--- base_model: vinai/phobert-base-v2 tags: - generated_from_trainer metrics: - accuracy - f1 model-index: - name: PhoBert_70KURL results: [] --- # PhoBert_70KURL This model is a fine-tuned version of [vinai/phobert-base-v2](https://huggingface.co/vinai/phobert-base-v2) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 0.3781 - Accuracy: 0.9371 - F1: 0.9417 ## 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: 2e-05 - train_batch_size: 64 - eval_batch_size: 64 - seed: 42 - gradient_accumulation_steps: 2 - total_train_batch_size: 128 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 2150 - num_epochs: 20 ### Training results | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | |:-------------:|:-------:|:----:|:---------------:|:--------:|:------:| | No log | 0.4651 | 200 | 0.3374 | 0.9302 | 0.9233 | | No log | 0.9302 | 400 | 0.1954 | 0.9353 | 0.9383 | | No log | 1.3953 | 600 | 0.1273 | 0.9598 | 0.9609 | | No log | 1.8605 | 800 | 0.1601 | 0.9423 | 0.9461 | | 0.3263 | 2.3256 | 1000 | 0.1059 | 0.9666 | 0.9675 | | 0.3263 | 2.7907 | 1200 | 0.0832 | 0.9742 | 0.9746 | | 0.3263 | 3.2558 | 1400 | 0.1073 | 0.9621 | 0.9637 | | 0.3263 | 3.7209 | 1600 | 0.1053 | 0.9644 | 0.9658 | | 0.1315 | 4.1860 | 1800 | 0.0994 | 0.9653 | 0.9663 | | 0.1315 | 4.6512 | 2000 | 0.1109 | 0.9617 | 0.9633 | | 0.1315 | 5.1163 | 2200 | 0.1350 | 0.9491 | 0.9523 | | 0.1315 | 5.5814 | 2400 | 0.1204 | 0.9543 | 0.9567 | | 0.0959 | 6.0465 | 2600 | 0.2663 | 0.8906 | 0.9034 | | 0.0959 | 6.5116 | 2800 | 0.1309 | 0.9510 | 0.9538 | | 0.0959 | 6.9767 | 3000 | 0.2027 | 0.9279 | 0.9340 | | 0.0959 | 7.4419 | 3200 | 0.1277 | 0.9509 | 0.9535 | | 0.0959 | 7.9070 | 3400 | 0.1892 | 0.9473 | 0.9507 | | 0.0689 | 8.3721 | 3600 | 0.2090 | 0.9350 | 0.9396 | | 0.0689 | 8.8372 | 3800 | 0.1247 | 0.9579 | 0.9596 | | 0.0689 | 9.3023 | 4000 | 0.2984 | 0.9132 | 0.9216 | | 0.0689 | 9.7674 | 4200 | 0.2277 | 0.9352 | 0.9401 | | 0.0483 | 10.2326 | 4400 | 0.2595 | 0.9266 | 0.9328 | | 0.0483 | 10.6977 | 4600 | 0.2725 | 0.9249 | 0.9313 | | 0.0483 | 11.1628 | 4800 | 0.2483 | 0.9341 | 0.9391 | | 0.0483 | 11.6279 | 5000 | 0.2195 | 0.9442 | 0.9478 | | 0.0379 | 12.0930 | 5200 | 0.7268 | 0.8450 | 0.8672 | | 0.0379 | 12.5581 | 5400 | 0.2236 | 0.9503 | 0.9530 | | 0.0379 | 13.0233 | 5600 | 0.2610 | 0.9438 | 0.9474 | | 0.0379 | 13.4884 | 5800 | 0.3155 | 0.9366 | 0.9412 | | 0.0379 | 13.9535 | 6000 | 0.3225 | 0.9374 | 0.9418 | | 0.0284 | 14.4186 | 6200 | 0.4604 | 0.9158 | 0.9238 | | 0.0284 | 14.8837 | 6400 | 0.3269 | 0.9372 | 0.9417 | | 0.0284 | 15.3488 | 6600 | 0.3419 | 0.9401 | 0.9442 | | 0.0284 | 15.8140 | 6800 | 0.2869 | 0.9465 | 0.9497 | | 0.0213 | 16.2791 | 7000 | 0.3623 | 0.9364 | 0.9410 | | 0.0213 | 16.7442 | 7200 | 0.2880 | 0.9433 | 0.9466 | | 0.0213 | 17.2093 | 7400 | 0.3010 | 0.9454 | 0.9484 | | 0.0213 | 17.6744 | 7600 | 0.3719 | 0.9344 | 0.9391 | | 0.0165 | 18.1395 | 7800 | 0.3160 | 0.9448 | 0.9481 | | 0.0165 | 18.6047 | 8000 | 0.3767 | 0.9388 | 0.9431 | | 0.0165 | 19.0698 | 8200 | 0.3702 | 0.9383 | 0.9426 | | 0.0165 | 19.5349 | 8400 | 0.3998 | 0.9343 | 0.9393 | | 0.0121 | 20.0 | 8600 | 0.3781 | 0.9371 | 0.9417 | ### Framework versions - Transformers 4.41.2 - Pytorch 2.1.2 - Datasets 2.19.2 - Tokenizers 0.19.1