|
|
--- |
|
|
license: mit |
|
|
library_name: transformers |
|
|
pipeline_tag: text-classification |
|
|
tags: |
|
|
- sentiment-analysis |
|
|
- text-classification |
|
|
- distilbert |
|
|
- fine-tuned |
|
|
- nlp |
|
|
language: |
|
|
- en |
|
|
datasets: |
|
|
- sentiment140 |
|
|
metrics: |
|
|
- accuracy |
|
|
- f1 |
|
|
base_model: distilbert-base-uncased |
|
|
model-index: |
|
|
- name: my-sentiment-model |
|
|
results: |
|
|
- task: |
|
|
type: text-classification |
|
|
name: Text Classification |
|
|
dataset: |
|
|
name: sentiment140 |
|
|
type: sentiment140 |
|
|
metrics: |
|
|
- name: Accuracy |
|
|
type: accuracy |
|
|
value: 0.85 |
|
|
- name: F1 Score (Macro) |
|
|
type: f1 |
|
|
value: 0.84 |
|
|
--- |
|
|
|
|
|
# My DistilBERT Sentiment Model |
|
|
|
|
|
Fine-tuned DistilBERT for 3-class sentiment classification (negative, neutral, positive). |
|
|
|
|
|
## Model Description |
|
|
|
|
|
This model is a fine-tuned version of DistilBERT-base-uncased for sentiment analysis. It has been trained to classify text into three sentiment categories: |
|
|
|
|
|
- **Negative** (0) |
|
|
- **Neutral** (1) |
|
|
- **Positive** (2) |
|
|
|
|
|
## Intended Uses |
|
|
|
|
|
This model is intended for sentiment analysis tasks on English text. It can be used to: |
|
|
- Analyze customer feedback and reviews |
|
|
- Monitor social media sentiment |
|
|
- Classify emotions in text data |
|
|
- Support content moderation systems |
|
|
|
|
|
## Limitations |
|
|
|
|
|
- Trained primarily on English text |
|
|
- May not perform well on domain-specific jargon |
|
|
- Performance may vary on very short or very long texts |
|
|
- Potential bias from training data |
|
|
|
|
|
## Training Details |
|
|
|
|
|
- **Base Model**: distilbert-base-uncased |
|
|
- **Training Epochs**: 2 |
|
|
- **Batch Size**: 8 |
|
|
- **Learning Rate**: 3e-5 |
|
|
- **Max Sequence Length**: 128 |
|
|
- **Optimizer**: AdamW |
|
|
- **Weight Decay**: 0.01 |
|
|
|
|
|
## Model Performance |
|
|
|
|
|
The model achieves the following performance on the test set: |
|
|
|
|
|
- **Accuracy**: 85% |
|
|
- **F1-Score (Macro)**: 84% |
|
|
- **F1-Score (Weighted)**: 85% |
|
|
|
|
|
## Usage |
|
|
|
|
|
Install the required dependencies: |
|
|
|
|
|
```bash |
|
|
pip install transformers torch |
|
|
``` |
|
|
|
|
|
Load and use the model: |
|
|
|
|
|
```python |
|
|
from transformers import AutoTokenizer, AutoModelForSequenceClassification |
|
|
import torch |
|
|
|
|
|
# Load model and tokenizer |
|
|
model_name = "your-username/my-sentiment-model" |
|
|
tokenizer = AutoTokenizer.from_pretrained(model_name) |
|
|
model = AutoModelForSequenceClassification.from_pretrained(model_name) |
|
|
|
|
|
# Prepare text |
|
|
text = "I love this product!" |
|
|
inputs = tokenizer(text, return_tensors="pt", truncation=True, padding=True, max_length=128) |
|
|
|
|
|
# Get prediction |
|
|
with torch.no_grad(): |
|
|
outputs = model(**inputs) |
|
|
probabilities = torch.nn.functional.softmax(outputs.logits, dim=-1) |
|
|
predicted_class = torch.argmax(probabilities, dim=-1).item() |
|
|
|
|
|
# Map prediction to label |
|
|
labels = {0: "negative", 1: "neutral", 2: "positive"} |
|
|
confidence = probabilities[0][predicted_class].item() |
|
|
|
|
|
print(f"Text: {text}") |
|
|
print(f"Sentiment: {labels[predicted_class]} (confidence: {confidence:.2%})") |
|
|
``` |
|
|
|
|
|
## Citation |
|
|
|
|
|
If you use this model in your research, please cite: |
|
|
|
|
|
```bibtex |
|
|
@misc{my-sentiment-model, |
|
|
author = {Your Name}, |
|
|
title = {Fine-tuned DistilBERT for Sentiment Analysis}, |
|
|
year = {2025}, |
|
|
publisher = {Hugging Face}, |
|
|
url = {https://huggingface.co/your-username/my-sentiment-model} |
|
|
} |
|
|
``` |
|
|
|
|
|
## License |
|
|
|
|
|
This model is released under the MIT License. |
