--- base_model: haryoaw/scenario-TCR-NER_data-univner_full library_name: transformers license: mit metrics: - precision - recall - f1 - accuracy tags: - generated_from_trainer model-index: - name: scenario-kd-scr-ner-half-xlmr_data-univner_full66 results: [] --- # scenario-kd-scr-ner-half-xlmr_data-univner_full66 This model is a fine-tuned version of [haryoaw/scenario-TCR-NER_data-univner_full](https://huggingface.co/haryoaw/scenario-TCR-NER_data-univner_full) on the None dataset. It achieves the following results on the evaluation set: - Loss: 1.2164 - Precision: 0.4268 - Recall: 0.3872 - F1: 0.4061 - Accuracy: 0.9451 ## 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: 3e-05 - train_batch_size: 8 - eval_batch_size: 32 - seed: 66 - gradient_accumulation_steps: 4 - total_train_batch_size: 32 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 10 ### Training results | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy | |:-------------:|:------:|:-----:|:---------------:|:---------:|:------:|:------:|:--------:| | 2.8528 | 0.2911 | 500 | 2.3182 | 0.5357 | 0.0022 | 0.0043 | 0.9241 | | 2.1207 | 0.5822 | 1000 | 2.0259 | 0.2078 | 0.0100 | 0.0190 | 0.9243 | | 1.8727 | 0.8732 | 1500 | 1.8417 | 0.2116 | 0.0853 | 0.1216 | 0.9265 | | 1.7263 | 1.1643 | 2000 | 1.7362 | 0.1592 | 0.0693 | 0.0965 | 0.9268 | | 1.6268 | 1.4554 | 2500 | 1.7251 | 0.2612 | 0.1594 | 0.1980 | 0.9303 | | 1.5859 | 1.7465 | 3000 | 1.6161 | 0.3013 | 0.1824 | 0.2272 | 0.9331 | | 1.4992 | 2.0375 | 3500 | 1.5799 | 0.3257 | 0.2106 | 0.2558 | 0.9348 | | 1.4072 | 2.3286 | 4000 | 1.5337 | 0.3402 | 0.2483 | 0.2871 | 0.9358 | | 1.374 | 2.6197 | 4500 | 1.5234 | 0.3113 | 0.2886 | 0.2995 | 0.9363 | | 1.3602 | 2.9108 | 5000 | 1.4697 | 0.3426 | 0.2717 | 0.3030 | 0.9376 | | 1.2661 | 3.2019 | 5500 | 1.4345 | 0.3421 | 0.2953 | 0.3170 | 0.9386 | | 1.2524 | 3.4929 | 6000 | 1.4146 | 0.3843 | 0.3011 | 0.3376 | 0.9395 | | 1.2057 | 3.7840 | 6500 | 1.4308 | 0.3767 | 0.2730 | 0.3166 | 0.9400 | | 1.2018 | 4.0751 | 7000 | 1.3902 | 0.3836 | 0.3069 | 0.3410 | 0.9406 | | 1.1243 | 4.3662 | 7500 | 1.3595 | 0.3811 | 0.3255 | 0.3511 | 0.9414 | | 1.1248 | 4.6573 | 8000 | 1.3407 | 0.3940 | 0.3122 | 0.3484 | 0.9414 | | 1.1234 | 4.9483 | 8500 | 1.3333 | 0.3802 | 0.3196 | 0.3473 | 0.9415 | | 1.0707 | 5.2394 | 9000 | 1.3303 | 0.3937 | 0.3301 | 0.3591 | 0.9422 | | 1.0384 | 5.5305 | 9500 | 1.2940 | 0.3962 | 0.3370 | 0.3642 | 0.9425 | | 1.0239 | 5.8216 | 10000 | 1.2959 | 0.3967 | 0.3486 | 0.3711 | 0.9420 | | 1.007 | 6.1126 | 10500 | 1.2798 | 0.4070 | 0.3653 | 0.3850 | 0.9430 | | 0.9654 | 6.4037 | 11000 | 1.2714 | 0.3904 | 0.3634 | 0.3764 | 0.9424 | | 0.9657 | 6.6948 | 11500 | 1.2591 | 0.3861 | 0.3774 | 0.3817 | 0.9428 | | 0.9678 | 6.9859 | 12000 | 1.2546 | 0.4209 | 0.3509 | 0.3827 | 0.9435 | | 0.9217 | 7.2770 | 12500 | 1.2610 | 0.4124 | 0.3686 | 0.3893 | 0.9433 | | 0.9056 | 7.5680 | 13000 | 1.2403 | 0.4238 | 0.3744 | 0.3976 | 0.9442 | | 0.9146 | 7.8591 | 13500 | 1.2396 | 0.4242 | 0.3779 | 0.3997 | 0.9445 | | 0.8974 | 8.1502 | 14000 | 1.2246 | 0.4213 | 0.3910 | 0.4056 | 0.9448 | | 0.8572 | 8.4413 | 14500 | 1.2233 | 0.4232 | 0.3831 | 0.4022 | 0.9447 | | 0.8703 | 8.7324 | 15000 | 1.2265 | 0.4228 | 0.3740 | 0.3969 | 0.9450 | | 0.8774 | 9.0234 | 15500 | 1.2190 | 0.4415 | 0.3806 | 0.4088 | 0.9454 | | 0.8581 | 9.3145 | 16000 | 1.2245 | 0.4251 | 0.3838 | 0.4034 | 0.9449 | | 0.8411 | 9.6056 | 16500 | 1.2153 | 0.4298 | 0.3982 | 0.4134 | 0.9453 | | 0.8466 | 9.8967 | 17000 | 1.2164 | 0.4268 | 0.3872 | 0.4061 | 0.9451 | ### Framework versions - Transformers 4.44.2 - Pytorch 2.1.1+cu121 - Datasets 2.14.5 - Tokenizers 0.19.1