QA_BERT_50_epoch

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

• Loss: 4.9109

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: 160
• eval_batch_size: 80
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss
No log	1.0	1	6.2090
No log	2.0	2	6.0772
No log	3.0	3	5.9414
No log	4.0	4	5.7888
No log	5.0	5	5.6358
No log	6.0	6	5.4781
No log	7.0	7	5.3259
No log	8.0	8	5.1904
No log	9.0	9	5.0758
No log	10.0	10	4.9715
No log	11.0	11	4.8743
No log	12.0	12	4.7854
No log	13.0	13	4.6994
No log	14.0	14	4.6228
No log	15.0	15	4.5529
No log	16.0	16	4.4915
No log	17.0	17	4.4442
No log	18.0	18	4.4073
No log	19.0	19	4.3770
No log	20.0	20	4.3537
No log	21.0	21	4.3292
No log	22.0	22	4.3063
No log	23.0	23	4.2858
No log	24.0	24	4.2697
No log	25.0	25	4.2673
No log	26.0	26	4.2773
No log	27.0	27	4.2956
No log	28.0	28	4.3234
No log	29.0	29	4.3491
No log	30.0	30	4.3719
No log	31.0	31	4.3902
No log	32.0	32	4.4187
No log	33.0	33	4.4387
No log	34.0	34	4.4576
No log	35.0	35	4.4799
No log	36.0	36	4.5204
No log	37.0	37	4.5620
No log	38.0	38	4.6024
No log	39.0	39	4.6346
No log	40.0	40	4.6566
No log	41.0	41	4.6756
No log	42.0	42	4.6900
No log	43.0	43	4.6994
No log	44.0	44	4.7294
No log	45.0	45	4.7631
No log	46.0	46	4.7912
No log	47.0	47	4.8192
No log	48.0	48	4.8414
No log	49.0	49	4.8611
No log	50.0	50	4.9109

Framework versions

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