roberta-base-rte

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

  • Loss: 0.5446
  • Accuracy: 0.7978

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: 1e-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_ratio: 0.06
  • num_epochs: 10.0

Training results

Training Loss Epoch Step Validation Loss Accuracy
No log 1.0 156 0.7023 0.4729
No log 2.0 312 0.6356 0.6895
No log 3.0 468 0.5177 0.7617
0.6131 4.0 624 0.6238 0.7473
0.6131 5.0 780 0.5446 0.7978
0.6131 6.0 936 0.9697 0.7545
0.2528 7.0 1092 1.1004 0.7690
0.2528 8.0 1248 1.1937 0.7726
0.2528 9.0 1404 1.3313 0.7726
0.1073 10.0 1560 1.3534 0.7726

Framework versions

  • Transformers 4.20.0.dev0
  • Pytorch 1.11.0+cu113
  • Datasets 2.1.0
  • Tokenizers 0.12.1
Downloads last month
40
Safetensors
Model size
125M params
Tensor type
I64
·
F32
·
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.

Dataset used to train JeremiahZ/roberta-base-rte

Evaluation results