STS-Lora-Fine-Tuning-Capstone-roberta-base-deepset-filtered-120-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: 0.7040
• Accuracy: 0.6854

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	57	1.0545	0.4551
No log	2.0	114	1.0270	0.5
No log	3.0	171	0.9823	0.5262
No log	4.0	228	0.9416	0.5356
No log	5.0	285	0.8917	0.5805
No log	6.0	342	0.7931	0.6180
No log	7.0	399	0.7627	0.6423
No log	8.0	456	0.7679	0.6573
0.7963	9.0	513	0.7380	0.6610
0.7963	10.0	570	0.7256	0.6760
0.7963	11.0	627	0.7223	0.6742
0.7963	12.0	684	0.7255	0.6779
0.7963	13.0	741	0.7132	0.6779
0.7963	14.0	798	0.7097	0.6835
0.7963	15.0	855	0.7116	0.6760
0.7963	16.0	912	0.7200	0.6760
0.7963	17.0	969	0.7176	0.6760
0.615	18.0	1026	0.7133	0.6798
0.615	19.0	1083	0.7121	0.6798
0.615	20.0	1140	0.7117	0.6873
0.615	21.0	1197	0.7028	0.6816
0.615	22.0	1254	0.7033	0.6854
0.615	23.0	1311	0.7054	0.6798
0.615	24.0	1368	0.7059	0.6854
0.615	25.0	1425	0.6996	0.6835
0.615	26.0	1482	0.7045	0.6835
0.5826	27.0	1539	0.7032	0.6854
0.5826	28.0	1596	0.7030	0.6854
0.5826	29.0	1653	0.7046	0.6854
0.5826	30.0	1710	0.7040	0.6854

Framework versions

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