bert_base_lda_50_v1_book
This model is a fine-tuned version of on the gokulsrinivasagan/processed_book_corpus-ld-50 dataset. It achieves the following results on the evaluation set:
- Loss: 3.9956
- Accuracy: 0.7552
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0001
- train_batch_size: 96
- eval_batch_size: 96
- seed: 10
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 10000
- num_epochs: 25
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 8.955 | 0.4215 | 10000 | 8.7407 | 0.1644 |
| 8.7137 | 0.8431 | 20000 | 8.5491 | 0.1653 |
| 8.6265 | 1.2646 | 30000 | 8.4783 | 0.1663 |
| 6.7021 | 1.6861 | 40000 | 6.0988 | 0.4554 |
| 5.3349 | 2.1077 | 50000 | 5.0105 | 0.6051 |
| 5.0794 | 2.5292 | 60000 | 4.7754 | 0.6390 |
| 4.9395 | 2.9507 | 70000 | 4.6509 | 0.6570 |
| 4.849 | 3.3723 | 80000 | 4.5686 | 0.6692 |
| 4.7838 | 3.7938 | 90000 | 4.5056 | 0.6787 |
| 4.7252 | 4.2153 | 100000 | 4.4582 | 0.6854 |
| 4.6807 | 4.6369 | 110000 | 4.4130 | 0.6913 |
| 4.6389 | 5.0584 | 120000 | 4.3734 | 0.6966 |
| 4.6076 | 5.4799 | 130000 | 4.3443 | 0.7008 |
| 4.5871 | 5.9014 | 140000 | 4.3176 | 0.7046 |
| 4.5675 | 6.3230 | 150000 | 4.3085 | 0.7070 |
| 4.5509 | 6.7445 | 160000 | 4.2861 | 0.7097 |
| 4.5201 | 7.1660 | 170000 | 4.2644 | 0.7131 |
| 4.5046 | 7.5876 | 180000 | 4.2521 | 0.7151 |
| 4.4859 | 8.0091 | 190000 | 4.2339 | 0.7175 |
| 4.4759 | 8.4306 | 200000 | 4.2231 | 0.7194 |
| 4.4563 | 8.8522 | 210000 | 4.2089 | 0.7215 |
| 4.4461 | 9.2737 | 220000 | 4.1986 | 0.7233 |
| 4.4263 | 9.6952 | 230000 | 4.1845 | 0.7251 |
| 4.4123 | 10.1168 | 240000 | 4.1760 | 0.7270 |
| 4.4131 | 10.5383 | 250000 | 4.1642 | 0.7284 |
| 4.3987 | 10.9598 | 260000 | 4.1552 | 0.7298 |
| 4.3866 | 11.3814 | 270000 | 4.1501 | 0.7306 |
| 4.3844 | 11.8029 | 280000 | 4.1384 | 0.7325 |
| 4.3661 | 12.2244 | 290000 | 4.1314 | 0.7336 |
| 4.3614 | 12.6460 | 300000 | 4.1207 | 0.7351 |
| 4.3483 | 13.0675 | 310000 | 4.1126 | 0.7365 |
| 4.3453 | 13.4890 | 320000 | 4.1087 | 0.7371 |
| 4.3391 | 13.9106 | 330000 | 4.1031 | 0.7379 |
| 4.3372 | 14.3321 | 340000 | 4.0942 | 0.7396 |
| 4.3255 | 14.7536 | 350000 | 4.0891 | 0.7403 |
| 4.3166 | 15.1751 | 360000 | 4.0857 | 0.7410 |
| 4.315 | 15.5967 | 370000 | 4.0752 | 0.7421 |
| 4.3041 | 16.0182 | 380000 | 4.0704 | 0.7431 |
| 4.2986 | 16.4397 | 390000 | 4.0649 | 0.7440 |
| 4.293 | 16.8613 | 400000 | 4.0620 | 0.7446 |
| 4.2881 | 17.2828 | 410000 | 4.0532 | 0.7457 |
| 4.282 | 17.7043 | 420000 | 4.0508 | 0.7465 |
| 4.2738 | 18.1259 | 430000 | 4.0469 | 0.7471 |
| 4.2676 | 18.5474 | 440000 | 4.0429 | 0.7476 |
| 4.2666 | 18.9689 | 450000 | 4.0364 | 0.7485 |
| 4.2598 | 19.3905 | 460000 | 4.0315 | 0.7493 |
| 4.258 | 19.8120 | 470000 | 4.0286 | 0.7499 |
| 4.2503 | 20.2335 | 480000 | 4.0250 | 0.7505 |
| 4.2446 | 20.6551 | 490000 | 4.0202 | 0.7513 |
| 4.2401 | 21.0766 | 500000 | 4.0157 | 0.7517 |
| 4.2359 | 21.4981 | 510000 | 4.0125 | 0.7524 |
| 4.2301 | 21.9197 | 520000 | 4.0108 | 0.7527 |
| 4.2318 | 22.3412 | 530000 | 4.0075 | 0.7532 |
| 4.2247 | 22.7627 | 540000 | 4.0060 | 0.7536 |
| 4.2241 | 23.1843 | 550000 | 4.0021 | 0.7540 |
| 4.2201 | 23.6058 | 560000 | 3.9990 | 0.7546 |
| 4.2151 | 24.0273 | 570000 | 3.9981 | 0.7549 |
| 4.2144 | 24.4488 | 580000 | 3.9943 | 0.7553 |
| 4.2123 | 24.8704 | 590000 | 3.9949 | 0.7552 |
Framework versions
- Transformers 4.46.3
- Pytorch 2.2.1+cu118
- Datasets 2.17.0
- Tokenizers 0.20.3