cellate-tapt_base-LR_1e-05

This model is a fine-tuned version of microsoft/BiomedNLP-BiomedBERT-base-uncased-abstract-fulltext on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 1.2573
  • Accuracy: 0.7397

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: 16
  • seed: 3407
  • gradient_accumulation_steps: 2
  • total_train_batch_size: 32
  • optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-06 and optimizer_args=No additional optimizer arguments
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_ratio: 0.1
  • num_epochs: 20
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Accuracy
1.3595 1.0 6 1.2847 0.7278
1.3505 2.0 12 1.2635 0.7379
1.3676 3.0 18 1.3146 0.7406
1.2871 4.0 24 1.2949 0.7277
1.26 5.0 30 1.3364 0.7255
1.2586 6.0 36 1.3626 0.7243
1.307 7.0 42 1.2961 0.7280
1.248 8.0 48 1.2761 0.7266
1.2549 9.0 54 1.2707 0.7302
1.2123 10.0 60 1.3332 0.7139
1.2096 11.0 66 1.2993 0.7317
1.2494 12.0 72 1.2903 0.7274
1.2028 13.0 78 1.4261 0.7158
1.1927 14.0 84 1.2636 0.7219
1.2186 15.0 90 1.3468 0.7187
1.2001 16.0 96 1.3215 0.7183
1.3421 16.7273 100 1.2573 0.7397

Framework versions

  • Transformers 4.48.2
  • Pytorch 2.4.1+cu121
  • Datasets 3.0.2
  • Tokenizers 0.21.0
Downloads last month
2
Safetensors
Model size
0.1B params
Tensor type
F32
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for Mardiyyah/cellate-tapt_base-LR_1e-05_grouped_txt