Skip to content

Latest commit

 

History

History
15 lines (8 loc) · 781 Bytes

README.md

File metadata and controls

15 lines (8 loc) · 781 Bytes

YOLO Car Detection

Week 3 project from Coursera's "Convolutional Neural Networks" course (DeepLearning.AI Deep Learning Specialization, 2018). Tested on a conda environment with TensorFlow 1.14.0 + Keras.

Important: you have to put the file yolo.h5 in the model_data folder. The file is ~196MB and can be obtained following these instructions:

  1. clone or download YAD2K (https://github.com/allanzelener/YAD2K);

  2. download weights and cfg files (YOLOv2 608x608) from https://pjreddie.com/darknet/yolo ;

  3. put the files in the YAD2K-master folder, open a shell, activate a TensorFlow 1.X + Keras environment and type

python yad2k.py yolov2.cfg yolov2.weights model_data/yolo.h5 ;

  1. put the yolo.h5 file in the model_data folder of the Car Detection project.