the-full-stack/fsdl-text-recognizer-2022-labs
Full Stack Deep Learning 2022 course labs: incrementally building a CNN/Transformer-based handwriting recognizer deployed as a Docker container.
Not currently ranked — collecting fresh signals.
star history
This repository contains Jupyter notebooks from the Full Stack Deep Learning 2022 course labs, guiding students through building a complete deep learning application for recognizing handwritten text in paragraphs. The project uses PyTorch and PyTorch Lightning to implement CNNs and Transformers for the recognition model, Weights & Biases for experiment tracking, and deploys the model as a Docker container on AWS Lambda. The pipeline includes pre-commit hooks, GitHub Actions CI, Gradio for the frontend interface, and Gantry for model monitoring.
Frequently asked
- What is the-full-stack/fsdl-text-recognizer-2022-labs?
- Full Stack Deep Learning 2022 course labs: incrementally building a CNN/Transformer-based handwriting recognizer deployed as a Docker container.
- Is fsdl-text-recognizer-2022-labs open source?
- Yes — the-full-stack/fsdl-text-recognizer-2022-labs is open source, released under the MIT license.
- What language is fsdl-text-recognizer-2022-labs written in?
- the-full-stack/fsdl-text-recognizer-2022-labs is primarily written in Jupyter Notebook.
- How popular is fsdl-text-recognizer-2022-labs?
- the-full-stack/fsdl-text-recognizer-2022-labs has 530 stars on GitHub.
- Where can I find fsdl-text-recognizer-2022-labs?
- the-full-stack/fsdl-text-recognizer-2022-labs is on GitHub at https://github.com/the-full-stack/fsdl-text-recognizer-2022-labs.