creating underwater sonar imaging for fisheries dataset for detection

in this repo you will find 10 training batch files containing average of 215 images. The first training batch file contains the annotation file which is a csv annotated in YOLO format. to use the training images extract them all in the same file naming it "Train". it is also provided test and validation files. the test file contains 200 images and the validation contains 100

• prepaining dataset good reference: http://christopher5106.github.io/computer/vision/2015/12/26/file-format-for-computer-vision-annotations.html