metadata
library_name: transformers
tags:
- text-classification
- modernbert
- generated-data
base_model: xlm-roberta-large
metrics:
- name: loss
type: loss
value: 0.4027690887451172
- name: accuracy
type: accuracy
value: 0.883
- name: f1
type: f1
value: 0.8826794072627964
- name: precision
type: precision
value: 0.8852544531362957
- name: recall
type: recall
value: 0.8831381197007152
- name: runtime
type: runtime
value: 43.8849
- name: samples_per_second
type: samples_per_second
value: 273.443
- name: steps_per_second
type: steps_per_second
value: 34.18
- name: epoch
type: epoch
value: 3
Gender Classifier (Fine-tuned xlm-roberta-large)
This model was fine-tuned to classify text into: male, female, neutral
Performance Metrics
| Metric | Value |
|---|---|
| loss | 0.4028 |
| accuracy | 0.8830 |
| f1 | 0.8827 |
| precision | 0.8853 |
| recall | 0.8831 |
| runtime | 43.8849 |
| samples_per_second | 273.4430 |
| steps_per_second | 34.1800 |
| epoch | 3.0000 |
Hyperparameters
- Batch Size: 8
- Learning Rate: 5e-05
- Epochs: 3
- Weight Decay: 0.01
- Mixed Precision (FP16): True
Quick Usage
from transformers import pipeline
# Load the model directly from this folder or HF Hub
classifier = pipeline('text-classification', model='.')
print(classifier('She is a great engineer.'))