Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Unable to run eval.py successfully #17

Closed
w804479595 opened this issue Apr 27, 2018 · 14 comments
Closed

Unable to run eval.py successfully #17

w804479595 opened this issue Apr 27, 2018 · 14 comments

Comments

@w804479595
Copy link

Hi，I followed your steps in tutorial 3-detection ,but I got some error when I try to run eval.py.
Here is the wrong log:
mask_scale: Using default '1.000000'
Loading weights from products/backup/yolo-chinese_final.weights...conv 5030 1 x 1 / 1 38 x 38 x1024 -> 38 x 38 x5030
30 detection
mask_scale: Using default '1.000000'
Loading weights from products/backup/yolo-chinese_final.weights...Done!
Learning Rate: 0.0001, Momentum: 0.9, Decay: 0.0005
4
Done!
Learning Rate: 0.0001, Momentum: 0.9, Decay: 0.0005
4
8
Exception in thread Thread-2:
Traceback (most recent call last):
File "/home/deeplearn/anaconda3/lib/python3.6/threading.py", line 916, in _bootstrap_inner
self.run()
File "/home/deeplearn/anaconda3/lib/python3.6/threading.py", line 864, in run
self._target(*self._args, **self._kwargs)
File "/home/gxwang2/ctw/ctw-baseline/detection/pythonapi/common_tools.py", line 77, in parallel_work
func(*args_list[i], tid=tid)
File "eval.py", line 70, in eval_yolo
assert 0 == p.returncode
AssertionError

Exception in thread Thread-1:
Traceback (most recent call last):
File "/home/deeplearn/anaconda3/lib/python3.6/threading.py", line 916, in _bootstrap_inner
self.run()
File "/home/deeplearn/anaconda3/lib/python3.6/threading.py", line 864, in run
self._target(*self._args, **self._kwargs)
File "/home/gxwang2/ctw/ctw-baseline/detection/pythonapi/common_tools.py", line 77, in parallel_work
func(*args_list[i], tid=tid)
File "eval.py", line 70, in eval_yolo
assert 0 == p.returncode
AssertionError

I changed the num_thread and TEST_NUM_GPU but that doesn't work.
Could you give me some help? Thank you
@yuantailing
Copy link
Owner

Test this command:

darknet/darknet detector valid products/chinese.0.data products/yolo-chinese-test.cfg products/backup/yolo-chinese_final.weights -out chinese.0

and check whether return code is 0.

@w804479595
Copy link
Author

I try your command and return code is 0.
Then I check the settings.py and found out the reason: 'DARKNET_RESULTS_OUT' has been changed into ‘products/results/chinese.txt’
I using 'DARKNET_RESULTS_OUT = chinese ' to run eval.py again and the return code is 0 now.
Thanks a lot.

@yuantailing
Copy link
Owner

yuantailing commented May 11, 2018 •
edited
Loading

Note: 'settings.DARKNET_RESULTS_OUT' parameter is directly passed to 'outfile' in https://github.com/pjreddie/darknet/blob/f6d861736038da22c9eb0739dca84003c5a5e275/examples/detector.c#L409 Line 409, that's why we cannot write a file path. And so does 'settings.DARKNET_RESULTS_DIR', it is passed to 'prefix'.

@peppapeppapeppapeppa
Copy link

您好，我在python3 eval.py时候出现了问题，请问您知道怎么解决吗？
`make -j16
make: Nothing to be done for 'all'.
CUDA_VISIBLE_DEVICES=0 darknet/darknet detector valid products/chinese.1.data products/yolo-chinese-test.cfg products/backup/yolo-chinese_final.weights -out chinese.1
CUDA_VISIBLE_DEVICES=0 darknet/darknet detector valid products/chinese.0.data products/yolo-chinese-test.cfg products/backup/yolo-chinese_final.weights -out chinese.0
layer filters size input output
layer filters size input output
0 0 conv 32 3 x 3 / 1 1216 x1216 x 3 -> 1216 x1216 x 32
1 max 2 x 2 / 2 1216 x1216 x 32 -> 608 x 608 x 32
2 conv 32 3 x 3 / 1 1216 x1216 x 3 -> 1216 x1216 x 32
1 max 2 x 2 / 2 1216 x1216 x 32 -> 608 x 608 x 32
2 conv 64 3 x 3 / 1 608 x 608 x 32 -> 608 x 608 x 64
3 max 2 x 2 / 2 608 x 608 x 64 -> 304 x 304 x 64
4 conv 64 3 x 3 / 1 608 x 608 x 32 -> 608 x 608 x 64
3 conv 128 3 x 3 / 1 304 x 304 x 64 -> 304 x 304 x 128
5 max 2 x 2 / 2 608 x 608 x 64 -> 304 x 304 x 64
4 conv 64 1 x 1 / 1 304 x 304 x 128 -> 304 x 304 x 64
6 conv 128 3 x 3 / 1 304 x 304 x 64 -> 304 x 304 x 128
5 conv 128 3 x 3 / 1 304 x 304 x 64 -> 304 x 304 x 128
7 conv 64 1 x 1 / 1 304 x 304 x 128 -> 304 x 304 x 64
6 max 2 x 2 / 2 304 x 304 x 128 -> 152 x 152 x 128
8 conv 128 3 x 3 / 1 304 x 304 x 64 -> 304 x 304 x 128
7 conv 256 3 x 3 / 1 152 x 152 x 128 -> 152 x 152 x 256
9 max 2 x 2 / 2 304 x 304 x 128 -> 152 x 152 x 128
8 conv 128 1 x 1 / 1 152 x 152 x 256 -> 152 x 152 x 128
10 conv 256 3 x 3 / 1 152 x 152 x 128 -> 152 x 152 x 256
9 conv 256 3 x 3 / 1 152 x 152 x 128 -> 152 x 152 x 256
11 max 2 x 2 / 2 152 x 152 x 256 -> 76 x 76 x 256
12 conv 128 1 x 1 / 1 152 x 152 x 256 -> 152 x 152 x 128
10 conv 256 3 x 3 / 1 152 x 152 x 128 -> 152 x 152 x 256
11 max 2 x 2 / 2 152 x 152 x 256 -> 76 x 76 x 256
12 conv 512 3 x 3 / 1 76 x 76 x 256 -> 76 x 76 x 512
13 conv 256 1 x 1 / 1 76 x 76 x 512 -> 76 x 76 x 256
14 conv 512 3 x 3 / 1 76 x 76 x 256 -> 76 x 76 x 512
13 conv 256 1 x 1 / 1 76 x 76 x 512 -> 76 x 76 x 256
14 conv 512 3 x 3 / 1 76 x 76 x 256 -> 76 x 76 x 512
15 conv 256 1 x 1 / 1 76 x 76 x 512 -> 76 x 76 x 256
16 conv 512 3 x 3 / 1 76 x 76 x 256 -> 76 x 76 x 512
15 conv 256 1 x 1 / 1 76 x 76 x 512 -> 76 x 76 x 256
16 conv 512 3 x 3 / 1 76 x 76 x 256 -> 76 x 76 x 512
17 max 2 x 2 / 2 76 x 76 x 512 -> 38 x 38 x 512
18 conv 512 3 x 3 / 1 76 x 76 x 256 -> 76 x 76 x 512
17 max 2 x 2 / 2 76 x 76 x 512 -> 38 x 38 x 512
18 conv 1024 3 x 3 / 1 38 x 38 x 512 -> 38 x 38 x1024
19 conv 512 1 x 1 / 1 38 x 38 x1024 -> 38 x 38 x 512
20 conv 1024 3 x 3 / 1 38 x 38 x 512 -> 38 x 38 x1024
19 conv 512 1 x 1 / 1 38 x 38 x1024 -> 38 x 38 x 512
20 conv 1024 3 x 3 / 1 38 x 38 x 512 -> 38 x 38 x1024
21 conv 512 1 x 1 / 1 38 x 38 x1024 -> 38 x 38 x 512
22 conv 1024 3 x 3 / 1 38 x 38 x 512 -> 38 x 38 x1024
21 conv 512 1 x 1 / 1 38 x 38 x1024 -> 38 x 38 x 512
22 conv 1024 3 x 3 / 1 38 x 38 x 512 -> 38 x 38 x1024
23 conv 1024 3 x 3 / 1 38 x 38 x 512 -> 38 x 38 x1024
23 conv 1024 3 x 3 / 1 38 x 38 x1024 -> 38 x 38 x1024
24 conv 1024 3 x 3 / 1 38 x 38 x1024 -> 38 x 38 x1024
24 conv 1024 3 x 3 / 1 38 x 38 x1024 -> 38 x 38 x1024
25 route 16
26 reorg / 2 76 x 76 x 512 -> 38 x 38 x2048
27 route 26 24
28 conv 1024 3 x 3 / 1 38 x 38 x1024 -> 38 x 38 x1024
25 route 16
26 reorg / 2 76 x 76 x 512 -> 38 x 38 x2048
27 route 26 24
28 conv 1024 3 x 3 / 1 38 x 38 x3072 -> 38 x 38 x1024
29 conv 1024 3 x 3 / 1 38 x 38 x3072 -> 38 x 38 x1024
29 conv 5030 1 x 1 / 1 38 x 38 x1024 -> 38 x 38 x5030
30 detection
mask_scale: Using default '1.000000'
Loading weights from products/backup/yolo-chinese_final.weights...conv 5030 1 x 1 / 1 38 x 38 x1024 -> 38 x 38 x5030
30 detection
mask_scale: Using default '1.000000'
Loading weights from products/backup/yolo-chinese_final.weights...Done!
Learning Rate: 0.0001, Momentum: 0.9, Decay: 0.0005
Done!
Learning Rate: 0.0001, Momentum: 0.9, Decay: 0.0005
Exception in thread Thread-2:
Traceback (most recent call last):
File "/home/smart/anaconda3/lib/python3.7/threading.py", line 917, in _bootstrap_inner
self.run()
File "/home/smart/anaconda3/lib/python3.7/threading.py", line 865, in run
self._target(*self._args, **self._kwargs)
File "/home/smart/zhuxuan/CTW/ctw-baseline-master/detection/pythonapi/common_tools.py", line 77, in parallel_work
func(*args_list[i], tid=tid)
File "eval.py", line 65, in eval_yolo
assert 0 == p.returncode
AssertionError

Exception in thread Thread-1:
Traceback (most recent call last):
File "/home/smart/anaconda3/lib/python3.7/threading.py", line 917, in _bootstrap_inner
self.run()
File "/home/smart/anaconda3/lib/python3.7/threading.py", line 865, in run
self._target(*self._args, **self._kwargs)
File "/home/smart/zhuxuan/CTW/ctw-baseline-master/detection/pythonapi/common_tools.py", line 77, in parallel_work
func(*args_list[i], tid=tid)
File "eval.py", line 65, in eval_yolo
assert 0 == p.returncode
AssertionError
`

@peppapeppapeppapeppa
Copy link

Hi，I followed your steps in tutorial 3-detection ,but I got some error when I try to run eval.py.
Here is the wrong log:
mask_scale: Using default '1.000000'
Loading weights from products/backup/yolo-chinese_final.weights...conv 5030 1 x 1 / 1 38 x 38 x1024 -> 38 x 38 x5030
30 detection
mask_scale: Using default '1.000000'
Loading weights from products/backup/yolo-chinese_final.weights...Done!
Learning Rate: 0.0001, Momentum: 0.9, Decay: 0.0005
4
Done!
Learning Rate: 0.0001, Momentum: 0.9, Decay: 0.0005
4
8
Exception in thread Thread-2:
Traceback (most recent call last):
File "/home/deeplearn/anaconda3/lib/python3.6/threading.py", line 916, in _bootstrap_inner
self.run()
File "/home/deeplearn/anaconda3/lib/python3.6/threading.py", line 864, in run
self._target(*self._args, **self._kwargs)
File "/home/gxwang2/ctw/ctw-baseline/detection/pythonapi/common_tools.py", line 77, in parallel_work
func(*args_list[i], tid=tid)
File "eval.py", line 70, in eval_yolo
assert 0 == p.returncode
AssertionError

Exception in thread Thread-1:
Traceback (most recent call last):
File "/home/deeplearn/anaconda3/lib/python3.6/threading.py", line 916, in _bootstrap_inner
self.run()
File "/home/deeplearn/anaconda3/lib/python3.6/threading.py", line 864, in run
self._target(*self._args, **self._kwargs)
File "/home/gxwang2/ctw/ctw-baseline/detection/pythonapi/common_tools.py", line 77, in parallel_work
func(*args_list[i], tid=tid)
File "eval.py", line 70, in eval_yolo
assert 0 == p.returncode
AssertionError

I changed the num_thread and TEST_NUM_GPU but that doesn't work.
Could you give me some help? Thank you

请问您这个问题解决了吗，是如何解决的呢？期待您的回答

@peppapeppapeppapeppa
Copy link

测试该命令时darknet/darknet detector valid products/chinese.0.data products/yolo-chinese-test.cfg products/backup/yolo-chinese_final.weights -out chinese.0
显示mask_scale: Using default '1.000000' Loading weights from products/backup/yolo-chinese_final.weights...Done! Learning Rate: 0.0001, Momentum: 0.9, Decay: 0.0005 段错误 期待您的回答，谢谢

@yuantailing
Copy link
Owner

先看一下使用的 darknet 版本是不是我改过的版本：
https://github.com/yuantailing/darknet/tree/801409731a46b072b8ee11f6f2acfa29cb16f165

这是 Git Submodule 引用的指定版本，并不是任何版本的 darknet 都行。

@peppapeppapeppapeppa
Copy link

先看一下使用的 darknet 版本是不是我改过的版本：
https://github.com/yuantailing/darknet/tree/801409731a46b072b8ee11f6f2acfa29cb16f165

这是 Git Submodule 引用的指定版本，并不是任何版本的 darknet 都行。

您好 我检查一下确实是您的改过的版本，但是还是出现了该错误，不知道问题出在了哪里

@peppapeppapeppapeppa
Copy link

而且只是在测试时执行python3 eval.py出现的问题，训练的时候没有报错。期待您的回答。

@yuantailing
Copy link
Owner

yuantailing commented Sep 5, 2019 •
edited
Loading

命令行里单进程调用 darknet/darknet 也出问题的话，请确认空闲显存大于 4GB。既然训练时都没报错，应该是够的。我没遇到过这个问题，这得用gdb或加断点查程序崩在哪了。

@yuantailing
Copy link
Owner

yuantailing commented Sep 5, 2019 •
edited
Loading

运行 darknet/darknet detector valid products/chinese.0.data products/yolo-chinese-test.cfg products/backup/yolo-chinese_final.weights -out chinese.0 前，

确认文件夹 products/results 存在

确认 products/test.0.txt 里引用的图片存在

确认以下文件存在，核对文件内容

products/chinese.0.data
classes = 1001
eval = chinese
names = products/chinese.names
results = products/results
valid = products/test.0.txt

products/chinese.names
0
1
2
...

products/yolo-chinese-test.cfg
[net]

batch=1
subdivisions=1

height=1216
width=1216
channels=3
...

@peppapeppapeppapeppa
Copy link

运行 darknet/darknet detector valid products/chinese.0.data products/yolo-chinese-test.cfg products/backup/yolo-chinese_final.weights -out chinese.0 前，

确认文件夹 products/results 存在

确认 products/test.0.txt 里引用的图片存在

确认以下文件存在，核对文件内容

products/chinese.0.data
classes = 1001
eval = chinese
names = products/chinese.names
results = products/results
valid = products/test.0.txt

products/chinese.names
0
1
2
...

products/yolo-chinese-test.cfg
[net]

batch=1
subdivisions=1

height=1216
width=1216
channels=3
...

感谢您的耐心解答。我按照您说的仔细核查了一遍，发现文件夹 products/test.0.txt中显示0字节，没有任何内容。其他的文件夹以及文件并没有什么问题。并且data/images/test文件夹中并没有生成测试集图片,之前的指令只生成了data/images/trainval文件夹中的训练集图片。不知是哪里出现了问题。

@yuantailing
Copy link
Owner

prepare_test_data.py 再运行一遍

@peppapeppapeppapeppa
Copy link

prepare_test_data.py 再运行一遍

运行 prepare_test_data.py之后没反应，我觉得问题出在cd ../prepare && python3 symlink_images.py这里，因为运行此命令之后只有文件夹 data/images/trainval中生成了图片，而文件夹data/images/test中并没有图片。我也没有更改过代码，试了很多次还是不行。

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
3 participants
@yuantailing @w804479595 @peppapeppapeppapeppa