Using TensorFlow backend.
Parsing annotation files
Training instances per class:
{'1': 396, 'bg': 0}
Num classes (including bg) = 2
Config has been written to fcnet_config.pickle, and can be loaded when testing to ensure correct results
Num train samples 264
Num val samples 0
Traceback (most recent call last):
File "D:\Python\fast R cnn\Traffic-Sign-Detection-Faster-RCNN-master\train_frcnn.py", line 127, in
data_gen_train = data_generators.get_anchor_gt(train_imgs, classes_count, C, nn.img_length_calc_function, K.image_dim_ordering(), mode='train')
AttributeError: module 'tensorflow.python.keras.api._v2.keras.backend' has no attribute 'image_dim_ordering'
Using TensorFlow backend. Parsing annotation files Training instances per class: {'1': 396, 'bg': 0} Num classes (including bg) = 2 Config has been written to fcnet_config.pickle, and can be loaded when testing to ensure correct results Num train samples 264 Num val samples 0 Traceback (most recent call last): File "D:\Python\fast R cnn\Traffic-Sign-Detection-Faster-RCNN-master\train_frcnn.py", line 127, in
data_gen_train = data_generators.get_anchor_gt(train_imgs, classes_count, C, nn.img_length_calc_function, K.image_dim_ordering(), mode='train')
AttributeError: module 'tensorflow.python.keras.api._v2.keras.backend' has no attribute 'image_dim_ordering'