Logs/000/trained_weights_final.h5 placement after training weight
Keras-yolo3-master
Keras/tensorflow + Python + yolo3 train your own dataset
Code: https://github.com/qqwweee/keras-yolo3
Modify the yolov3.cfg file: 79695109
Use yolo3 to train your own dataset for Target Detection
Vocdevkit/voc2007/Annotations XML file
Vocdevkit/voc2007/javasimages jpgimage
Four files under vocdevkit/voc2007/imagesets/Main, create the file test. py under voc2007,
Run voc_annotation.py, modify classes, and set the running result to 2007_train.txt. manually delete 2007_and generate train.txt,
Modify the parameter file: yolo3.cfg
Filters for each of the three Yolo modifications: 3*(5 + Len (classes ));
Classes: Len (classes)
Random: the original value is 1, and the memory size is changed to 0,
Modify the model_datafile, coco_classes.txt: Put the class name voc_classes.txt: Put the class name, be sure not to have blank lines or spaces
Training:
Python convert. py-W yolov3.cfg yolov3.weights model_data/yolo_weights.h5
Python train. py
Test:
Python Yolo. py or Python yolo_video.pyvideo_path
Keras-yolo3-master