📝 Model Card: bart-imdb-finetuned

🔍 Introduction

The wakaflocka17/bart-imdb-finetuned model is a fine-tuned version of facebook/bart-base for the sentiment classification task on the IMDb dataset. Trained on movie reviews, it can distinguish between positive and negative sentiment with excellent accuracy. Below you will find its model card, evaluation metrics, training parameters, and a practical example of its use in Google Colab.

📊 Evaluation Metrics

Metric	Value
Accuracy	0.87968
Precision	0.8839
Recall	0.8742
F1-score	0.8790

⚙️ Training Parameters

Parameter	Values
Base model	facebook/bart-base
Repo pretrained	facebook/bart-base
Repo finetuned	models/bart_base
Repo downloaded	models/downloaded/bart_base
Epochs	3
Batch size (train)	8
Batch size (eval)	16
Labels number	2

🚀 Example of use in Colab

Installing dependencies

!pip install --upgrade transformers huggingface_hub

(Optional) Authentication for private models.

from huggingface_hub import login
login(token="hf_yourhftoken")

Loading tokenizer and model

from transformers import AutoTokenizer, AutoModelForSequenceClassification, TextClassificationPipeline

repo_id   = "wakaflocka17/bart-imdb-finetuned"
tokenizer = AutoTokenizer.from_pretrained(repo_id)
model     = AutoModelForSequenceClassification.from_pretrained(repo_id)

# Override default labels
model.config.id2label = {0: 'NEGATIVE', 1: 'POSITIVE'}
model.config.label2id = {'NEGATIVE': 0, 'POSITIVE': 1}

# Create the classification pipeline
pipe = TextClassificationPipeline(model=model, tokenizer=tokenizer, return_all_scores=True)

Inference on a text example

testo     = "This movie was absolutely fantastic—wonderful performances and a gripping story!"
risultati = pipe(testo)
print(risultati)
# Example output:
# [{'label': 'POSITIVE', 'score': 0.95}, {'label': 'NEGATIVE', 'score': 0.05}]

📖 How to cite If you use this model in your work, you can cite it as:

@misc{Sentiment-Project,
  author       = {Francesco Congiu},
  title        = {Sentiment Analysis with Pretrained, Fine-tuned and Ensemble Transformer Models},
  howpublished = {\url{https://github.com/wakaflocka17/DLA_LLMSANALYSIS}},
  year         = {2025}
}

🔗 Reference Repository

All the file structure and script examples can be found at: https://github.com/wakaflocka17/DLA_LLMSANALYSIS/tree/main