Skip to content

API_tutorial.md

Latest commit

 

History

History
35 lines (25 loc) · 1017 Bytes

API_tutorial.md

File metadata and controls

35 lines (25 loc) · 1017 Bytes

Tensorflow object detection API with custom dataset

Tensorflow object detection on custom dataset

For korean version, check out README_KOR.md

This project was conducted based on the official document and tutorial from tensorflow.
official document link: https://tensorflow-object-detection-api-tutorial.readthedocs.io/en/latest/
tutorials: https://github.com/tensorflow/models/tree/master/research/object_detection/

Basic Settings

environment: Ubuntu 18.04
Cuda:10.1
CUDNN:7.6.5
tensorflow version: 1.14(when the api only supported tf1) -> 2.3

  • How to check the version if Cuda and CUDNN?

  • for windows environment check out, I found it useful when I tried in windows.

Datasets

Process

Detailed steps for each process can be found in official documentation. 1) 2) 3) 4) 5)

Performance

Error log

I recorded all the errors I faced while using api. Those error logs can be found here.
Check out stackoverflow or issus tab in official git repositories when you face unknown error!