Edit model card

deberta-v3-large-otat-recommened-hp

This model is a fine-tuned version of microsoft/deberta-v3-large on the DandinPower/review_onlytitleandtext dataset. It achieves the following results on the evaluation set:

  • Loss: 0.8169
  • Accuracy: 0.6686
  • Macro F1: 0.6662

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: 6e-06
  • train_batch_size: 8
  • eval_batch_size: 8
  • seed: 42
  • gradient_accumulation_steps: 8
  • total_train_batch_size: 64
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 50
  • num_epochs: 5

Training results

Training Loss Epoch Step Validation Loss Accuracy Macro F1
0.7726 1.14 500 0.8107 0.6613 0.6602
0.6983 2.29 1000 0.7739 0.669 0.6662
0.6504 3.43 1500 0.7891 0.6726 0.6725
0.6067 4.57 2000 0.8169 0.6686 0.6662

Framework versions

  • Transformers 4.39.3
  • Pytorch 2.2.2+cu121
  • Datasets 2.18.0
  • Tokenizers 0.15.2
Downloads last month
12
Safetensors
Model size
435M params
Tensor type
F32
·

Finetuned from

Dataset used to train DandinPower/deberta-v3-large-otat-recommened-hp

Collection including DandinPower/deberta-v3-large-otat-recommened-hp

Evaluation results