hazarsozer/distilbert-finetuned-imdb-sentiment

Model description

This model is a fine-tuned version of distilbert-base-uncased on an IMDB dataset.

Intended uses & limitations

This model can be used for sentiment analysis of comments, and general texts. But since it is trained on an IMDB dataset it might not perform as well on other text than movie comments.

Training and evaluation data

This model has trained and evulauated on labeled IMDB data.

Training procedure

It has went through supervised training, as the labeled imdb data first tokenized then fed into the model with small batches.

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Use adamw_torch_fused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 1

Training results

It achieves following sentiment analysis results on the evaluation dataset:

• Loss: 0.1973
• Accuracy: 0.9279
• Precision: 0.9369
• Recall: 0.9175
• F1: 0.9271

Framework versions

• Transformers 4.55.0
• Pytorch 2.8.0+cu128
• Datasets 4.0.0
• Tokenizers 0.21.4