Downloads · 30 days
0
mikagross/cnn-fashion-mnist-augmented-92acc
cnn-fashion-mnist-augmented-92acc is a machine learning model from mikagross. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
This is a custom Convolutional Neural Network (CNN) trained from scratch to classify clothing items using the FashionMNIST dataset.
Downloads · 30 days
0
Access
Public
Updated Aug 14, 2026
Parameters
422K
1.7 MB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors1.7 MB · 100%
From the Hugging Face model README
This is a custom Convolutional Neural Network (CNN) trained from scratch to classify clothing items using the FashionMNIST dataset.
It achieves 92.7% accuracy on the test set, utilizing custom data augmentation and early-stopping checkpointing.
To achieve the 92.7% accuracy, the model was trained with the following configuration to prevent overfitting and improve generalization:
v2.RandomHorizontalFlip(p=0.5) and v2.RandomRotation(degrees=10) to the training pipeline.The model was evaluated on the standard 10,000-image FashionMNIST test split (with no augmentations applied during testing).
The model outputs a tensor of 10 logits corresponding to the following classes:
0: T-shirt/top, 1: Trouser, 2: Pullover, 3: Dress, 4: Coat, 5: Sandal, 6: Shirt, 7: Sneaker, 8: Bag, 9: Ankle boot
You can load this model directly from the Hugging Face Hub using the huggingface_hub mixin.
(Note: You must define the CNNModel class in your script before loading).
from your_script import CNNModel
# Load the trained weights securely via Safetensors
model = CNNModel.from_pretrained("mikagross/cnn-fashion-mnist-augmented-92acc")
model.eval()