Compare revisions

#  /!\  TO DO before : globox convert input_YOLO/labels/val/ output_../classifier/test_convert_labels/ --format yolo --save_fmt labelme --img_folder YOLO/images/val/ #path to images

import argparse

import os

import json

import labelme

import base64

import pandas as pd

mode = 'train' #or val

img_dir = str('../annot/YOLO/images/'+mode+'/')

parser = argparse.ArgumentParser(formatter_class=argparse.ArgumentDefaultsHelpFormatter, description='TODO')

parser.add_argument('-p','--path_to_json', type=str, help = 'Path of the folder that contain the .json',required=True)

parser.add_argument('-i','--path_to_img', type=str, help = 'Path of the folder that contain the .jpg',required=True)

parser.add_argument('-d','--direction', type=str, help = 'Directory to wich modified .json files will be stored',required=True)

args = parser.parse_args()

filename = 'test_convert_labels_train/' #path to .json

out_file = 'test_convert_labels_train_2/' #direction

img_path = '/nfs/NAS7/manip_stephane/annot/YOLO/images/train/'

filename = args.path_to_json

out_file = args.direction

img_path = args.path_to_img

liste_file = os.listdir(filename) 

liste_file = pd.DataFrame(liste_file, columns =['fn'])

for i, row in liste_file.iterrows():

    if len(row.fn) > 30:

        data = labelme.LabelFile.load_image_file(img_dir+row.fn[:-4]+'jpg')

        data = labelme.LabelFile.load_image_file(str(img_path+row.fn[:-4]+'jpg'))

        image_data = base64.b64encode(data).decode('utf-8')

    else:

        continue