metadata

license: mit
tags:
  - generated_from_trainer
datasets:
  - amazon_reviews_multi
metrics:
  - accuracy
model-index:
  - name: deberta_v3_amazon_reviews
    results:
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: amazon_reviews_multi
          type: amazon_reviews_multi
          args: en
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.61

deberta_v3_amazon_reviews

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

Loss: 0.9723
Accuracy: 0.61

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: 2e-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: 200
num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.9339	0.2	5000	0.9879	0.5876
0.9386	0.4	10000	0.9408	0.5992
0.9127	0.6	15000	0.9118	0.6004
0.8997	0.8	20000	0.9192	0.607
0.8853	1.0	25000	0.9167	0.6018
0.8159	1.2	30000	0.9364	0.6064
0.8367	1.4	35000	0.9215	0.6174
0.8322	1.6	40000	0.9076	0.6108
0.8142	1.8	45000	0.9305	0.6148
0.8139	2.0	50000	0.9394	0.6092
0.7279	2.2	55000	0.9868	0.605
0.715	2.4	60000	0.9865	0.6072
0.7515	2.6	65000	0.9783	0.606
0.7363	2.8	70000	0.9765	0.6096
0.7405	3.0	75000	0.9723	0.61

Framework versions

Transformers 4.17.0
Pytorch 1.10.0+cu111
Datasets 2.0.0
Tokenizers 0.11.6