README.md

metadata

library_name: transformers
license: mit
base_model: microsoft/mdeberta-v3-base
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - precision
  - recall
  - f1
model-index:
  - name: mdeberta-v3-base-name-classifier-v2
    results: []

mdeberta-v3-base-name-classifier-v2

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

• Loss: 0.0216
• Accuracy: 0.9942
• Precision: 0.9983
• Recall: 0.9913
• F1: 0.9948

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: 128
• eval_batch_size: 128
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 2000
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
0.047	0.0359	2000	0.0390	0.9903	0.9974	0.9851	0.9912
0.0322	0.0718	4000	0.0346	0.9921	0.9968	0.9890	0.9929
0.0325	0.1076	6000	0.0293	0.9924	0.9970	0.9893	0.9932
0.0342	0.1435	8000	0.0264	0.9927	0.9973	0.9895	0.9934
0.0301	0.1794	10000	0.0260	0.9929	0.9967	0.9905	0.9936
0.0291	0.2153	12000	0.0259	0.9931	0.9984	0.9893	0.9938
0.0246	0.2511	14000	0.0263	0.9931	0.9971	0.9905	0.9938
0.0321	0.2870	16000	0.0264	0.9934	0.9988	0.9893	0.9940
0.0256	0.3229	18000	0.0250	0.9935	0.9980	0.9903	0.9941
0.0234	0.3588	20000	0.0260	0.9934	0.9969	0.9912	0.9940
0.0246	0.3946	22000	0.0246	0.9935	0.9975	0.9909	0.9942
0.0238	0.4305	24000	0.0252	0.9932	0.9961	0.9917	0.9938
0.0263	0.4664	26000	0.0238	0.9936	0.9976	0.9910	0.9943
0.0234	0.5023	28000	0.0250	0.9936	0.9972	0.9913	0.9943
0.0241	0.5382	30000	0.0230	0.9939	0.9978	0.9912	0.9945
0.0238	0.5740	32000	0.0228	0.9939	0.9984	0.9907	0.9945
0.0243	0.6099	34000	0.0239	0.9939	0.9993	0.9897	0.9945
0.023	0.6458	36000	0.0228	0.9939	0.9980	0.9911	0.9945
0.0252	0.6817	38000	0.0230	0.9941	0.9987	0.9907	0.9947
0.0251	0.7175	40000	0.0223	0.9940	0.9977	0.9915	0.9946
0.0217	0.7534	42000	0.0226	0.9940	0.9976	0.9916	0.9946
0.0269	0.7893	44000	0.0220	0.9941	0.9981	0.9914	0.9947
0.0227	0.8252	46000	0.0224	0.9939	0.9972	0.9918	0.9945
0.026	0.8610	48000	0.0216	0.9942	0.9986	0.9911	0.9948
0.0213	0.8969	50000	0.0220	0.9942	0.9983	0.9913	0.9948
0.0233	0.9328	52000	0.0217	0.9942	0.9982	0.9913	0.9948
0.0239	0.9687	54000	0.0216	0.9942	0.9983	0.9913	0.9948

Framework versions

• Transformers 4.57.1
• Pytorch 2.9.0+cu128
• Datasets 4.4.1
• Tokenizers 0.22.1