---
language:
- en
license: apache-2.0
tags:
- tensorflow
- keras
- computer-vision
- medical-imaging
- brain-tumor
- mobilevit
- image-classification
datasets:
- brain-tumor-mri
metrics:
- accuracy
model-index:
- name: MobileViT Brain Tumor Classifier
  results:
  - task:
      type: image-classification
      name: Brain Tumor Classification
    dataset:
      type: brain-tumor-mri
      name: Brain Tumor MRI Images
    metrics:
    - type: accuracy
      value: 0.9850
      name: Accuracy
---

# MobileViT Brain Tumor Classifier

This MobileViT model classifies brain MRI scans into:

- Healthy
- Tumor

Accuracy: **98.5%**

⚠️ **Note**: For research/educational purposes only. Not for clinical use.

## Model Files

- `model.keras`: Native Keras format (recommended)
- `model.h5`: Legacy H5 format
- `saved_model/`: TensorFlow SavedModel format
- `model.weights.h5`: Model weights only
- `model_config.json`: Model architecture configuration
- `class_names.json`: Class label mappings

## Usage

```python
import tensorflow as tf
from huggingface_hub import hf_hub_download

# Download and load model
model_path = hf_hub_download(repo_id="abdo1176/brain-model-test", filename="model.keras")
model = tf.keras.models.load_model(model_path)

# Or load weights only
weights_path = hf_hub_download(repo_id="abdo1176/brain-model-test", filename="model.weights.h5")
# model.load_weights(weights_path)
```