Edit model card

deberta-v3-small-otat-recommened-hp

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

  • Loss: 1.6500
  • Accuracy: 0.6229
  • Macro F1: 0.6240

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: 4.5e-05
  • train_batch_size: 16
  • eval_batch_size: 16
  • seed: 42
  • gradient_accumulation_steps: 4
  • 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: 1500
  • num_epochs: 10

Training results

Training Loss Epoch Step Validation Loss Accuracy Macro F1
0.8942 1.14 500 0.8753 0.6316 0.6330
0.7816 2.29 1000 0.8880 0.633 0.6216
0.7231 3.43 1500 0.8827 0.632 0.6322
0.6145 4.57 2000 0.9674 0.6369 0.6329
0.4694 5.71 2500 1.0903 0.6249 0.6200
0.3611 6.86 3000 1.2490 0.6216 0.6249
0.278 8.0 3500 1.4194 0.6201 0.6230
0.1689 9.14 4000 1.6500 0.6229 0.6240

Framework versions

  • Transformers 4.39.3
  • Pytorch 2.2.2+cu121
  • Datasets 2.18.0
  • Tokenizers 0.15.2
Downloads last month
8
Safetensors
Model size
142M 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 DandinPower/deberta-v3-small-otat-recommened-hp

Finetuned
(106)
this model

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

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

Evaluation results