JobPredictor2 / README.md
akzaidan's picture
Upload folder using huggingface_hub
22472aa verified
---
language: en
tags:
- job-classification
- salary-prediction
- experience-prediction
- deberta
---
# JobPredictor1
Fine-tuned DeBERTa-v3-base model that predicts:
- **Expected years of experience** required for a job
- **Expected salary** (USD)
## Input Format
```
[LOCATION] <Remote | United States (State) | Country> [TITLE]: <job title> [DESC]: <job description>
```
## Outputs
| Output | Type | Description |
|---|---|---|
| expected_experience_years | int | Years of experience required |
| expected_salary | int | Expected salary (USD) |
## Normalization
Experience is z-score normalized:
```python
real_value = pred * norm_stats["expected_experience_years"]["std"] + norm_stats["expected_experience_years"]["mean"]
```
Salary is log1p + z-score normalized:
```python
real_salary = np.expm1(pred * norm_stats["expected_salary"]["std"] + norm_stats["expected_salary"]["mean"])
```
## Test Set Performance
| Metric | Value |
|---|---|
| Experience MAE | 0.57 years |
| Experience Within 1yr | 83.1% |
| Salary MAE | $15,511 |
| Salary Within $20k | 84.5% |
## Base Model
microsoft/deberta-v3-base