metadata

library_name: transformers
language:
  - en
license: apache-2.0
base_model: google/bert_uncased_L-2_H-256_A-4
tags:
  - generated_from_trainer
datasets:
  - glue
metrics:
  - accuracy
  - f1
model-index:
  - name: bert_uncased_L-2_H-256_A-4_mrpc
    results:
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: GLUE MRPC
          type: glue
          args: mrpc
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.7475490196078431
          - name: F1
            type: f1
            value: 0.835725677830941

bert_uncased_L-2_H-256_A-4_mrpc

This model is a fine-tuned version of google/bert_uncased_L-2_H-256_A-4 on the GLUE MRPC dataset. It achieves the following results on the evaluation set:

Loss: 0.5344
Accuracy: 0.7475
F1: 0.8357
Combined Score: 0.7916

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: 5e-05
train_batch_size: 256
eval_batch_size: 256
seed: 10
optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Combined Score
0.619	1.0	15	0.5956	0.6887	0.8146	0.7517
0.5893	2.0	30	0.5835	0.7010	0.8179	0.7594
0.5612	3.0	45	0.5597	0.7059	0.8171	0.7615
0.5397	4.0	60	0.5398	0.7377	0.8320	0.7849
0.5063	5.0	75	0.5358	0.7426	0.8336	0.7881
0.476	6.0	90	0.5344	0.7475	0.8357	0.7916
0.4361	7.0	105	0.5515	0.7451	0.8349	0.7900
0.4014	8.0	120	0.5508	0.75	0.8365	0.7933
0.3684	9.0	135	0.5901	0.7304	0.8254	0.7779
0.3396	10.0	150	0.5755	0.7426	0.8276	0.7851
0.3061	11.0	165	0.5943	0.75	0.8317	0.7908

Framework versions

Transformers 4.46.3
Pytorch 2.2.1+cu118
Datasets 2.17.0
Tokenizers 0.20.3