echoboi/veganism_and_vegetarianism-distilbert-classifier

Veganism & Vegetarianism Classifier (Distilbert)

This model classifies content related to veganism and vegetarianism on climate change subreddits.

Model Details

• Model Type: Distilbert
• Task: Multilabel text classification
• Sector: Veganism & Vegetarianism
• Base Model: Distilbert base uncased
• Labels: 7
• Training Data: Sample from 1000 GPT 4o-mini-labeled Reddit posts from climate subreddits (2010-2023)

Labels

The model predicts 7 labels simultaneously:

1. Animal Welfare: Cites animal suffering, cruelty, or ethics as motivation.
2. Environmental Impact: Links diet choice to climate change, land, water, or emissions.
3. Health: Claims physical health benefits or risks of eating less meat / going vegan.
4. Lab Grown And Alt Proteins: References cultivated meat, precision fermentation, insect protein or plant-based substitutes.
5. Psychology And Identity: Diet as part of personal identity, moral virtue signalling or tribal politics.
6. Systemic Vs Individual Action: Calls for policy, corporate reform or large-scale funding instead of just personal diet shifts.
7. Taste And Convenience: Talks about flavour, texture, cooking ease, availability of vegan options, or social convenience.

Note: Label order in predictions matches the order above.

Usage

import torch, sys, os, tempfile
from transformers import DistilBertTokenizer
from huggingface_hub import snapshot_download

device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')

def print_sorted_label_scores(label_scores):
    # Sort label_scores dict by score descending
    sorted_items = sorted(label_scores.items(), key=lambda x: x[1], reverse=True)
    for label, score in sorted_items:
        print(f"  {label}: {score:.6f}")

# Model link and examples for this specific model
model_link = 'sanchow/veganism_and_vegetarianism-distilbert-classifier'
examples = [
    "Plant-based diets have a much lower carbon footprint than meat-heavy diets."
]

print(f"\n{'='*60}")
print("MODEL: VEGANISM & VEGETARIANISM SECTOR")
print(f"{'='*60}")

print(f"Downloading model: {model_link}")
with tempfile.TemporaryDirectory() as temp_dir:
    snapshot_download(
        repo_id=model_link,
        local_dir=temp_dir,
        local_dir_use_symlinks=False
    )
    model_class_path = os.path.join(temp_dir, 'model_class.py')
    if not os.path.exists(model_class_path):
        print(f"model_class.py not found in downloaded files")
        print(f"   Available files: {os.listdir(temp_dir)}")
    else:
        sys.path.insert(0, temp_dir)
        from model_class import MultilabelClassifier
        tokenizer = DistilBertTokenizer.from_pretrained(temp_dir)
        checkpoint = torch.load(os.path.join(temp_dir, 'model.pt'), map_location='cpu', weights_only=False)
        model = MultilabelClassifier(checkpoint['model_name'], len(checkpoint['label_names']))
        model.load_state_dict(checkpoint['model_state_dict'])
        model.to(device)
        model.eval()
        print("Model loaded successfully")
        print(f"   Labels: {checkpoint['label_names']}")
        print("\nVeganism & Vegetarianism classifier results:\n")
        for i, test_text in enumerate(examples):
            inputs = tokenizer(
                test_text, 
                return_tensors="pt", 
                truncation=True, 
                max_length=512,
                padding=True
            ).to(device)
            with torch.no_grad():
                outputs = model(**inputs)
                predictions = outputs.cpu().numpy() if isinstance(outputs, (tuple, list)) else outputs.cpu().numpy()
            label_scores = {label: float(score) for label, score in zip(checkpoint['label_names'], predictions[0])}
            print(f"Example {i+1}: '{test_text}'")
            print("Predictions (all label scores, highest first):")
            print_sorted_label_scores(label_scores)
            print("-" * 40)

Performance

Best model performance:

• Micro Jaccard: 0.5584
• Macro Jaccard: 0.6710
• F1 Score: 0.8906
• Accuracy: 0.8906

Dataset: ~900 GPT-labeled samples per sector (600 train, 150 validation, 150 test)

Optimal Thresholds

optimal_thresholds = {'Animal Welfare': 0.48107979620047003, 'Environmental Impact': 0.45919171852850427, 'Health': 0.20115313966833437, 'Lab Grown And Alt Proteins': 0.3414601502146817, 'Psychology And Identity': 0.5246278637433214, 'Systemic Vs Individual Action': 0.37517437676211585, 'Taste And Convenience': 0.6635140143644325}
for label, score in zip(label_names, predictions[0]):
    threshold = optimal_thresholds.get(label, 0.5)
    if score > threshold:
        print(f"{label}: {score:.3f}")

Training

Trained on GPT-labeled Reddit data:

1. Data collection from climate subreddits
2. keyword based filtering for sector-specific content
3. GPT labeling for multilabel classification
4. 80/10/10 train/validation/test split
5. Fine-tuning with threshold optimization

Citation

If you use this model in your research, please cite:

@misc{veganism_and_vegetarianism_distilbert_classifier,
  title={Veganism & Vegetarianism Classifier for Climate Change Analysis},
  author={Sandeep Chowdhary},
  year={2025},
  publisher={Hugging Face},
  journal={Hugging Face Hub},
  howpublished={\url{https://huggingface.co/echoboi/veganism_and_vegetarianism-distilbert-classifier}},
}

Limitations

• Trained on data from specific climate change subreddits and limited to English content
• Performance depends on GPT-generated labels