ResNet18-IM_ink2k Model

Overview

The ResNet18-IM_ink2k is a highly efficient image classification model trained on an extensive dataset of 2,000 ink-based images using the Gluon framework. This model leverages the power of Residual Networks, ensuring swift convergence during training and outstanding performance in real-world scenarios.

Features

• Pre-trained on a diverse dataset, ensuring robust recognition of various ink patterns.
• Utilizes the Gluon framework for a seamless training and inference experience.
• Built upon the robust ResNet-18 architecture known for its excellent balance of performance and computational efficiency.

Requirements

• PyTorch 1.8.0 or higher
• Gluon 0.10.0 or higher

Usage

Clone this repository and navigate to the project directory:

git clone https://huggingface.co/PretrainingFactory/resnet18-im_ink2k.gluon.18k
cd resnet18-im_ink2k.gluon.18k

License

This model is licensed under the Apache-2.0 License. Please see the LICENSE file for more details.

Contribution

We welcome contributions to improve this model. Please feel free to open an issue or submit a pull request.

Citation

If you use this model in your work, please cite:

@misc{resnet18-im_ink2k,
  author = {PretrainingFactory},
  title = {ResNet18-IM_ink2k: A Robust Ink Image Classifier},
  year = {2023},
  publisher = {Hugging Face},
  howpublished = {\url{https://huggingface.co/PretrainingFactory/resnet18-im_ink2k.gluon.18k}}
}