metadata
library_name: transformers
base_model: allenai/scibert_scivocab_cased
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: arxiv-scibert-11class
results: []
arxiv-scibert-11class
This model is a fine-tuned version of allenai/scibert_scivocab_cased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.2082
- Accuracy: 0.7990
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: 16
- eval_batch_size: 32
- seed: 42
- 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
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.7767 | 1.0 | 592 | 0.6513 | 0.6605 |
| 0.5454 | 2.0 | 1184 | 0.6570 | 0.7297 |
| 0.3584 | 3.0 | 1776 | 0.7839 | 0.7914 |
| 0.2228 | 4.0 | 2368 | 0.8626 | 0.8074 |
| 0.1584 | 5.0 | 2960 | 0.9993 | 0.8218 |
Framework versions
- Transformers 5.0.0
- Pytorch 2.10.0+cu128
- Datasets 4.0.0
- Tokenizers 0.22.2