Text Classification
Transformers
Safetensors
roberta
mental-health
multi-label
mental-roberta-base
text-embeddings-inference
Instructions to use mahir2021/mental-roberta-depEMO-v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use mahir2021/mental-roberta-depEMO-v2 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="mahir2021/mental-roberta-depEMO-v2")# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForSequenceClassification tokenizer = AutoTokenizer.from_pretrained("mahir2021/mental-roberta-depEMO-v2") model = AutoModelForSequenceClassification.from_pretrained("mahir2021/mental-roberta-depEMO-v2", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Mental-RoBERTa
Multi-label emotion classifier trained on depression-related text.
- Base model:
mental/mental-roberta-base - Architecture: Standard AutoModelForSequenceClassification
- Labels (8): anger, brain dysfunction (forget), emptiness, hopelessness, loneliness, sadness, suicide intent, worthlessness
- Seed: 42, Epochs: 5, LR: 2e-05, MaxLen: 256
Test metrics
{
"eval_loss": 0.7149868607521057,
"eval_P-Mic": 0.8186006334581054,
"eval_R-Mic": 0.8499252615844544,
"eval_F1-Mic": 0.833968905837489,
"eval_P-Mac": 0.7866288741283867,
"eval_R-Mac": 0.7956996729339153,
"eval_F1-Mac": 0.7842684893606211
}
- Downloads last month
- 54