QA_BERT

This model was trained from scratch on the None dataset. It achieves the following results on the evaluation set:

• Loss: 3.7595

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: 320
• eval_batch_size: 40
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 25

Training results

Training Loss	Epoch	Step	Validation Loss
No log	1.0	1	5.5075
No log	2.0	2	4.8600
No log	3.0	3	4.5288
No log	4.0	4	4.2908
No log	5.0	5	4.1877
No log	6.0	6	4.1192
No log	7.0	7	4.0090
No log	8.0	8	3.8800
No log	9.0	9	3.7745
No log	10.0	10	3.7043
No log	11.0	11	3.6591
No log	12.0	12	3.6300
No log	13.0	13	3.6148
No log	14.0	14	3.6134
No log	15.0	15	3.6236
No log	16.0	16	3.6432
No log	17.0	17	3.6669
No log	18.0	18	3.6900
No log	19.0	19	3.7081
No log	20.0	20	3.7238
No log	21.0	21	3.7352
No log	22.0	22	3.7451
No log	23.0	23	3.7520
No log	24.0	24	3.7569
No log	25.0	25	3.7595

Framework versions

• Transformers 4.36.2
• Pytorch 2.1.2+cu121
• Datasets 2.14.7
• Tokenizers 0.15.0