nystromformer-4096-medqa-usmle-MiniLM-IR-cs

This model is a fine-tuned version of GBaker/nystromformer-4096-medqa-usmle-MiniLM-IR-cs on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.8436
• Accuracy: 0.2812

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: 5e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 32
• total_train_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Accuracy	Validation Loss
No log	0.99	79	0.2372	1.3863
No log	1.99	158	0.2655	1.3861
No log	2.99	237	0.2545	1.3859
No log	3.99	316	0.2765	1.3837
No log	4.99	395	0.2820	1.3876
No log	5.99	474	1.3819	0.2639
1.3342	6.99	553	1.4875	0.2694
1.3342	7.99	632	1.6126	0.2718
1.3342	8.99	711	1.7637	0.2804
1.3342	9.99	790	1.8436	0.2812

Framework versions

• Transformers 4.26.1
• Pytorch 1.13.1+cu116
• Datasets 2.9.0
• Tokenizers 0.13.2