cosmetic3_roberta

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

  • Loss: 0.3549
  • Accuracy: 0.8520
  • F1: 0.8563
  • Precision: 0.8640
  • Recall: 0.8526

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: 8
  • 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_steps: 500
  • num_epochs: 5

Training results

Training Loss Epoch Step Validation Loss Accuracy F1 Precision Recall
0.4186 1.0 277 0.2708 0.8877 0.8911 0.8933 0.8897
0.3237 2.0 554 0.4175 0.8623 0.8624 0.8755 0.8727
0.2054 3.0 831 0.4479 0.8986 0.9008 0.9087 0.9055
0.3151 4.0 1108 0.3885 0.9130 0.9154 0.9163 0.9173
0.1832 5.0 1385 0.4418 0.9094 0.9120 0.9122 0.9132

Framework versions

  • Transformers 4.38.2
  • Pytorch 2.2.1+cu121
  • Datasets 2.18.0
  • Tokenizers 0.15.2
Downloads last month
28
Safetensors
Model size
111M params
Tensor type
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.

Model tree for yaesol/cosmetic3_roberta

Base model

klue/roberta-base
Finetuned
(130)
this model