metadata

license: mit
base_model: microsoft/deberta-v3-small
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - f1
  - precision
  - recall
model-index:
  - name: DeBERTaV3_model_V3
    results: []

DeBERTaV3_model_V3

This model is a fine-tuned version of microsoft/deberta-v3-small on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.1015
Accuracy: 0.9693
F1: 0.8766
Precision: 0.8803
Recall: 0.8729

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: 5
eval_batch_size: 5
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 12

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
No log	1.0	100	0.3499	0.875	0.0	0.0	0.0
No log	2.0	200	0.2517	0.9068	0.4211	0.9412	0.2712
No log	3.0	300	0.1835	0.9396	0.7077	0.8961	0.5847
No log	4.0	400	0.1338	0.9587	0.8219	0.8911	0.7627
0.2507	5.0	500	0.1043	0.9640	0.8522	0.875	0.8305
0.2507	6.0	600	0.1076	0.9629	0.8472	0.8739	0.8220
0.2507	7.0	700	0.1061	0.9619	0.8475	0.8475	0.8475
0.2507	8.0	800	0.1015	0.9693	0.8766	0.8803	0.8729
0.2507	9.0	900	0.1099	0.9650	0.8596	0.8632	0.8559
0.0434	10.0	1000	0.1101	0.9661	0.8632	0.8707	0.8559
0.0434	11.0	1100	0.1054	0.9693	0.8766	0.8803	0.8729
0.0434	12.0	1200	0.1066	0.9682	0.8729	0.8729	0.8729

Framework versions

Transformers 4.41.2
Pytorch 2.3.1+cu121
Datasets 2.19.2
Tokenizers 0.19.1