Deberta Auto Grading

--- library_name: transformers license: mit base_model: microsoft/deberta-v3-base tags: - generated_from_trainer metrics: - accuracy model-index: - name: deberta-auto-grading results: [] --- # deberta-auto-grading This model is a fine-tuned version of [microsoft/deberta-v3-base](https://huggingface.co/microsoft/deberta-v3-base) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 0.4526 - Accuracy: 0.8205 - F1 Macro: 0.8005 - F1 Incorrect: 0.8267 - F1 Partial: 0.6995 - F1 Correct: 0.8753 ## 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: 32 - eval_batch_size: 64 - seed: 42 - gradient_accumulation_steps: 2 - total_train_batch_size: 64 - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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: 0.1 - num_epochs: 5 ### Training results | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro | F1 Incorrect | F1 Partial | F1 Correct | |:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|:------------:|:----------:|:----------:| | 1.7184 | 1.0 | 354 | 0.5803 | 0.7485 | 0.7210 | 0.6997 | 0.6398 | 0.8235 | | 1.0207 | 2.0 | 708 | 0.4883 | 0.7802 | 0.7586 | 0.75 | 0.6778 | 0.8481 | | 0.7490 | 3.0 | 1062 | 0.4664 | 0.8125 | 0.7928 | 0.7887 | 0.7178 | 0.8719 | | 0.5996 | 4.0 | 1416 | 0.4734 | 0.8060 | 0.7887 | 0.7870 | 0.7150 | 0.8641 | | 0.5094 | 5.0 | 1770 | 0.5118 | 0.8085 | 0.7904 | 0.7915 | 0.7127 | 0.8670 | ### Framework versions - Transformers 5.0.0 - Pytorch 2.10.0+cu128 - Datasets 4.8.3 - Tokenizers 0.22.2