Edit model card

distilbert-base-uncased-finetuned-squad

This model is a fine-tuned version of distilbert-base-uncased on the squad dataset. It achieves the following results on the evaluation set:

  • Loss: 4.3629

Model description

Base model weights were frozen leaving only to finetune the last layer (qa outputs).

Training and evaluation data

Achieved EM: 4.7776726584673606, F1: 11.440882287905591

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 2e-05
  • train_batch_size: 16
  • eval_batch_size: 16
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 3

Training results

Training Loss Epoch Step Validation Loss
4.679 1.0 5533 4.6713
4.4171 2.0 11066 4.4218
4.3464 3.0 16599 4.3629

Framework versions

  • Transformers 4.15.0
  • Pytorch 1.10.0+cu111
  • Datasets 1.17.0
  • Tokenizers 0.10.3
Downloads last month
6

Dataset used to train ericRosello/distilbert-base-uncased-finetuned-squad-frozen-v1