---
license: apache-2.0
tags:
  - classification
  - deep-learning
  - cnn
model-index:
  - name: Digit Recognizer
    results:
      - task:
          type: image-classification
          name: Image Classification
        dataset:
          name: Kaggle - MNIST dataset
          type: mnist
          link: https://www.kaggle.com/competitions/digit-recognizer/data
        metrics:
          - type: accuracy
            value: 0.985
            name: Accuracy
---

# Digit Recognizer v1.0.0

This repository hosts the trained model for **digit recognition** in images. The model is a CNN-based architecture designed to classify images containing single digits between 0 and 9.

## Model Details

- **Architecture:** A CNN model that classifies handwritten digits between 0 and 9.
- **Dataset:** [Kaggle - MNIST dataset](https://www.kaggle.com/c/digit-recognizer/data).
- **Version:** v1.0.0
- **Task:** Image Classification
- **License:** Apache 2.0

## Usage

To use this model for inference, you can load it using the `tensorflow` library.

Requires: [Pip](https://pypi.org/project/pip/)

```bash
# Clones the repository and installs dependencies
!git clone https://huggingface.co/preethamganesh/digit-recognizer-v1.0.0
!pip install tensorflow

# Imports TensorFlow
import tensorflow as tf

# Loads the pre-trained model from the cloned directory
model_path = "digit-recognizer-v1.0.0"
exported_model = tf.saved_model.load(model_path)

# Retrieves the default serving function from the loaded model
model = exported_model.signatures["serving_default"]

# Prepares a dummy input tensor for inference (batch size: 1, height: 28, width: 28, channels: 1)
input_data = tf.ones((1, 28, 28, 1), dtype=tf.float32)

# Performs inference using the model. The output will be a dictionary, with the classification logits in the key 'output_0'
output = model(input_data)["output_0"]

# Prints the predicted class (e.g., 0 for normal, 1 for abnormal)
predicted_digit = tf.argmax(output, axis=-1).numpy()[0]
print("Predicted digit: ", predicted_digit)
```

## Training Details

### Compute

- The model was trained on a GeForce 4070Ti GPU with 16GB VRAM.
- Training completed in approximately 20.3 seconds over 9 epochs.

### Dataset

- The model was trained on the [Kaggle - MNIST dataset](https://www.kaggle.com/c/digit-recognizer/data), which includes images containing digits between 0 - 9.

### Performance on test set

- **Accuracy:** 0.985

## Citation

If you use this model in your research, please cite the repository:

```bash
@misc{preethamganesh2024digitrecog,
  title={Digit Recognizer - v1.0.0},
  author={Preetham Ganesh},
  year={2025},
  url={https://huggingface.co/preethamganesh/digit-recognizer-v1.0.0},
  note={Apache-2.0 License}
}
```

## Contact

For any questions or support, please contact preetham.ganesh2021@gmail.com.