Custom data training error about yolo_tensorflow HOT 4 OPEN

Without looking at the code, my guess it that you haven't changed the number of classes as the main difference between the two shapes is 25 and 7.

from yolo_tensorflow.

@ashleyjsands Can you tel me what i need to change?
python train.py
WARNING:tensorflow:From /home/g5cdevs_gcp/yolo_tensorflow/yolo/yolo_net.py:186: calling reduce_max (from tensorflow.python.ops.math_ops) with keep_dims is deprecated and will be removed in a future version.
Instructions for updating:
keep_dims is deprecated, use keepdims instead
Processing gt_labels from: data/pascal_voc/VOCdevkit/VOC2007
Traceback (most recent call last):
File "train.py", line 164, in
main()
File "train.py", line 152, in main
pascal = pascal_voc('train')
File "/home/g5cdevs_gcp/yolo_tensorflow/utils/pascal_voc.py", line 26, in init
self.prepare()
File "/home/g5cdevs_gcp/yolo_tensorflow/utils/pascal_voc.py", line 57, in prepare
gt_labels = self.load_labels()
File "/home/g5cdevs_gcp/yolo_tensorflow/utils/pascal_voc.py", line 102, in load_labels
label, num = self.load_pascal_annotation(index)
File "/home/g5cdevs_gcp/yolo_tensorflow/utils/pascal_voc.py", line 138, in load_pascal_annotation
cls_ind = self.class_to_ind[obj.find('name').text.lower().strip()]
KeyError: 'linear'

from yolo_tensorflow.

@Mahibro I have no idea why you are getting that error. Having a guess: maybe an activation function of 'linear' is missing or not supported?

from yolo_tensorflow.

@ashleyjsands @hizhangp What all i need to make change(In files),when i create XML files for my own dataset.

from yolo_tensorflow.