jackmedda/google-bigbird-roberta-base_finetuned_augmented_augmented_gemini

This model is a fine-tuned version of google/bigbird-roberta-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.6858
• Accuracy: 0.8824
• F1: 0.9231
• Precision: 0.9231
• Recall: 0.9231

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: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
0.5114	1.0	46	0.8325	0.7	0.8235	0.7	1.0
0.6362	2.0	92	0.7559	0.7	0.8235	0.7	1.0
0.54	3.0	138	0.6932	0.7	0.8235	0.7	1.0
0.4266	4.0	184	0.5539	0.7	0.8	0.75	0.8571
0.2418	5.0	230	0.7236	0.7	0.8235	0.7	1.0
0.2088	6.0	276	0.6816	0.7	0.8235	0.7	1.0
0.1444	7.0	322	0.5416	0.9	0.9333	0.875	1.0
0.1138	8.0	368	0.5843	0.9	0.9333	0.875	1.0
0.0024	9.0	414	0.6265	0.9	0.9333	0.875	1.0
0.1536	10.0	460	0.9787	0.8	0.8571	0.8571	0.8571
0.0016	11.0	506	0.6553	0.9	0.9333	0.875	1.0
0.1379	12.0	552	0.5226	0.9	0.9333	0.875	1.0
0.0011	13.0	598	0.4243	0.9	0.9333	0.875	1.0
0.0009	14.0	644	0.5967	0.9	0.9333	0.875	1.0
0.0007	15.0	690	0.0412	1.0	1.0	1.0	1.0
0.0006	16.0	736	0.3025	0.9	0.9333	0.875	1.0
0.0005	17.0	782	0.5500	0.9	0.9333	0.875	1.0
0.0005	18.0	828	0.6036	0.9	0.9333	0.875	1.0
0.0004	19.0	874	0.6402	0.9	0.9333	0.875	1.0
0.0004	20.0	920	0.6644	0.9	0.9333	0.875	1.0
0.0003	21.0	966	0.6896	0.9	0.9333	0.875	1.0
0.0003	22.0	1012	0.7052	0.9	0.9333	0.875	1.0
0.0003	23.0	1058	0.7198	0.9	0.9333	0.875	1.0

Framework versions

• Transformers 4.48.2
• Pytorch 2.3.0+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0