prajwalJumde's picture
Update README.md
30436e1
metadata
license: mit
base_model: xlm-roberta-base
tags:
  - generated_from_trainer
model-index:
  - name: QA_SYNTHETIC_DATA_TRAIN_REAL_DATA_TEST_xlm_roberta-base
    results: []
datasets:
  - squad

QA_SYNTHETIC_DATA_TRAIN_REAL_DATA_TEST_xlm_roberta-base

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

  • Loss: 13.6582

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: 2e-05
  • train_batch_size: 2
  • eval_batch_size: 2
  • seed: 42
  • 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
0.018 1.0 19000 13.0396
0.0481 2.0 38000 10.5261
0.032 3.0 57000 10.4920
0.0025 4.0 76000 12.6251
0.0005 5.0 95000 12.7548
0.0 6.0 114000 12.1423
0.0 7.0 133000 11.4364
0.0044 8.0 152000 11.7476
0.0 9.0 171000 13.3022
0.0 10.0 190000 13.6582

Framework versions

  • Transformers 4.32.0.dev0
  • Pytorch 2.0.1+cu117
  • Datasets 2.14.4
  • Tokenizers 0.13.3