因此，当使用Darkflow和我自己的100个图像数据集进行训练时，我遇到了一个挑战：我的系统似乎没有任何东西在流动：
Vinces-MBP:darkflow vinceparis$ python3.6 flow --model cfg/yolo-new.cfg --labels one_label.txt --train --trainer adam --dataset
"/Users/vinceparis/xfz/100pics/Images/002" --annotation
"/Users/vinceparis/xfz/BBox-Label-Tool/AnnotationsXML/002"
Parsing cfg/yolo-new.cfg
Loading None ...
Finished in 0.00015306472778320312s
Building net ...
Source | Train? | Layer description | Output size
-------+--------+----------------------------------+---------------
| | input | (?, 608, 608, 3)
Init | Yep! | conv 3x3p1_1 +bnorm leaky | (?, 608, 608, 32)
Load | Yep! | maxp 2x2p0_2 | (?, 304, 304, 32)
Init | Yep! | conv 3x3p1_1 +bnorm leaky | (?, 304, 304, 64)
Load | Yep! | maxp 2x2p0_2 | (?, 152, 152, 64)
Init | Yep! | conv 3x3p1_1 +bnorm leaky | (?, 152, 152, 128)
Init | Yep! | conv 1x1p0_1 +bnorm leaky | (?, 152, 152, 64)
Init | Yep! | conv 3x3p1_1 +bnorm leaky | (?, 152, 152, 128)
Load | Yep! | maxp 2x2p0_2 | (?, 76, 76, 128)
Init | Yep! | conv 3x3p1_1 +bnorm leaky | (?, 76, 76, 256)
Init | Yep! | conv 1x1p0_1 +bnorm leaky | (?, 76, 76, 128)
Init | Yep! | conv 3x3p1_1 +bnorm leaky | (?, 76, 76, 256)
Load | Yep! | maxp 2x2p0_2 | (?, 38, 38, 256)
Init | Yep! | conv 3x3p1_1 +bnorm leaky | (?, 38, 38, 512)
Init | Yep! | conv 1x1p0_1 +bnorm leaky | (?, 38, 38, 256)
Init | Yep! | conv 3x3p1_1 +bnorm leaky | (?, 38, 38, 512)
Init | Yep! | conv 1x1p0_1 +bnorm leaky | (?, 38, 38, 256)
Init | Yep! | conv 3x3p1_1 +bnorm leaky | (?, 38, 38, 512)
Load | Yep! | maxp 2x2p0_2 | (?, 19, 19, 512)
Init | Yep! | conv 3x3p1_1 +bnorm leaky | (?, 19, 19, 1024)
Init | Yep! | conv 1x1p0_1 +bnorm leaky | (?, 19, 19, 512)
Init | Yep! | conv 3x3p1_1 +bnorm leaky | (?, 19, 19, 1024)
Init | Yep! | conv 1x1p0_1 +bnorm leaky | (?, 19, 19, 512)
Init | Yep! | conv 3x3p1_1 +bnorm leaky | (?, 19, 19, 1024)
Init | Yep! | conv 3x3p1_1 +bnorm leaky | (?, 19, 19, 1024)
Init | Yep! | conv 3x3p1_1 +bnorm leaky | (?, 19, 19, 1024)
Load | Yep! | concat [16] | (?, 38, 38, 512)
Init | Yep! | conv 1x1p0_1 +bnorm leaky | (?, 38, 38, 64)
Load | Yep! | local flatten 2x2 | (?, 19, 19, 256)
Load | Yep! | concat [27, 24] | (?, 19, 19, 1280)
Init | Yep! | conv 3x3p1_1 +bnorm leaky | (?, 19, 19, 1024)
Init | Yep! | conv 1x1p0_1 linear | (?, 19, 19, 30)
-------+--------+----------------------------------+---------------
Running entirely on CPU
cfg/yolo-new.cfg loss hyper-parameters:
H = 19
W = 19
box = 5
classes = 2
scales = [1.0, 5.0, 1.0, 1.0]
Building cfg/yolo-new.cfg loss
Building cfg/yolo-new.cfg train op
2018-12-09 23:53:49.842284: I
tensorflow/core/platform/cpu_feature_guard.cc:141] Your CPU supports
instructions that this TensorFlow binary was not compiled to use: AVX2 FMA
Finished in 15.812134027481079s
Enter training ...
cfg/yolo-new.cfg parsing /Users/vinceparis/xfz/BBox-Label-
Tool/AnnotationsXML/002
Parsing for ['map_text', 'map_tet2']
[====================>]100% image0.xml
Statistics:
Dataset size: 1
Dataset of 1 instance(s)
image0.jpg__preview copy 88.jpg
Traceback (most recent call last):
File "flow", line 6, in <module>
cliHandler(sys.argv)
File "/Users/vinceparis/xfz/darkflow/darkflow/cli.py", line 33, in
cliHandler
print('Enter training ...'); tfnet.train()
File "/Users/vinceparis/xfz/darkflow/darkflow/net/flow.py", line 39, in train
for i, (x_batch, datum) in enumerate(batches):
File "/Users/vinceparis/xfz/darkflow/darkflow/net/yolo/data.py", line 114, in shuffle
inp, new_feed = self._batch(train_instance)
File "/Users/vinceparis/xfz/darkflow/darkflow/net/yolov2/data.py", line 28, in _batch
img = self.preprocess(path, allobj)
File "/Users/vinceparis/xfz/darkflow/darkflow/net/yolo/predict.py", line 62, in preprocess
result = imcv2_affine_trans(im)
File "/Users/vinceparis/xfz/darkflow/darkflow/utils/im_transform.py", line
20, in imcv2_affine_trans
h, w, c = im.shape
AttributeError: 'NoneType' object has no attribute 'shape'****
我检查了我的BBox-Label-Tool，确认了使用该工具时仅生成了一个xml文件！通过注释过程时，如何获取BBox-Label工具生成多个XML？