The incredible module mmsegmentation that we can use the more green segmentation model as much as we want. Also this latest model?
A Python module that makes it easy to use the latest models just announced
It is a module that can use the latest model like “announced this year” with a simple code.
Machine learning is making progress every day, but it seems difficult to catch up and use it.
A machine learning model in which papers are published steadily and accuracy is steadily increasing.
The code will be released, but it’s hard to visit GitHub and run the code every time it’s announced. .. ..
Do we have to refer to the code for each study?
Even if we try to compare which model is better for our service, the
code format may differ for each project, such as the execution method, execution environment, and required modules, so it takes time to refer to each one.
With mmsegmentation, it’s easy, just select a model and execute it in one shot
mmsegmentation supports many models and can be used in the same execution format.
It supports many new models up to the model announced this year.
Moreover, it can be executed with less code.
It supports the following models. At the time of writing the article, it also includes the most accurate model at the dissertation level. Supports all compatible data sets.
How to use
1. Install mmsegmentation according to the official procedure. You can do it with pip.
mmsegmentation/get_started.md at master · open-mmlab/mmsegmentation
Linux or macOS (Windows is in experimental support) Python 3.6+ PyTorch 1.3+ CUDA 9.2+ (If you build PyTorch from…
2. You can get the config file and checkpoint file from the official lipo model zoo.
Official sample colab:
Compare the old and new models and easily use the one that suits your service.
With mmsegmentation, you can access the latest models and choose the one that suits your service.
I think it’s very convenient to be able to use it in a unified module.
I’m a freelance engineer.
Please feel free to contact us with a brief development description.
I am making an app that uses Core ML and ARKit.
We send machine learning / AR related information.