kmcs-casulit/department-classification

my_awesome_model

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: 0.0808
• Accuracy: 0.8289
• F1: 0.8595
• Precision: 0.8864
• Recall: 0.8342

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: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
No log	1.0	47	0.1056	0.7059	0.7788	0.8684	0.7059
No log	2.0	94	0.0961	0.7219	0.7895	0.8710	0.7219
No log	3.0	141	0.1042	0.7594	0.8045	0.8554	0.7594
No log	4.0	188	0.0899	0.8021	0.8427	0.8876	0.8021
No log	5.0	235	0.0911	0.8182	0.8540	0.8807	0.8289
No log	6.0	282	0.0808	0.8289	0.8595	0.8864	0.8342
No log	7.0	329	0.0885	0.8503	0.8689	0.8883	0.8503
No log	8.0	376	0.0873	0.8396	0.8634	0.8827	0.8449
No log	9.0	423	0.0926	0.8342	0.8579	0.8771	0.8396
No log	10.0	470	0.0904	0.8342	0.8603	0.8820	0.8396

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.0+cu121
• Datasets 2.19.2
• Tokenizers 0.19.1