README.md

metadata

license: mit
base_model: bobbyw/deberta-v3-large_v3_relationships
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - f1
  - precision
  - recall
model-index:
  - name: deberta-v3-large_v3_relationships
    results: []

deberta-v3-large_v3_relationships

This model is a fine-tuned version of bobbyw/deberta-v3-large_v3_relationships on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0124
• Accuracy: 0.0038
• F1: 0.0038
• Precision: 0.0019
• Recall: 0.4242
• Learning Rate: 0.002

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-06
• train_batch_size: 3
• eval_batch_size: 3
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: constant
• num_epochs: 8

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall	Rate
No log	1.0	382	0.0109	0.0027	0.0046	0.0023	0.5152	0.002
0.0043	2.0	764	0.0106	0.0026	0.0046	0.0023	0.5152	0.002
0.0041	3.0	1146	0.0115	0.0028	0.0051	0.0026	0.5758	0.002
0.004	4.0	1528	0.0132	0.0031	0.0048	0.0024	0.5455	0.002
0.004	5.0	1910	0.0140	0.0032	0.0046	0.0023	0.5152	0.002
0.0043	6.0	2292	0.0116	0.0032	0.0046	0.0023	0.5152	0.002
0.0044	7.0	2674	0.0129	0.0027	0.0043	0.0022	0.4848	0.002
0.0038	8.0	3056	0.0124	0.0038	0.0038	0.0019	0.4242	0.002

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1