Instructions to use kamel-usp/jbcs2025_phi4-balanced-C1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use kamel-usp/jbcs2025_phi4-balanced-C1 with PEFT:
from peft import PeftModel from transformers import AutoModelForSequenceClassification base_model = AutoModelForSequenceClassification.from_pretrained("microsoft/phi-4") model = PeftModel.from_pretrained(base_model, "kamel-usp/jbcs2025_phi4-balanced-C1") - Notebooks
- Google Colab
- Kaggle
File size: 1,267 Bytes
877cacf 81535f0 877cacf 81535f0 877cacf 81535f0 877cacf 81535f0 877cacf | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 |
---
language:
- pt
- en
tags:
- aes
datasets:
- kamel-usp/aes_enem_dataset
base_model: microsoft/phi-4
metrics:
- accuracy
- qwk
library_name: peft
model-index:
- name: phi4-balanced-C1
results:
- task:
type: text-classification
name: Automated Essay Score
dataset:
name: Automated Essay Score ENEM Dataset
type: kamel-usp/aes_enem_dataset
config: JBCS2025
split: test
metrics:
- name: Macro F1 (ignoring nan)
type: f1
value: 0.5725084663763909
- name: QWK
type: qwk
value: 0.6652455160246986
- name: Weighted Macro F1
type: f1
value: 0.6013651374602975
---
# Model ID: phi4-balanced-C1
## Results
| | test_data |
|:-----------------------------|------------:|
| eval_accuracy | 0.601449 |
| eval_RMSE | 27.6626 |
| eval_QWK | 0.665246 |
| eval_Macro_F1 | 0.458007 |
| eval_Macro_F1_(ignoring_nan) | 0.572508 |
| eval_Weighted_F1 | 0.601365 |
| eval_Micro_F1 | 0.601449 |
| eval_HDIV | 0.00724638 |
|