---
license: mit
library_name: pytorch
tags:
- image-classification
- cifar10
- cnn
- pytorch
- ai-generated
datasets:
- uoft-cs/cifar10
metrics:
- accuracy
model-index:
- name: cifar10-cnn-ai-made
  results:
  - task: {type: image-classification}
    dataset: {name: CIFAR-10, type: uoft-cs/cifar10}
    metrics:
    - {type: accuracy, value: 0.9152}
---

# CIFAR-10 CNN — 🤖 Fully AI-Made

> **This model — the architecture, training code, evaluation, plots, and this very card — was created end-to-end by an AI agent (Claude, via the Hermes ML research loop).** No human wrote the code. Kaggle T4 GPU did the compute; the artifact was pushed to the Hugging Face Hub automatically.

A compact 3-block VGG-style CNN trained from scratch on [CIFAR-10](https://huggingface.co/datasets/uoft-cs/cifar10).

## Results

| metric | value |
|---|---|
| **Test accuracy** | **91.52%** |
| Test loss | 0.542 |
| Best val accuracy | 91.52% |
| Parameters | 1.15M |
| Epochs | 18 |
| Train time (T4) | 5.9 min |

### Training curves
![training curves](training_curves.png)

### Confusion matrix
![confusion matrix](confusion_matrix.png)

### Per-class accuracy
![per-class accuracy](per_class_accuracy.png)

| class | accuracy |
|---|---|
| automobile | 96.0% |
| truck | 95.2% |
| frog | 94.9% |
| ship | 94.6% |
| horse | 93.7% |
| airplane | 92.7% |
| deer | 91.7% |
| dog | 86.5% |
| bird | 86.5% |
| cat | 83.4% |

### Sample predictions
![samples](sample_predictions.png)

## Architecture
3 conv blocks (64→128→256 ch, each = 2×[Conv3×3–BN–ReLU]+MaxPool) → GAP → Dropout(0.3) → Linear(10).
AdamW + OneCycleLR, AMP, label smoothing 0.05, random crop + h-flip augmentation.

## Usage
```python
import torch, torch.nn as nn, torch.nn.functional as F
# (define the CNN class as in config.json), then:
sd = torch.load("cifar10_cnn.pt", map_location="cpu")
model.load_state_dict(sd); model.eval()
```

## Provenance
- Dataset: `uoft-cs/cifar10`
- Compute: Kaggle Notebook, 1× NVIDIA Tesla T4
- Pipeline: Hermes `ml-research-loop` (Kaggle GPU → Hugging Face)
- Author: AI agent (Claude). Human involvement: pressed go.