--- license: apache-2.0 tags: - generated_from_trainer metrics: - f1 model-index: - name: bert-zs-sentence-classifier results: [] --- # bert-zs-sentence-classifier This model is a fine-tuned version of [bert-base-cased](https://huggingface.co/bert-base-cased) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.3663 - F1: 0.8483 ## 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: 1e-05 - train_batch_size: 16 - eval_batch_size: 8 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 1 ### Training results | Training Loss | Epoch | Step | Validation Loss | F1 | |:-------------:|:-----:|:-----:|:---------------:|:------:| | 0.5973 | 0.01 | 500 | 0.5186 | 0.7538 | | 0.5021 | 0.03 | 1000 | 0.4646 | 0.7996 | | 0.4741 | 0.04 | 1500 | 0.4634 | 0.8064 | | 0.4656 | 0.06 | 2000 | 0.4485 | 0.8142 | | 0.4567 | 0.07 | 2500 | 0.4345 | 0.8160 | | 0.4448 | 0.09 | 3000 | 0.4239 | 0.8228 | | 0.4403 | 0.1 | 3500 | 0.4155 | 0.8294 | | 0.4163 | 0.12 | 4000 | 0.4021 | 0.8290 | | 0.4205 | 0.13 | 4500 | 0.4057 | 0.8283 | | 0.416 | 0.14 | 5000 | 0.4049 | 0.8319 | | 0.4115 | 0.16 | 5500 | 0.4095 | 0.8280 | | 0.4156 | 0.17 | 6000 | 0.3927 | 0.8349 | | 0.4042 | 0.19 | 6500 | 0.4003 | 0.8392 | | 0.4057 | 0.2 | 7000 | 0.3929 | 0.8385 | | 0.3977 | 0.22 | 7500 | 0.3915 | 0.8406 | | 0.4049 | 0.23 | 8000 | 0.3785 | 0.8433 | | 0.4027 | 0.24 | 8500 | 0.3807 | 0.8424 | | 0.4096 | 0.26 | 9000 | 0.3768 | 0.8435 | | 0.3958 | 0.27 | 9500 | 0.3846 | 0.8420 | | 0.4037 | 0.29 | 10000 | 0.3808 | 0.8381 | | 0.3813 | 0.3 | 10500 | 0.4004 | 0.8415 | | 0.3934 | 0.32 | 11000 | 0.3821 | 0.8422 | | 0.3895 | 0.33 | 11500 | 0.3844 | 0.8428 | | 0.3907 | 0.35 | 12000 | 0.3847 | 0.8435 | | 0.3862 | 0.36 | 12500 | 0.3803 | 0.8431 | | 0.3958 | 0.37 | 13000 | 0.3739 | 0.8392 | | 0.3845 | 0.39 | 13500 | 0.3817 | 0.8422 | | 0.3914 | 0.4 | 14000 | 0.3857 | 0.8424 | | 0.3814 | 0.42 | 14500 | 0.3793 | 0.8438 | | 0.3816 | 0.43 | 15000 | 0.3843 | 0.8395 | | 0.4022 | 0.45 | 15500 | 0.3737 | 0.8436 | | 0.3879 | 0.46 | 16000 | 0.3750 | 0.8424 | | 0.3794 | 0.48 | 16500 | 0.3743 | 0.8410 | | 0.393 | 0.49 | 17000 | 0.3733 | 0.8461 | | 0.384 | 0.5 | 17500 | 0.3765 | 0.8476 | | 0.3782 | 0.52 | 18000 | 0.3748 | 0.8451 | | 0.3931 | 0.53 | 18500 | 0.3807 | 0.8454 | | 0.3889 | 0.55 | 19000 | 0.3653 | 0.8463 | | 0.386 | 0.56 | 19500 | 0.3707 | 0.8445 | | 0.3802 | 0.58 | 20000 | 0.3700 | 0.8474 | | 0.3883 | 0.59 | 20500 | 0.3646 | 0.8463 | | 0.3825 | 0.61 | 21000 | 0.3665 | 0.8513 | | 0.382 | 0.62 | 21500 | 0.3620 | 0.8508 | | 0.3795 | 0.63 | 22000 | 0.3692 | 0.8493 | | 0.367 | 0.65 | 22500 | 0.3704 | 0.8479 | | 0.3825 | 0.66 | 23000 | 0.3723 | 0.8472 | | 0.3902 | 0.68 | 23500 | 0.3681 | 0.8465 | | 0.3813 | 0.69 | 24000 | 0.3668 | 0.8515 | | 0.3878 | 0.71 | 24500 | 0.3632 | 0.8506 | | 0.3743 | 0.72 | 25000 | 0.3728 | 0.8463 | | 0.3826 | 0.73 | 25500 | 0.3746 | 0.8465 | | 0.3892 | 0.75 | 26000 | 0.3602 | 0.8518 | | 0.3767 | 0.76 | 26500 | 0.3722 | 0.8513 | | 0.3724 | 0.78 | 27000 | 0.3716 | 0.8499 | | 0.3767 | 0.79 | 27500 | 0.3651 | 0.8483 | | 0.3846 | 0.81 | 28000 | 0.3753 | 0.8493 | | 0.3748 | 0.82 | 28500 | 0.3720 | 0.8458 | | 0.3768 | 0.84 | 29000 | 0.3663 | 0.8508 | | 0.3716 | 0.85 | 29500 | 0.3635 | 0.8531 | | 0.3673 | 0.86 | 30000 | 0.3659 | 0.8485 | | 0.3805 | 0.88 | 30500 | 0.3608 | 0.8518 | | 0.3718 | 0.89 | 31000 | 0.3695 | 0.8520 | | 0.374 | 0.91 | 31500 | 0.3631 | 0.8485 | | 0.3871 | 0.92 | 32000 | 0.3659 | 0.8485 | | 0.3724 | 0.94 | 32500 | 0.3584 | 0.8518 | | 0.3756 | 0.95 | 33000 | 0.3587 | 0.8492 | | 0.3709 | 0.97 | 33500 | 0.3700 | 0.8488 | | 0.376 | 0.98 | 34000 | 0.3657 | 0.8492 | | 0.372 | 0.99 | 34500 | 0.3663 | 0.8483 | ### Framework versions - Transformers 4.19.2 - Pytorch 1.11.0+cu113 - Datasets 2.2.2 - Tokenizers 0.12.1