Skip to content

Latest commit

 

History

History
39 lines (35 loc) · 1.14 KB

README.md

File metadata and controls

39 lines (35 loc) · 1.14 KB

Model training using TF's object detection API

Given that this project primarily uses scripts, it's helpful to have an idea of the directory structure.

.
├── annotations
├── external
│   ├── labelImg
│   │   ├── build-tools
│   │   ├── data
│   │   ├── demo
│   │   ├── libs
│   │   ├── readme
│   │   ├── requirements
│   │   ├── resources
│   │   ├── tests
│   │   └── tools
│   └── models
│       ├── community
│       ├── official
│       ├── orbit
│       └── research
├── models
│   ├── export
│   └── pretrained
├── records
├── resources
│   └── images
│       ├── test
│       └── train
└── scripts

Now that that is out of the way - the project houses three notebooks that build on each other. The notebooks are self-documnting and should be reviewed in the order:

  1. train.ipynb
  2. detect.ipynb
  3. export.ipynbe