STS-Lora-Fine-Tuning-Capstone-roberta-base-deepset-filtered-115-with-higher-r-mid

This model is a fine-tuned version of deepset/roberta-base-squad2 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.9056
• Accuracy: 0.6040

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: 3e-05
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	76	1.2568	0.4161
No log	2.0	152	1.1655	0.4603
No log	3.0	228	1.0574	0.5120
No log	4.0	304	0.9846	0.5574
No log	5.0	380	0.9665	0.5675
No log	6.0	456	0.9544	0.5738
1.0456	7.0	532	0.9503	0.5763
1.0456	8.0	608	0.9269	0.5876
1.0456	9.0	684	0.9233	0.5889
1.0456	10.0	760	0.9264	0.5927
1.0456	11.0	836	0.9092	0.5927
1.0456	12.0	912	0.9187	0.5914
1.0456	13.0	988	0.9122	0.6003
0.8486	14.0	1064	0.9091	0.5977
0.8486	15.0	1140	0.9079	0.5965
0.8486	16.0	1216	0.9144	0.5952
0.8486	17.0	1292	0.9049	0.5977
0.8486	18.0	1368	0.9257	0.5939
0.8486	19.0	1444	0.9006	0.5952
0.8112	20.0	1520	0.9008	0.6015
0.8112	21.0	1596	0.9044	0.6040
0.8112	22.0	1672	0.9008	0.6053
0.8112	23.0	1748	0.9052	0.6028
0.8112	24.0	1824	0.9065	0.6028
0.8112	25.0	1900	0.9015	0.6053
0.8112	26.0	1976	0.9141	0.5965
0.7992	27.0	2052	0.9072	0.6053
0.7992	28.0	2128	0.9042	0.6053
0.7992	29.0	2204	0.9054	0.6040
0.7992	30.0	2280	0.9056	0.6040

Framework versions

• PEFT 0.10.0
• Transformers 4.38.2
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2