BERT2BERT_pretrained_LC-QuAD_2.0

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.0810
• Bleu: 74.7723
• Em: 0.0349
• Gen Len: 46.1685

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: 16
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 2500
• num_epochs: 8.0

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Em	Gen Len
6.9199	1.4	2000	6.3751	3.4801	0.0	220.966
3.0222	2.8	4000	2.8796	27.8543	0.0	36.8
1.5982	4.19	6000	1.7495	56.0747	0.0017	43.9021
1.0717	5.59	8000	1.2625	69.9606	0.0199	46.0722
0.7765	6.99	10000	1.0810	74.7723	0.0349	46.1685

Framework versions

• Transformers 4.30.0.dev0
• Pytorch 2.0.1+cu118
• Datasets 2.12.0
• Tokenizers 0.13.3