heldenkombinat / logodetect Goto Github PK

Hi! Please help!
I have compiled a version of logodetect. Executed the logodetect init command . There are files in the models folder, but for some reason I get the following error:

logodetect image -i /home/kali/cybersquat/screenshots/http-31.31.205.163-80.png Traceback (most recent call last): File "/usr/local/bin/logodetect", line 33, in <module> sys.exit(load_entry_point('logodetect', 'console_scripts', 'logodetect')()) File "/home/kali/.local/lib/python3.7/site-packages/click/core.py", line 1128, in __call__ return self.main(*args, **kwargs) File "/home/kali/.local/lib/python3.7/site-packages/click/core.py", line 1053, in main rv = self.invoke(ctx) File "/home/kali/.local/lib/python3.7/site-packages/click/core.py", line 1659, in invoke return _process_result(sub_ctx.command.invoke(sub_ctx)) File "/home/kali/.local/lib/python3.7/site-packages/click/core.py", line 1395, in invoke return ctx.invoke(self.callback, **ctx.params) File "/home/kali/.local/lib/python3.7/site-packages/click/core.py", line 754, in invoke return __callback(*args, **kwargs) File "/home/kali/logodetect/Logodetect/logodetect/cli.py", line 64, in image recognizer = Recognizer(exemplars, config) File "/home/kali/logodetect/Logodetect/logodetect/recognizer.py", line 41, in __init__ ).Detector(config=self.config) File "/home/kali/logodetect/Logodetect/logodetect/detectors/faster_rcnn.py", line 25, in __init__ self.config.get("DETECTOR_DEVICE"), self.config.get("DETECTOR_WEIGHTS") File "/home/kali/logodetect/Logodetect/logodetect/detectors/__init__.py", line 47, in _binary_fasterrcnn_resnet50 checkpoint = torch.load(model_weights, map_location=device) File "/home/kali/.local/lib/python3.7/site-packages/torch/serialization.py", line 699, in load with _open_file_like(f, 'rb') as opened_file: File "/home/kali/.local/lib/python3.7/site-packages/torch/serialization.py", line 231, in _open_file_like return _open_file(name_or_buffer, mode) File "/home/kali/.local/lib/python3.7/site-packages/torch/serialization.py", line 212, in __init__ super(_open_file, self).__init__(open(name, mode)) FileNotFoundError: [Errno 2] No such file or directory: '/home/kali/.hkt/logodetect/models/detector.pth'

Whats wrong? Thanks!

Use click to expose a CLI that works like this:

logo-detect image foo.img
logo-detect video bar.mp4

optional output path/name and whether to immediately show the result would be desirable

• High-level description of what the project does and why I should care
• link to demo (other issue)
• More in-depth usage
• How to run examples
• Logo/brand for the project
• CI batches, pypi version etc.
• License
• Authors

Something quick and dirty, but relatively sexy, so that we can pitch with it. Can be hosted on GitHub pages, or netlify if the setup becomes more complicated.

Traceback (most recent call last):
File "C:\Users\AjayRawat\anaconda3\envs\logodetect\lib\runpy.py", line 197, in _run_module_as_main
return run_code(code, main_globals, None,
File "C:\Users\AjayRawat\anaconda3\envs\logodetect\lib\runpy.py", line 87, in run_code
exec(code, run_globals)
File "C:\Users\AjayRawat\anaconda3\envs\logodetect\Scripts\logodetect.exe_main.py", line 4, in
File "C:\Users\AjayRawat\anaconda3\envs\logodetect\lib\site-packages\logodetect_init.py", line 8, in
os.path.join(os.environ["HOME"], ".hkt", "logodetect")
File "C:\Users\AjayRawat\anaconda3\envs\logodetect\lib\os.py", line 679, in getitem
raise KeyError(key) from None
KeyError: 'HOME'

Apparently, some output images are mislabeled when using the GUI. For example:

.

I realize Logo-Net is not publicly available. However, how is it structured?

Related to #22, maybe in the end one and the same. Maybe this one will be embedded in the landing page. Idea is to build an e.g. Flask-based web application serving our model that can be deployed easily. Maybe have a Google Cloud Run button or similar that we can put on the readme.

Dear authors,

I installed logodetect using pip, had a first error state that the constants package was missing, installed constants, then have the following error message:

> logodetect
Traceback (most recent call last):
  File "/usr/local/bin/logodetect", line 5, in <module>
    from logodetect.cli import cli
  File "/usr/local/lib/python3.8/dist-packages/logodetect/cli.py", line 50, in <module>
    default=constants.IMAGE_FILENAME,
AttributeError: module 'constants' has no attribute 'IMAGE_FILENAME'

I also tried with an image as first argument and got the same error.

Nicolas

I am on Windows.

When pip installing logodetect or logodetect==1.1.2, I get
ERROR: Could not find a version that satisfies the requirement torch>=1.9.0 (from logodetect) (from versions: none)
ERROR: No matching distribution found for torch>=1.9.0

Tried on WSL as well, got similar errors.

when installing pip install logodetect we should create a new folder at ~/.hkt/logodetect in which data is stored. the necessary models for the detection demo should then lazily be downloaded from an appropriate storage solution (or could be triggered explicitly with logodetect download or so)

working with constants.py works fine as long a people clone the repo and work from source. we want to aim at installing via pip, in which case a config file should be provided external to the installed library.

Hi thanks for such an impressive model!
I am interested on running the provided notebooks. It looks like the LOGO-Net dataset is not publicly available. Is there a sample of this dataset I could use to run these notebooks? Or an explanation of how this datasets is structured so I can make a similar small dataset and run the notebooks on it.
Thanks!