deberta_pf_multi

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

• Loss: 0.1231
• F1 Weighted: 0.8298
• F1 Samples: 0.8413
• F1 Macro: 0.6383
• F1 Micro: 0.8462
• Accuracy: 0.8166

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
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	F1 Weighted	F1 Samples	F1 Macro	F1 Micro	Accuracy
0.2836	0.3381	500	0.2080	0.5860	0.6270	0.2691	0.6650	0.6198
0.1898	0.6761	1000	0.1638	0.7330	0.7293	0.3928	0.7699	0.7118
0.1617	1.0142	1500	0.1448	0.7745	0.7888	0.5089	0.8054	0.7618
0.1418	1.3523	2000	0.1401	0.7987	0.8014	0.6172	0.8164	0.7781
0.1289	1.6903	2500	0.1336	0.8058	0.8156	0.6149	0.8256	0.7909
0.1315	2.0284	3000	0.1231	0.8213	0.8339	0.6221	0.8410	0.8092
0.1116	2.3665	3500	0.1250	0.8209	0.8272	0.6217	0.8375	0.8058
0.1106	2.7045	4000	0.1231	0.8298	0.8413	0.6383	0.8462	0.8166

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.1+cu121
• Datasets 2.19.2
• Tokenizers 0.19.1