Deberta Auto Grading Newfinal

--- library_name: transformers license: mit base_model: microsoft/deberta-v3-base tags: - generated_from_trainer metrics: - accuracy model-index: - name: deberta-auto-grading-newfinal results: [] --- # deberta-auto-grading-newfinal 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: 0.5761 - Accuracy: 0.8292 - F1 Macro: 0.8088 - F1 Incorrect: 0.8100 - F1 Partial: 0.7366 - F1 Correct: 0.8797 ## 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: 7 ### Training results | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro | F1 Incorrect | F1 Partial | F1 Correct | |:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|:------------:|:----------:|:----------:| | 1.7530 | 1.0 | 354 | 0.5601 | 0.7406 | 0.7173 | 0.6901 | 0.6490 | 0.8127 | | 1.0009 | 2.0 | 708 | 0.5010 | 0.7609 | 0.7491 | 0.7415 | 0.6868 | 0.8189 | | 0.7233 | 3.0 | 1062 | 0.5282 | 0.7852 | 0.7585 | 0.7433 | 0.6764 | 0.8558 | | 0.5744 | 4.0 | 1416 | 0.5181 | 0.7896 | 0.7700 | 0.76 | 0.6994 | 0.8506 | | 0.4661 | 5.0 | 1770 | 0.6180 | 0.7918 | 0.7790 | 0.7717 | 0.7185 | 0.8466 | | 0.3881 | 6.0 | 2124 | 0.6160 | 0.8052 | 0.7898 | 0.7815 | 0.7281 | 0.8598 | | 0.3194 | 7.0 | 2478 | 0.6706 | 0.8052 | 0.7886 | 0.7784 | 0.7272 | 0.8603 | ### Framework versions - Transformers 5.0.0 - Pytorch 2.10.0+cu128 - Datasets 4.8.3 - Tokenizers 0.22.2