mobilebert-uncased-squad-v2-30-10-22

This model is a fine-tuned version of csarron/mobilebert-uncased-squad-v2 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 13.5041

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: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss
No log	1.0	228	3.1250
No log	2.0	456	3.5844
1.1867	3.0	684	3.9242
1.1867	4.0	912	4.4160
0.842	5.0	1140	5.3645
0.842	6.0	1368	5.7769
0.5983	7.0	1596	5.6078
0.5983	8.0	1824	6.2784
0.4518	9.0	2052	7.7741
0.4518	10.0	2280	7.4468
0.332	11.0	2508	8.0155
0.332	12.0	2736	9.0584
0.332	13.0	2964	9.9912
0.2125	14.0	3192	11.2215
0.2125	15.0	3420	11.8451
0.1525	16.0	3648	12.2774
0.1525	17.0	3876	12.5938
0.1232	18.0	4104	13.8241
0.1232	19.0	4332	13.4653
0.1152	20.0	4560	13.5041

Framework versions

• Transformers 4.35.0
• Pytorch 2.1.0+cu118
• Datasets 2.14.6
• Tokenizers 0.14.1