deberta-v3-base-financial-inc-dec-ner

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

• Loss: 0.0416
• Precision: 0.9632
• Recall: 0.9704
• F1: 0.9668
• Accuracy: 0.9933

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 100
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	92	0.1193	0.625	0.7407	0.6780	0.9588
No log	2.0	184	0.0522	0.8643	0.8963	0.88	0.9798
No log	3.0	276	0.0554	0.8897	0.8963	0.8930	0.9835
No log	4.0	368	0.0362	0.9416	0.9556	0.9485	0.9910
No log	5.0	460	0.0315	0.9286	0.9630	0.9455	0.9918
0.1731	6.0	552	0.0416	0.9632	0.9704	0.9668	0.9933
0.1731	7.0	644	0.0496	0.9420	0.9630	0.9524	0.9910
0.1731	8.0	736	0.0527	0.9420	0.9630	0.9524	0.9910
0.1731	9.0	828	0.0604	0.9348	0.9556	0.9451	0.9895
0.1731	10.0	920	0.0564	0.9420	0.9630	0.9524	0.9910
0.0028	11.0	1012	0.0571	0.9493	0.9704	0.9597	0.9918
0.0028	12.0	1104	0.0570	0.9493	0.9704	0.9597	0.9918
0.0028	13.0	1196	0.0559	0.9493	0.9704	0.9597	0.9918
0.0028	14.0	1288	0.0574	0.9493	0.9704	0.9597	0.9918
0.0028	15.0	1380	0.0576	0.9493	0.9704	0.9597	0.9918

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu124
• Datasets 2.21.0
• Tokenizers 0.19.1