Edit model card

TExAS-SQuAD-de

This model is a fine-tuned version of xlm-roberta-base on the TExAS-SQuAD-de dataset. It achieves the following results on the evaluation set:

  • Exact match: 61.45%
  • F1-score: 66.12%

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 2e-05
  • train_batch_size: 8
  • eval_batch_size: 8
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 32
  • 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
1.8084 1.0 4233 1.5897
1.5696 2.0 8466 1.5478
1.4196 3.0 12699 1.5754

Framework versions

  • Transformers 4.12.2
  • Pytorch 1.8.1+cu101
  • Datasets 1.12.1
  • Tokenizers 0.10.3
Downloads last month
14
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for saattrupdan/xlmr-base-texas-squad-de

Finetuned
(2489)
this model