Downloads · 30 days
0
FER-CNN
NipunD58/FER-CNN
FER-CNN is a machine learning model from NipunD58. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
Trained on FER2013, RAF-DB, and AffectNet. Merges all into a unified training pipeline. Uses a ResNet-style CNN with SE attention, Mixup, label smoothing, cosine annealing, and comprehensive graphical statistics. Accu…
Access
Public
Updated Aug 6, 2026
Repo size
47 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.pth45.6 MB · 96%
At a glance
- Access
- Public
- Created
- Aug 6, 2026
- Updated
- Aug 6, 2026
- SHA
- 1bcf16d0
Model card
From the Hugging Face model README
FER-CNN
Facial Emotion Recognition Convolutional Neural Network
Trained on FER2013, RAF-DB, and AffectNet. Merges all into a unified training pipeline. Uses a ResNet-style CNN with SE attention, Mixup, label smoothing, cosine annealing, and comprehensive graphical statistics. Accuracy - 56.71% Usage - Use
python predict.py path/to/image.jpg
for making predictions using pretrained model Use
python predict.py path/to/image.jpg
for training the model from scratch
- Created
- Aug 6, 2026
- Updated
- Aug 6, 2026