File size: 3,425 Bytes
d8c7540 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 | # DistilBERT-Base-Uncased Quantized Model for Scientific Paper Classification
This repository hosts a quantized version of the **DistilBERT** model, fine-tuned for **scientific paper classification** into three categories: **Biology, Mathematics, and Physics**. The model has been optimized for efficient deployment while maintaining high accuracy, making it suitable for real-world applications, including academic research and automated categorization of scientific literature.
## Model Details
- **Model Architecture:** DistilBERT Base Uncased
- **Task:** Scientific Paper Classification
- **Dataset:** Custom dataset labeled with three categories: Biology, Mathematics, and Physics
- **Quantization:** Float16 (FP16)
- **Fine-tuning Framework:** Hugging Face Transformers
## Usage
### Installation
```sh
pip install transformers torch
```
### Loading the Model
```python
from transformers import DistilBertForSequenceClassification, DistilBertTokenizer
import torch
# Load quantized model
quantized_model_path = "/kaggle/working/distilbert_finetuned_fp16"
quantized_model = DistilBertForSequenceClassification.from_pretrained(quantized_model_path)
quantized_model.eval() # Set to evaluation mode
quantized_model.half() # Convert model to FP16
# Load tokenizer
tokenizer = DistilBertTokenizer.from_pretrained("distilbert-base-uncased")
# Define a test input
test_paper = "The quantum mechanics of atomic structures are governed by Schrödinger's equation."
# Tokenize input
inputs = tokenizer(test_paper, return_tensors="pt", padding=True, truncation=True, max_length=512)
# Ensure input tensors are in correct dtype
inputs["input_ids"] = inputs["input_ids"].long() # Convert to long type
inputs["attention_mask"] = inputs["attention_mask"].long() # Convert to long type
# Make prediction
with torch.no_grad():
outputs = quantized_model(**inputs)
# Get predicted class
predicted_class = torch.argmax(outputs.logits, dim=1).item()
# Class labels
label_mapping = {0: "Biology", 1: "Mathematics", 2: "Physics"}
predicted_label = label_mapping[predicted_class]
print(f"Predicted Label: {predicted_label}")
```
## Performance Metrics
- **Accuracy:** 0.95 (after fine-tuning)
- **F1-Score:** 0.91 (weighted)
## Fine-Tuning Details
### Dataset
The dataset consists of **scientific papers** categorized into three domains:
- **Biology**
- **Mathematics**
- **Physics**
The dataset was preprocessed and tokenized using the **DistilBERT tokenizer**.
### Training
- Number of epochs: 3
- Batch size: 8
- Learning rate: 2e-5
- Optimizer: AdamW
- Evaluation strategy: epoch
### Quantization
Post-training quantization was applied using PyTorch’s built-in quantization framework to reduce the model size and improve inference efficiency.
## Repository Structure
```
.
├── model/ # Contains the quantized model files
├── tokenizer_config/ # Tokenizer configuration and vocabulary files
├── model.safensors/ # Fine-Tuned Model
├── README.md # Model documentation
```
## Limitations
- The model is trained on a limited dataset and may not generalize well to niche scientific subdomains.
- Quantization may result in slight accuracy degradation compared to full-precision models.
## Contributing
Contributions are welcome! Feel free to open an issue or submit a pull request if you have suggestions or improvements.
|