fine-tuned-DatasetQAS-IDK-MRC-with-xlm-roberta-large-with-ITTL-without-freeze-LR-1e-05

This model is a fine-tuned version of xlm-roberta-large on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.8090
• Exact Match: 74.0838
• F1: 80.8517

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: 1e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 16
• 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	Validation Loss	Exact Match	F1
6.1945	0.49	36	2.3430	50.0	50.0
3.5404	0.98	72	1.8265	48.5602	52.5772
1.9365	1.48	108	1.2750	59.4241	67.2505
1.9365	1.97	144	1.0492	66.0995	73.9501
1.2265	2.46	180	0.9042	69.7644	77.5779
0.9482	2.95	216	0.8393	71.2042	78.9342
0.7866	3.45	252	0.8805	70.8115	78.2310
0.7866	3.94	288	0.9333	69.5026	76.5121
0.6871	4.44	324	0.8045	75.3927	82.4815
0.6086	4.92	360	0.7908	75.5236	81.8859
0.6086	5.42	396	0.8351	73.0366	80.3624
0.5449	5.91	432	0.8090	74.0838	80.8517

Framework versions

• Transformers 4.26.1
• Pytorch 1.13.1+cu117
• Datasets 2.2.0
• Tokenizers 0.13.2