depression-roberta

RoBERTa - SentiMind

Fine-tuned RoBERTa model for binary classification of depressive vs. non-depressive tweets.

Model Description

Robustly optimized BERT. Trained without Next Sentence Prediction.

Performance

Usage

from transformers import AutoTokenizer, AutoModelForSequenceClassification
import torch

model_id = "RaxzellMornov/depression-roberta"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForSequenceClassification.from_pretrained(model_id)
model.eval()

text = "I feel so empty and hopeless."
inputs = tokenizer(text, return_tensors="pt", truncation=True, max_length=128)

with torch.no_grad():
    logits = model(**inputs).logits

predicted_class = logits.argmax(-1).item()
labels = {0: "Non-Depressive", 1: "Depressive"}
print(labels[predicted_class])

Training Details

• Task: Binary sequence classification
• Framework: PyTorch + Hugging Face Transformers
• Fine-tuned on: Depressive Twitter dataset

Disclaimer

SentiMind is intended for research and educational purposes only. It should not be used as a substitute for professional mental health diagnosis.