This is a fork of GabrielDornelles/pytorch-ocr.
- Use dataclasses for config
- Use SQLite database for labelling (with a crappy GUI classifier)
- Experimental ONNX support
- Limited Tensorboard logging support
- Experimental Docker support
- Removed some fancy log messages during training
- Quantization?
- ./docker/build.sh: build images we need
- ./docker/train.sh: start training session and a tensorboard session listening on port 6006
- ./docker/interactive.sh: an interactive bash shell for custom commands
Docker compose isn't suitable for one-time tasks like this. However we do have a sample docker-compose.sample.yml under the root directory, you can copy and create your own docker-compose.yml in your flavor.
Before starting, it's suggested to run docker compose build to build images.
To start training, try docker compose up. This will start python train.py immediately, and a tensorboard session listening on port 6006.
If you want an interactive shell, run docker compose run --rm interactive.
The original project, pytorch-ocr, is licensed under the MIT license, Copyright (c) 2021 Nanohana. See LICENSE for details.
Copyright (c) 2024 Lin He
This file is part of Pytorch CRNN OCR model, along with all the accessories included in the project, such as samples classifier, dataset loader, model trainer and so on (all referred as "this software" below).
This software is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
This software is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.
You should have received a copy of the GNU General Public License along with this software. If not, see https://www.gnu.org/licenses/.