heartkilla / yolo-v3 Goto Github PK

After follow step by step the installation instructions I try to run :

python detect.py images 0.5 0.5 data/images/office.jpg

And I get this error :

File "detect.py", line 39 inputs = tf.placeholder(tf.float32, [batch_size, *_MODEL_SIZE, 3]) SyntaxError: invalid syntax

( ^ show "*" on *_MODEL_SIZE)

How can I fix that ?

in file "detect.py", line 100, in
main(sys.argv[1], float(sys.argv[2]), float(sys.argv[3]), sys.argv[4:])
IndexError: list index out of range

while i was trying to run this code, i'm getting following error

SystemError: <class 'cv2.VideoWriter'> returned a result with an error set
warning: Error opening file (/build/opencv/modules/videoio/src/cap_ffmpeg_impl.hpp:834)
warning: C (/build/opencv/modules/videoio/src/cap_ffmpeg_impl.hpp:835)

i've installed ffmpeg library , also added to path. still i'm unable to resolve

from yolo_v3 import Yolo_v3
ModuleNotFoundError: No module named 'yolo_v3'

on running python load_weights.py

OpenCV: FFMPEG: tag 0x34363258/'X264' is not supported with codec id 27 and format 'mp4 / MP4 (MPEG-4 Part 14)'
OpenCV: FFMPEG: fallback to use tag 0x31637661/'avc1'

        OpenH264 Video Codec provided by Cisco Systems, Inc.

This is the log I have on cmd.
Video got stuck here ... Its not detecting anything. Why its taking so long?

How to train model on custom dataset?

When I compile load_weights.py, I get the below warning and the same happens when I compile detect.py as well. Ultimately detections are not placed in detections folder.

ERROR DETAILS:

UserWarning: h5py is running against HDF5 1.10.5 when it was built against 1.10.4, this may cause problems
'{0}.{1}.{2}'.format(*version.hdf5_built_version_tuple)
Warning! HDF5 library version mismatched error
The HDF5 header files used to compile this application do not match
the version used by the HDF5 library to which this application is linked.
Data corruption or segmentation faults may occur if the application continues.
This can happen when an application was compiled by one version of HDF5 but
linked with a different version of static or shared HDF5 library.
You should recompile the application or check your shared library related
settings such as 'LD_LIBRARY_PATH'.
You can, at your own risk, disable this warning by setting the environment
variable 'HDF5_DISABLE_VERSION_CHECK' to a value of '1'.
Setting it to 2 or higher will suppress the warning messages totally.
Headers are 1.10.4, library is 1.10.5

SUMMARY OF THE HDF5 CONFIGURATION
=================================

General Information:

               HDF5 Version: 1.10.5
              Configured on: 2019-03-04
              Configured by: Visual Studio 14 2015 Win64
                Host system: Windows-10.0.17763
          Uname information: Windows
                   Byte sex: little-endian
         Installation point: C:/Program Files/HDF5

python detect.py video 0.5 0.5 data/video/shinjuku.mp4

when i run this command, i got this error.

2019-04-12 16:46:20.595148: E tensorflow/core/common_runtime/bfc_allocator.cc:373] tried to deallocate nullptr

I successfully compiled detect.py, but it will not pop up the picture display detection box
but i see it have draw_box function

Hi)) Maybe you know how can I solve this problem:
ValueError: cannot reshape array of size 15472 into shape (128,64,3,3)

Hello, how important are photo resolutions when preparing my own dataset? So should the size of my photos for yolov3 be 416 x 416? (608 x 608 for Yolov4?) How important is this? What are the important parameters when creating the dataset?

My second question is what kind of way can I follow to recognize only certain objects. For example, I just want to know the "black mercedes cla" class. İt shouldn't know when he sees another car. He only needs to know the trademarks, models and colors I have defined. Like face recognition systems.

Thank you for all your feedbacks.

stuck at running load_weights.py

traceback (most recent call last):
  File "load_weights.py", line 117, in <module>
    main()
  File "load_weights.py", line 101, in main
    inputs = tf.placeholder(tf.float32, [1, 416, 416, 3])
AttributeError: module 'tensorflow' has no attribute 'placeholder'

i use tf ver 2.0.0
thx