metadata

license: apache-2.0
base_model: distilbert-base-uncased
tags:
  - generated_from_trainer
datasets:
  - twitter-sentiment-analysis
metrics:
  - accuracy
  - f1
model-index:
  - name: twitter-sentiment-analysis-v2
    results:
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: twitter-sentiment-analysis
          type: twitter-sentiment-analysis
          config: default
          split: test
          args: default
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.8366721507145392
          - name: F1
            type: f1
            value: 0.8366721507145392

twitter-sentiment-analysis-v2

This model is a fine-tuned version of distilbert-base-uncased on the twitter-sentiment-analysis dataset. It achieves the following results on the evaluation set:

Loss: 0.3771
Accuracy: 0.8367
F1: 0.8367

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: 16
eval_batch_size: 16
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: reduce_lr_on_plateau
lr_scheduler_warmup_steps: 500
num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
0.3957	0.13	1000	0.4273	0.8075	0.8005
0.4086	0.27	2000	0.4081	0.8211	0.8139
0.4085	0.4	3000	0.3971	0.8274	0.8237
0.3936	0.53	4000	0.3857	0.8304	0.8307
0.3783	0.67	5000	0.3978	0.8317	0.8300
0.3858	0.8	6000	0.3887	0.8281	0.8182
0.3779	0.93	7000	0.3771	0.8367	0.8367
0.2971	1.07	8000	0.4023	0.8352	0.8310
0.2994	1.2	9000	0.3865	0.8326	0.8342
0.293	1.33	10000	0.4454	0.8299	0.8197
0.3053	1.47	11000	0.3929	0.8364	0.8349
0.3125	1.6	12000	0.4141	0.8366	0.8314

Framework versions

Transformers 4.38.2
Pytorch 2.1.0+cu121
Datasets 2.18.0
Tokenizers 0.15.2