salohnana2018's picture
Upload tokenizer
236cc5d verified
metadata
license: apache-2.0
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - f1
  - precision
  - recall
base_model: salohnana2018/HARD_without_dp_4248_camel_prepocessed_ASC
model-index:
  - name: ABSA-SentencePair-DAPT-HARD-4248-bert-base-Camel-MSA-ru2
    results: []

ABSA-SentencePair-DAPT-HARD-4248-bert-base-Camel-MSA-ru2

This model is a fine-tuned version of salohnana2018/HARD_without_dp_4248_camel_prepocessed_ASC on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.4529
  • Accuracy: 0.8956
  • F1: 0.8956
  • Precision: 0.8956
  • Recall: 0.8956

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: 32
  • eval_batch_size: 32
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 5

Training results

Training Loss Epoch Step Validation Loss Accuracy F1 Precision Recall
0.4175 1.0 265 0.3482 0.8696 0.8696 0.8696 0.8696
0.2889 2.0 530 0.3280 0.8819 0.8819 0.8819 0.8819
0.1995 3.0 795 0.3343 0.8908 0.8908 0.8908 0.8908
0.1319 4.0 1060 0.3856 0.8932 0.8932 0.8932 0.8932
0.0855 5.0 1325 0.4529 0.8956 0.8956 0.8956 0.8956

Framework versions

  • Transformers 4.38.2
  • Pytorch 2.2.1+cu121
  • Datasets 2.18.0
  • Tokenizers 0.15.2