metadata

license: mit
base_model: microsoft/mdeberta-v3-base
tags:
  - generated_from_trainer
datasets:
  - tweet_sentiment_multilingual
metrics:
  - accuracy
  - f1
model-index:
  - name: scenario-MDBT-TCR_data-cardiffnlp_tweet_sentiment_multilingual_all
    results:
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: tweet_sentiment_multilingual
          type: tweet_sentiment_multilingual
          config: all
          split: validation
          args: all
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.6512345679012346
          - name: F1
            type: f1
            value: 0.6483011417314103

scenario-MDBT-TCR_data-cardiffnlp_tweet_sentiment_multilingual_all

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

Loss: 1.7268
Accuracy: 0.6512
F1: 0.6483

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: 64
seed: 66
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
0.8937	1.09	500	0.8922	0.6304	0.6189
0.6912	2.17	1000	0.8900	0.6551	0.6516
0.527	3.26	1500	0.9088	0.6593	0.6583
0.3874	4.35	2000	1.1089	0.6516	0.6470
0.2977	5.43	2500	1.2137	0.6408	0.6433
0.2397	6.52	3000	1.2022	0.6431	0.6409
0.203	7.61	3500	1.4913	0.6454	0.6469
0.1658	8.7	4000	1.7268	0.6512	0.6483

Framework versions

Transformers 4.33.3
Pytorch 2.1.1+cu121
Datasets 2.14.5
Tokenizers 0.13.3