metadata
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-C2
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: 0.4230106879189448
- name: QWK
type: qwk
value: 0.4118587182355762
- name: Weighted Macro F1
type: f1
value: 0.4331935675507009
Model ID: phi4-balanced-C2
Results
| test_data | |
|---|---|
| eval_accuracy | 0.456522 |
| eval_RMSE | 60.911 |
| eval_QWK | 0.411859 |
| eval_Macro_F1 | 0.282007 |
| eval_Macro_F1_(ignoring_nan) | 0.423011 |
| eval_Weighted_F1 | 0.433194 |
| eval_Micro_F1 | 0.456522 |
| eval_HDIV | 0.0797101 |