Receiving Illegal instruction (core dumped)

[MajorLOL]

trafficstars

Greetings, here is my setup: can anyone tell me what i am doing wrong?

also dont see the env file, i can open it tho.

This is the error i am getting

Thank you.

[cjekel]

Fixed the problem myself,

Seems like models needed to be in same folder and also folder needed to be directed to docker/tinder/tinder:/tindetheus

Problem now is receiving:

forward to solutions :) ill keep you posted who know maybe someone els will need it in the furture

[MajorLOL]

did i not delete that? :) seemed not to be the answer

[cjekel]

Hmm seems the ubuntu build has changed a bit. This was my first try in docker:

root@ac1f190e018a:/tinder# tindetheus browse
Traceback (most recent call last):
  File "/usr/local/bin/tindetheus", line 33, in <module>
    sys.exit(load_entry_point('tindetheus', 'console_scripts', 'tindetheus')())
  File "/usr/local/bin/tindetheus", line 25, in importlib_load_entry_point
    return next(matches).load()
  File "/usr/local/lib/python3.6/dist-packages/importlib_metadata/__init__.py", line 93, in load
    module = import_module(match.group('module'))
  File "/usr/lib/python3.6/importlib/__init__.py", line 126, in import_module
    return _bootstrap._gcd_import(name[level:], package, level)
  File "<frozen importlib._bootstrap>", line 994, in _gcd_import
  File "<frozen importlib._bootstrap>", line 971, in _find_and_load
  File "<frozen importlib._bootstrap>", line 955, in _find_and_load_unlocked
  File "<frozen importlib._bootstrap>", line 665, in _load_unlocked
  File "<frozen importlib._bootstrap_external>", line 678, in exec_module
  File "<frozen importlib._bootstrap>", line 219, in _call_with_frames_removed
  File "/tindetheus/tindetheus/__init__.py", line 1, in <module>
    from .tindetheus import *  # noqa F401
  File "/tindetheus/tindetheus/tindetheus.py", line 46, in <module>
    from tindetheus import export_embeddings
  File "/tindetheus/tindetheus/export_embeddings.py", line 88, in <module>
    from tindetheus.facenet_clone.align import detect_face
  File "/tindetheus/tindetheus/facenet_clone/align/detect_face.py", line 34, in <module>
    import cv2
  File "/usr/local/lib/python3.6/dist-packages/cv2/__init__.py", line 5, in <module>
    from .cv2 import *
ImportError: libGL.so.1: cannot open shared object file: No such file or directory

which was fixed with

apt install libgl1-mesa-glx

And now I get the authentication issue here:

root@ac1f190e018a:/tinder# tindetheus browse
Traceback (most recent call last):
  File "/usr/local/bin/tindetheus", line 33, in <module>
    sys.exit(load_entry_point('tindetheus', 'console_scripts', 'tindetheus')())
  File "/tindetheus/tindetheus/tindetheus.py", line 279, in command_line_run
    retries=defaults['retries'])
  File "/tindetheus/tindetheus/tindetheus.py", line 80, in main
    likes_left=args.likes, x_auth_token=x_auth_token)
  File "/tindetheus/tindetheus/tinder_client.py", line 89, in __init__
    self.session = self.login(facebook_token, x_auth_token)
  File "/tindetheus/tindetheus/tinder_client.py", line 127, in login
    XAuthToken=x_auth_token)
  File "/usr/local/lib/python3.6/dist-packages/pynder/session.py", line 18, in __init__
    self._api.auth(facebook_id, facebook_token)
  File "/usr/local/lib/python3.6/dist-packages/pynder/api.py", line 41, in auth
    raise errors.RequestError(err)
pynder.errors.RequestError: Couldn't authenticate request status code: 401

Which I think is further than you were able to get, and thindetheus train worked successfully. (I don't currently have a method to test authentication).

I'm going to push a change to the docker build to include libgl. That docker build is two months old. Then we will see what happens.

I didn't run into your problem, but it's not impossible with the magnitude of dependencies. I suspect it can be fixed with python3 -m pip install importlib-metadata

[cjekel]

@MajorLOL You are still getting core dumps?

[MajorLOL]

Still the same problem.

@MajorLOL You are still getting core dumps?

[cjekel]

does tindetheus --help give core dump too

[MajorLOL]

does tindetheus --help give core dump too

Yes

[cjekel]

Just a sanity check while I wait for the new docker build.

Just make sure plane python works.

python3 -c 'print("hello world")'

[MajorLOL]

Just a sanity check while I wait for the new docker build.

Just make sure plane python works.

python3 -c 'print("hello world")'

Gotta to home from work now ill check this post when i am home 30 min. let me know if you need me to do other thing to test.

[MajorLOL]

Just a sanity check while I wait for the new docker build.

Just make sure plane python works.

python3 -c 'print("hello world")'

Greetings, i saw you did an update. Currently i am receiving the same error. Can you contact me on Discord maybe? so i can screen share? id: 206496897022558209 Anónymous #7301

[MajorLOL]

Even when running fresh install and not putting in the folders give the same result

[cjekel]

Sorry I got caught up in my work.

I tested the latest automatic build from docker and it was working fine. I take it you have been building this yourself?

https://hub.docker.com/r/cjekel/tindetheus

So I just run docker pull cjekel/tindetheus instead of docker build. I think the image is like 10 Gb of data, so watch your bandwidth.

Are you using any unusual hardware (Arm? apple m1?)

[MajorLOL]

Greetings,

Running it on my Synology, i tried everything but all with the same results. It says latest: Pulling from cjekel/tindetheus Digest: sha256:b6dc4ca572432a3628f1b050e760171b9273a5222a62d0e88a5d92ae257a5ba3 Status: Image is up to date for cjekel/tindetheus:latest

[cjekel]

Sorry for the late reply.

I suspect something docker on that machine isn't working correctly on that hardware. Most likely it's something with the old tensorflow.

Is it possible to do the native-setup on your machine? (I know that docker is suppose to avoid these issues, but I can't reproduce your docker problems on my machines :( )

[MajorLOL]

Sorry for the late reaction, Used image: https://registry.hub.docker.com/_/python/

followed commands in link and received: ERROR: Could not find a version that satisfies the requirement tensorflow<2.0.0 ERROR: No matching distribution found for tensorflow<2.0.0 using docker 18.09.0-0519 (latest) on synology

used command: pip install tensorflow

Receiving same message before doing install tensorflow or after both on new images. Will search more info later.

[cjekel]

@MajorLOL I think it's too new of a python version, that tensorflow 1.15 was never release for? Tensorflow 1 only goes up to python 3.7 support https://pypi.org/project/tensorflow/1.15.5/

So can you check which version of python that is using?

[MajorLOL]

@cjekel can you please point me in the right direction? I tryed many differend once without succes: https://registry.hub.docker.com/_/python/?tab=tags&page=1&ordering=last_updated