finsynth/deberta-v3-base-financial-metric-ner

deberta-v3-base-financial-metric-test-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: 0.0667
• Precision: 0.9639
• Recall: 0.9639
• F1: 0.9639
• Accuracy: 0.9832

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: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	69	0.3878	0.7363	0.8072	0.7701	0.9142
No log	2.0	138	0.0801	0.94	0.9438	0.9419	0.9767
No log	3.0	207	0.0892	0.9472	0.9357	0.9414	0.9729
No log	4.0	276	0.0678	0.9325	0.9438	0.9381	0.9748
No log	5.0	345	0.0827	0.948	0.9518	0.9499	0.9767
No log	6.0	414	0.0735	0.9486	0.9639	0.9562	0.9823
No log	7.0	483	0.0821	0.9555	0.9478	0.9516	0.9776
0.2226	8.0	552	0.0685	0.9562	0.9639	0.96	0.9813
0.2226	9.0	621	0.0667	0.9639	0.9639	0.9639	0.9832
0.2226	10.0	690	0.0674	0.96	0.9639	0.9619	0.9823

Framework versions

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