README.md

metadata

license: mit
base_model: microsoft/deberta-base
tags:
  - generated_from_trainer
metrics:
  - f1
  - accuracy
model-index:
  - name: output
    results: []

output

This model is a fine-tuned version of microsoft/deberta-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0014
• F1: 0.9009
• Accuracy: 0.89

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: 3e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 2
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Accuracy
0.3529	0.16	10	0.2584	0.6667	0.5
0.2348	0.32	20	0.0989	0.6667	0.5
0.0492	0.48	30	0.0314	0.9615	0.96
0.0336	0.64	40	0.0132	0.6849	0.54
0.0185	0.8	50	0.0345	0.6667	0.5
0.0114	0.96	60	0.0490	0.9524	0.95
0.0118	1.12	70	0.0235	0.7042	0.58
0.01	1.28	80	0.0352	0.7299	0.63
0.0061	1.44	90	0.0195	0.8	0.75
0.0067	1.6	100	0.0108	0.8547	0.83
0.0055	1.76	110	0.0186	0.7874	0.73
0.0052	1.92	120	0.0141	0.9174	0.91

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu121
• Datasets 2.16.1
• Tokenizers 0.15.0