fudnet

This model is a fine-tuned version of setu4993/LaBSE on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1370
• Accuracy: 0.9537

Model description

FUDNet is for Follow-Up Detector Net. This model takes two consequtive questions of a multidoc2dial question answering conversation and determines whether those two questions are from the same documents or not.

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: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 50
• num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.1946	1.0	732	0.1493	0.9453
0.0945	2.0	1464	0.1370	0.9537

Framework versions

• Transformers 4.17.0
• Pytorch 1.12.1+cu113
• Datasets 2.0.0
• Tokenizers 0.12.1