2023-10-08 22:45:42,132 epoch 3 - iter 24/121 - loss 0.62273678 - time (sec): 18.06 - samples/sec: 252.35 - lr: 0.000139 - momentum: 0.000000
2023-10-08 22:45:52,256 epoch 3 - iter 36/121 - loss 0.60202538 - time (sec): 28.18 - samples/sec: 255.55 - lr: 0.000137 - momentum: 0.000000
2023-10-08 22:46:00,961 epoch 3 - iter 48/121 - loss 0.58615023 - time (sec): 36.89 - samples/sec: 256.78 - lr: 0.000135 - momentum: 0.000000
2023-10-08 22:46:10,364 epoch 3 - iter 60/121 - loss 0.56309441 - time (sec): 46.29 - samples/sec: 257.96 - lr: 0.000134 - momentum: 0.000000
2023-10-08 22:46:19,908 epoch 3 - iter 72/121 - loss 0.54683763 - time (sec): 55.83 - samples/sec: 260.54 - lr: 0.000132 - momentum: 0.000000
2023-10-08 22:46:29,867 epoch 3 - iter 84/121 - loss 0.53010442 - time (sec): 65.79 - samples/sec: 263.51 - lr: 0.000130 - momentum: 0.000000
2023-10-08 22:46:40,013 epoch 3 - iter 96/121 - loss 0.50209518 - time (sec): 75.94 - samples/sec: 264.69 - lr: 0.000128 - momentum: 0.000000
2023-10-08 22:46:48,542 epoch 3 - iter 108/121 - loss 0.48786290 - time (sec): 84.47 - samples/sec: 262.61 - lr: 0.000127 - momentum: 0.000000
2023-10-08 22:46:57,688 epoch 3 - iter 120/121 - loss 0.47452859 - time (sec): 93.61 - samples/sec: 263.07 - lr: 0.000125 - momentum: 0.000000
2023-10-08 22:46:58,213 ----------------------------------------------------------------------------------------------------
2023-10-08 22:46:58,214 EPOCH 3 done: loss 0.4731 - lr: 0.000125
2023-10-08 22:47:04,647 DEV : loss 0.37881988286972046 - f1-score (micro avg) 0.2965
2023-10-08 22:47:04,652 saving best model
2023-10-08 22:47:05,526 ----------------------------------------------------------------------------------------------------
2023-10-08 22:47:14,444 epoch 4 - iter 12/121 - loss 0.46738746 - time (sec): 8.92 - samples/sec: 263.79 - lr: 0.000123 - momentum: 0.000000
2023-10-08 22:47:24,059 epoch 4 - iter 24/121 - loss 0.40614776 - time (sec): 18.53 - samples/sec: 269.22 - lr: 0.000121 - momentum: 0.000000
2023-10-08 22:47:34,401 epoch 4 - iter 36/121 - loss 0.36038870 - time (sec): 28.87 - samples/sec: 266.58 - lr: 0.000120 - momentum: 0.000000
2023-10-08 22:47:43,578 epoch 4 - iter 48/121 - loss 0.34632050 - time (sec): 38.05 - samples/sec: 264.10 - lr: 0.000118 - momentum: 0.000000
2023-10-08 22:47:53,536 epoch 4 - iter 60/121 - loss 0.33528371 - time (sec): 48.01 - samples/sec: 263.91 - lr: 0.000116 - momentum: 0.000000
2023-10-08 22:48:03,709 epoch 4 - iter 72/121 - loss 0.31801268 - time (sec): 58.18 - samples/sec: 265.59 - lr: 0.000114 - momentum: 0.000000
2023-10-08 22:48:12,957 epoch 4 - iter 84/121 - loss 0.30951747 - time (sec): 67.43 - samples/sec: 265.74 - lr: 0.000113 - momentum: 0.000000
2023-10-08 22:48:21,926 epoch 4 - iter 96/121 - loss 0.30492446 - time (sec): 76.40 - samples/sec: 265.88 - lr: 0.000111 - momentum: 0.000000
2023-10-08 22:48:30,226 epoch 4 - iter 108/121 - loss 0.30375952 - time (sec): 84.70 - samples/sec: 266.17 - lr: 0.000109 - momentum: 0.000000
2023-10-08 22:48:38,206 epoch 4 - iter 120/121 - loss 0.29955442 - time (sec): 92.68 - samples/sec: 266.23 - lr: 0.000107 - momentum: 0.000000
2023-10-08 22:48:38,604 ----------------------------------------------------------------------------------------------------
2023-10-08 22:48:38,604 EPOCH 4 done: loss 0.3000 - lr: 0.000107
2023-10-08 22:48:44,517 DEV : loss 0.26859453320503235 - f1-score (micro avg) 0.5389
2023-10-08 22:48:44,523 saving best model
2023-10-08 22:48:45,439 ----------------------------------------------------------------------------------------------------
2023-10-08 22:48:53,562 epoch 5 - iter 12/121 - loss 0.22313744 - time (sec): 8.12 - samples/sec: 283.57 - lr: 0.000105 - momentum: 0.000000
2023-10-08 22:49:02,121 epoch 5 - iter 24/121 - loss 0.21957236 - time (sec): 16.68 - samples/sec: 280.87 - lr: 0.000104 - momentum: 0.000000
2023-10-08 22:49:11,118 epoch 5 - iter 36/121 - loss 0.22202082 - time (sec): 25.68 - samples/sec: 286.47 - lr: 0.000102 - momentum: 0.000000
2023-10-08 22:49:19,763 epoch 5 - iter 48/121 - loss 0.22785480 - time (sec): 34.32 - samples/sec: 283.63 - lr: 0.000100 - momentum: 0.000000
2023-10-08 22:49:28,751 epoch 5 - iter 60/121 - loss 0.22395787 - time (sec): 43.31 - samples/sec: 284.52 - lr: 0.000098 - momentum: 0.000000
2023-10-08 22:49:37,399 epoch 5 - iter 72/121 - loss 0.22025587 - time (sec): 51.96 - samples/sec: 284.09 - lr: 0.000097 - momentum: 0.000000
2023-10-08 22:49:46,172 epoch 5 - iter 84/121 - loss 0.22892588 - time (sec): 60.73 - samples/sec: 284.96 - lr: 0.000095 - momentum: 0.000000
2023-10-08 22:49:55,134 epoch 5 - iter 96/121 - loss 0.22823501 - time (sec): 69.69 - samples/sec: 284.01 - lr: 0.000093 - momentum: 0.000000
2023-10-08 22:50:03,052 epoch 5 - iter 108/121 - loss 0.22364187 - time (sec): 77.61 - samples/sec: 282.40 - lr: 0.000091 - momentum: 0.000000
2023-10-08 22:50:11,800 epoch 5 - iter 120/121 - loss 0.21942528 - time (sec): 86.36 - samples/sec: 283.42 - lr: 0.000090 - momentum: 0.000000
2023-10-08 22:50:12,655 ----------------------------------------------------------------------------------------------------
2023-10-08 22:50:12,655 EPOCH 5 done: loss 0.2185 - lr: 0.000090
2023-10-08 22:50:18,519 DEV : loss 0.21061980724334717 - f1-score (micro avg) 0.6275
2023-10-08 22:50:18,525 saving best model
2023-10-08 22:50:19,443 ----------------------------------------------------------------------------------------------------
2023-10-08 22:50:27,761 epoch 6 - iter 12/121 - loss 0.17277091 - time (sec): 8.32 - samples/sec: 286.54 - lr: 0.000088 - momentum: 0.000000
2023-10-08 22:50:36,616 epoch 6 - iter 24/121 - loss 0.16954790 - time (sec): 17.17 - samples/sec: 287.39 - lr: 0.000086 - momentum: 0.000000
2023-10-08 22:50:45,486 epoch 6 - iter 36/121 - loss 0.17882791 - time (sec): 26.04 - samples/sec: 288.85 - lr: 0.000084 - momentum: 0.000000
2023-10-08 22:50:53,736 epoch 6 - iter 48/121 - loss 0.16960253 - time (sec): 34.29 - samples/sec: 284.15 - lr: 0.000082 - momentum: 0.000000
2023-10-08 22:51:02,448 epoch 6 - iter 60/121 - loss 0.17113200 - time (sec): 43.00 - samples/sec: 284.14 - lr: 0.000081 - momentum: 0.000000
2023-10-08 22:51:11,093 epoch 6 - iter 72/121 - loss 0.17490968 - time (sec): 51.65 - samples/sec: 283.28 - lr: 0.000079 - momentum: 0.000000
2023-10-08 22:51:19,751 epoch 6 - iter 84/121 - loss 0.17136526 - time (sec): 60.31 - samples/sec: 283.91 - lr: 0.000077 - momentum: 0.000000
2023-10-08 22:51:28,848 epoch 6 - iter 96/121 - loss 0.16690048 - time (sec): 69.40 - samples/sec: 284.07 - lr: 0.000075 - momentum: 0.000000
2023-10-08 22:51:37,186 epoch 6 - iter 108/121 - loss 0.16544678 - time (sec): 77.74 - samples/sec: 283.37 - lr: 0.000074 - momentum: 0.000000
2023-10-08 22:51:45,939 epoch 6 - iter 120/121 - loss 0.16563812 - time (sec): 86.49 - samples/sec: 283.03 - lr: 0.000072 - momentum: 0.000000
2023-10-08 22:51:46,748 ----------------------------------------------------------------------------------------------------
2023-10-08 22:51:46,748 EPOCH 6 done: loss 0.1658 - lr: 0.000072
2023-10-08 22:51:52,688 DEV : loss 0.16975745558738708 - f1-score (micro avg) 0.7804
2023-10-08 22:51:52,694 saving best model
2023-10-08 22:51:53,580 ----------------------------------------------------------------------------------------------------
2023-10-08 22:52:02,885 epoch 7 - iter 12/121 - loss 0.11579701 - time (sec): 9.30 - samples/sec: 286.55 - lr: 0.000070 - momentum: 0.000000
2023-10-08 22:52:10,925 epoch 7 - iter 24/121 - loss 0.13622268 - time (sec): 17.34 - samples/sec: 277.56 - lr: 0.000068 - momentum: 0.000000
2023-10-08 22:52:19,329 epoch 7 - iter 36/121 - loss 0.12809804 - time (sec): 25.75 - samples/sec: 276.22 - lr: 0.000066 - momentum: 0.000000
2023-10-08 22:52:28,313 epoch 7 - iter 48/121 - loss 0.12708384 - time (sec): 34.73 - samples/sec: 274.47 - lr: 0.000065 - momentum: 0.000000
2023-10-08 22:52:36,945 epoch 7 - iter 60/121 - loss 0.12872714 - time (sec): 43.36 - samples/sec: 275.56 - lr: 0.000063 - momentum: 0.000000
2023-10-08 22:52:45,849 epoch 7 - iter 72/121 - loss 0.12749752 - time (sec): 52.27 - samples/sec: 274.80 - lr: 0.000061 - momentum: 0.000000
2023-10-08 22:52:54,917 epoch 7 - iter 84/121 - loss 0.12729279 - time (sec): 61.34 - samples/sec: 274.29 - lr: 0.000059 - momentum: 0.000000
2023-10-08 22:53:03,443 epoch 7 - iter 96/121 - loss 0.12653363 - time (sec): 69.86 - samples/sec: 273.77 - lr: 0.000058 - momentum: 0.000000
2023-10-08 22:53:13,270 epoch 7 - iter 108/121 - loss 0.12821719 - time (sec): 79.69 - samples/sec: 275.75 - lr: 0.000056 - momentum: 0.000000
2023-10-08 22:53:22,756 epoch 7 - iter 120/121 - loss 0.12880684 - time (sec): 89.17 - samples/sec: 276.49 - lr: 0.000054 - momentum: 0.000000
2023-10-08 22:53:23,237 ----------------------------------------------------------------------------------------------------
2023-10-08 22:53:23,238 EPOCH 7 done: loss 0.1292 - lr: 0.000054
2023-10-08 22:53:29,464 DEV : loss 0.15908175706863403 - f1-score (micro avg) 0.8025
2023-10-08 22:53:29,469 saving best model
2023-10-08 22:53:30,346 ----------------------------------------------------------------------------------------------------
2023-10-08 22:53:39,723 epoch 8 - iter 12/121 - loss 0.09380565 - time (sec): 9.38 - samples/sec: 277.43 - lr: 0.000052 - momentum: 0.000000
2023-10-08 22:53:48,462 epoch 8 - iter 24/121 - loss 0.09475660 - time (sec): 18.11 - samples/sec: 263.16 - lr: 0.000051 - momentum: 0.000000
2023-10-08 22:53:57,857 epoch 8 - iter 36/121 - loss 0.09983123 - time (sec): 27.51 - samples/sec: 269.28 - lr: 0.000049 - momentum: 0.000000
2023-10-08 22:54:06,734 epoch 8 - iter 48/121 - loss 0.10566214 - time (sec): 36.39 - samples/sec: 268.40 - lr: 0.000047 - momentum: 0.000000
2023-10-08 22:54:15,480 epoch 8 - iter 60/121 - loss 0.11078771 - time (sec): 45.13 - samples/sec: 265.31 - lr: 0.000045 - momentum: 0.000000
2023-10-08 22:54:24,874 epoch 8 - iter 72/121 - loss 0.11381718 - time (sec): 54.53 - samples/sec: 266.02 - lr: 0.000044 - momentum: 0.000000
2023-10-08 22:54:34,065 epoch 8 - iter 84/121 - loss 0.11202327 - time (sec): 63.72 - samples/sec: 265.25 - lr: 0.000042 - momentum: 0.000000
2023-10-08 22:54:43,568 epoch 8 - iter 96/121 - loss 0.11127488 - time (sec): 73.22 - samples/sec: 266.00 - lr: 0.000040 - momentum: 0.000000
2023-10-08 22:54:53,279 epoch 8 - iter 108/121 - loss 0.11004380 - time (sec): 82.93 - samples/sec: 265.40 - lr: 0.000038 - momentum: 0.000000
2023-10-08 22:55:02,926 epoch 8 - iter 120/121 - loss 0.10894085 - time (sec): 92.58 - samples/sec: 264.89 - lr: 0.000037 - momentum: 0.000000
2023-10-08 22:55:03,654 ----------------------------------------------------------------------------------------------------
2023-10-08 22:55:03,655 EPOCH 8 done: loss 0.1085 - lr: 0.000037
2023-10-08 22:55:10,189 DEV : loss 0.14787454903125763 - f1-score (micro avg) 0.8292
2023-10-08 22:55:10,194 saving best model
2023-10-08 22:55:11,094 ----------------------------------------------------------------------------------------------------
2023-10-08 22:55:22,036 epoch 9 - iter 12/121 - loss 0.07924901 - time (sec): 10.94 - samples/sec: 271.28 - lr: 0.000035 - momentum: 0.000000
2023-10-08 22:55:31,021 epoch 9 - iter 24/121 - loss 0.08290001 - time (sec): 19.93 - samples/sec: 266.95 - lr: 0.000033 - momentum: 0.000000
2023-10-08 22:55:39,752 epoch 9 - iter 36/121 - loss 0.08487672 - time (sec): 28.66 - samples/sec: 265.87 - lr: 0.000031 - momentum: 0.000000
2023-10-08 22:55:48,962 epoch 9 - iter 48/121 - loss 0.09640981 - time (sec): 37.87 - samples/sec: 264.35 - lr: 0.000029 - momentum: 0.000000
2023-10-08 22:55:58,261 epoch 9 - iter 60/121 - loss 0.09671133 - time (sec): 47.17 - samples/sec: 264.09 - lr: 0.000028 - momentum: 0.000000
2023-10-08 22:56:06,918 epoch 9 - iter 72/121 - loss 0.09540228 - time (sec): 55.82 - samples/sec: 262.65 - lr: 0.000026 - momentum: 0.000000
2023-10-08 22:56:16,764 epoch 9 - iter 84/121 - loss 0.09700245 - time (sec): 65.67 - samples/sec: 262.68 - lr: 0.000024 - momentum: 0.000000
2023-10-08 22:56:26,084 epoch 9 - iter 96/121 - loss 0.09708608 - time (sec): 74.99 - samples/sec: 263.15 - lr: 0.000022 - momentum: 0.000000
2023-10-08 22:56:35,194 epoch 9 - iter 108/121 - loss 0.09763745 - time (sec): 84.10 - samples/sec: 263.20 - lr: 0.000021 - momentum: 0.000000
2023-10-08 22:56:44,596 epoch 9 - iter 120/121 - loss 0.09566562 - time (sec): 93.50 - samples/sec: 263.13 - lr: 0.000019 - momentum: 0.000000
2023-10-08 22:56:45,173 ----------------------------------------------------------------------------------------------------
2023-10-08 22:56:45,174 EPOCH 9 done: loss 0.0952 - lr: 0.000019
2023-10-08 22:56:51,652 DEV : loss 0.14321769773960114 - f1-score (micro avg) 0.8257
2023-10-08 22:56:51,658 ----------------------------------------------------------------------------------------------------
2023-10-08 22:57:01,966 epoch 10 - iter 12/121 - loss 0.09113587 - time (sec): 10.31 - samples/sec: 273.12 - lr: 0.000017 - momentum: 0.000000
2023-10-08 22:57:10,536 epoch 10 - iter 24/121 - loss 0.09256184 - time (sec): 18.88 - samples/sec: 259.59 - lr: 0.000015 - momentum: 0.000000
2023-10-08 22:57:19,728 epoch 10 - iter 36/121 - loss 0.08650274 - time (sec): 28.07 - samples/sec: 261.00 - lr: 0.000013 - momentum: 0.000000
2023-10-08 22:57:28,755 epoch 10 - iter 48/121 - loss 0.08409242 - time (sec): 37.10 - samples/sec: 262.32 - lr: 0.000012 - momentum: 0.000000
2023-10-08 22:57:37,846 epoch 10 - iter 60/121 - loss 0.08236105 - time (sec): 46.19 - samples/sec: 263.13 - lr: 0.000010 - momentum: 0.000000
2023-10-08 22:57:46,841 epoch 10 - iter 72/121 - loss 0.08181150 - time (sec): 55.18 - samples/sec: 267.35 - lr: 0.000008 - momentum: 0.000000
2023-10-08 22:57:55,303 epoch 10 - iter 84/121 - loss 0.08382245 - time (sec): 63.64 - samples/sec: 270.10 - lr: 0.000006 - momentum: 0.000000
2023-10-08 22:58:03,840 epoch 10 - iter 96/121 - loss 0.08569249 - time (sec): 72.18 - samples/sec: 270.82 - lr: 0.000005 - momentum: 0.000000
2023-10-08 22:58:12,915 epoch 10 - iter 108/121 - loss 0.08968322 - time (sec): 81.26 - samples/sec: 273.57 - lr: 0.000003 - momentum: 0.000000
2023-10-08 22:58:21,198 epoch 10 - iter 120/121 - loss 0.08682648 - time (sec): 89.54 - samples/sec: 273.96 - lr: 0.000001 - momentum: 0.000000
2023-10-08 22:58:21,877 ----------------------------------------------------------------------------------------------------
2023-10-08 22:58:21,877 EPOCH 10 done: loss 0.0878 - lr: 0.000001
2023-10-08 22:58:27,712 DEV : loss 0.14467351138591766 - f1-score (micro avg) 0.8109
2023-10-08 22:58:28,561 ----------------------------------------------------------------------------------------------------
2023-10-08 22:58:28,563 Loading model from best epoch ...
2023-10-08 22:58:31,430 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
2023-10-08 22:58:37,176
Results:
- F-score (micro) 0.7828
- F-score (macro) 0.4692
- Accuracy 0.6744
By class:
precision recall f1-score support
pers 0.7852 0.8417 0.8125 139
scope 0.7945 0.8992 0.8436 129
work 0.6484 0.7375 0.6901 80
loc 0.0000 0.0000 0.0000 9
date 0.0000 0.0000 0.0000 3
micro avg 0.7565 0.8111 0.7828 360
macro avg 0.4456 0.4957 0.4692 360
weighted avg 0.7320 0.8111 0.7694 360
2023-10-08 22:58:37,176 ----------------------------------------------------------------------------------------------------
|