| --- |
| library_name: transformers |
| tags: |
| - text-classification |
| - bert |
| - sentiment-analysis |
| license: mit |
| language: |
| - en |
| metrics: |
| - accuracy |
| - f1 |
| - precision |
| - recall |
| base_model: |
| - google-bert/bert-base-uncased |
| pipeline_tag: text-classification |
| --- |
| |
|
|
| ## Model Details |
|
|
| - **Model type**: BERT-based sequence classification |
| - **Base model**: `bert-base-uncased` |
| - **Number of classes**: 3 |
| - **Trained on**: Custom labeled dataset |
| - **Framework**: PyTorch with 🤗 Transformers |
| - **Max sequence length**: 512 |
| - **Tokenizer**: `bert-base-uncased` |
|
|
| ### Model Description |
|
|
| This is a fine-tuned `bert-base-uncased` model for **multi-class text classification**. The model was trained on a custom dataset to classify text into 3 categories: `Negative`, `Neutral`, and `Positive`. |
|
|
| - **Developed by:** Divyansh Rajput |
| - **Model type:** Transformer(Bert-Base-Uncased) |
| - **Language(s) (NLP):** English |
| - **License:** Mit |
| - **Finetuned from model :** Bert-Base-Uncased |
|
|
| ### Model Sources |
|
|
| - **Repository:** https://huggingface.co/divyansh126/SentimentAnalysisBert/ |
|
|
| ## 📊 Classes |
|
|
| | Label | Description | |
| |-------|-------------| |
| | 0 | Negative | |
| | 1 | Neutral | |
| | 2 | Positive | |
|
|
| ## 🚀 How to Get Started with the Model## |
|
|
| from transformers import AutoTokenizer, AutoModelForSequenceClassification |
|
|
| tokenizer = AutoTokenizer.from_pretrained("divyansh126/") |
| model = AutoModelForSequenceClassification.from_pretrained("your-username/your-model-name") |
|
|
| inputs = tokenizer("I loved the product!", return_tensors="pt") |
| outputs = model(**inputs) |
| predicted_class = outputs.logits.argmax(dim=-1).item() |
| |
| |
| ## 🧪 Training Details |
| |
| - **Training framework**: Hugging Face `Trainer` |
| - **Epochs**: 1 |
| - **Batch size**: 32 |
| - **Learning rate**: 2e-5 |
| - **Device**: Trained on GPU (`cuda`) |
| |
| |
| ### Training Data |
| |
| Dataset Link - https://www.kaggle.com/datasets/jp797498e/twitter-entity-sentiment-analysis |
| |
| ## Evaluation |
| |
| ## 📈 Evaluation |
| |
| | Metric | Score | |
| |------------|---------| |
| | Accuracy | 0.85 | |
| | F1 Score | 0.85 | |
| | Precision | 0.85 | |
| | Recall | 0.85 | |
| |
| #### Summary |
| |
| This model is a fine-tuned version of BERT (bert-base-uncased) for multi-class text classification. It has been trained to classify input text into three sentiment-based categories: Negative (0), Neutral (1), and Positive (2). |
| The model was trained on a custom dataset using the Hugging Face Trainer API with PyTorch. It uses standard preprocessing with a maximum sequence length of 512 tokens. This fine-tuned BERT model achieves strong performance in sentiment classification tasks and is suitable for real-time predictions in production environments or APIs. |
