Installing imgaug on Jetson Nano breaks CSIC camera #970

Ezward · 2021-12-17T17:46:58Z

This has been found using Jetpack 4.6. In order to use the 'CROP' transformation with an autopilot the imgaug dependency needs to be installed into the environment. It installs without issue. however, once installed the CSIC camera will no longer return frames. This is not an issue in RaspberryPi. I'm not sure if it is an issue in earlier versions of Jetpack.

Using 'CROP' is a best practice when training an autopilot. So it is a problem that one of our supported platforms cannot use it. Worse, the error that is produced provides no help in understanding the underlying issue. We should try to find a fix or work-around for this issue.

Determine if earlier versions of jetpack also suffer from this issue
Determine if there is an specific version of imgaug or one of it's dependencies that can be installed to avoid this issue.
If we can't find a way to make imgaug work then open a issue for implementing CROP and trapezoidal masks that don't use imgaug. There is an alternative library that used to use imgaug but has removed it because of this bug. It may be a good replacement for us https://github.com/albumentations-team/albumentations#list-of-augmentations

It has also been determined that this breaks training on the nano, since the training code always tries to import imgaug. If it cannot import then it fails. If we do the pip install then it gets an runtime illegal instruction error. So a fix for that, if we don't get the imgaug replaced, is to fix the training code to only import imgaug if it is used. That may require some mock stub of the imgaug api we use; so if imgaug fails to import then use a mock that warns the user that augmentations are disabled.

Ezward · 2021-12-17T18:26:16Z

See this related issue #749

Ezward · 2021-12-20T18:48:23Z

when installing imgaug on the nano, I see a series of these errors while it is trying to install

WARNING: Discarding https://files.pythonhosted.org/packages/44/ec/4eddbf9d17a917c51fb4ad159aa7137f506681e91ab559cf87d120e1d78d/Shapely-1.7.0.tar.gz#sha256=e21a9fe1a416463ff11ae037766fe410526c95700b9e545372475d2361cc951e (from https://pypi.org/simple/shapely/). Command errored out with exit status 1: python setup.py egg_info Check the logs for full command output.
  Using cached Shapely-1.6.4.post2.tar.gz (225 kB)
  Preparing metadata (setup.py) ... error
  ERROR: Command errored out with exit status 1:
   command: /home/ed/env/bin/python -c 'import io, os, sys, setuptools, tokenize; sys.argv[0] = '"'"'/tmp/pip-install-dnuepv9f/shapely_3a52b345fc0746389fb9807cad9cd11a/setup.py'"'"'; __file__='"'"'/tmp/pip-install-dnuepv9f/shapely_3a52b345fc0746389fb9807cad9cd11a/setup.py'"'"';f = getattr(tokenize, '"'"'open'"'"', open)(__file__) if os.path.exists(__file__) else io.StringIO('"'"'from setuptools import setup; setup()'"'"');code = f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, __file__, '"'"'exec'"'"'))' egg_info --egg-base /tmp/pip-pip-egg-info-204d_l62
       cwd: /tmp/pip-install-dnuepv9f/shapely_3a52b345fc0746389fb9807cad9cd11a/
  Complete output (11 lines):
  Failed `CDLL(libgeos_c.so.1)`
  Failed `CDLL(libgeos_c.so)`
  Traceback (most recent call last):
    File "<string>", line 1, in <module>
    File "/tmp/pip-install-dnuepv9f/shapely_3a52b345fc0746389fb9807cad9cd11a/setup.py", line 80, in <module>
      from shapely._buildcfg import geos_version_string, geos_version, \
    File "/tmp/pip-install-dnuepv9f/shapely_3a52b345fc0746389fb9807cad9cd11a/shapely/_buildcfg.py", line 167, in <module>
      fallbacks=['libgeos_c.so.1', 'libgeos_c.so'])
    File "/tmp/pip-install-dnuepv9f/shapely_3a52b345fc0746389fb9807cad9cd11a/shapely/_buildcfg.py", line 161, in load_dll
      libname, fallbacks or []))
  OSError: Could not find library geos_c or load any of its variants ['libgeos_c.so.1', 'libgeos_c.so']
  ----------------------------------------

I apperas to be looking for geos_c.so. running this fixes those errors

sudo apt-get install libgeos-dev

However, it does not fix the camera issue, it still won't startup.

Ezward · 2021-12-20T19:23:32Z

This is a simple way to reproduce the but that does not include all the donkeycar code.

$ sudo apt-get install libgeos-dev
$ cd ~/projects
$ git clone https://github.com/JetsonHacksNano/CSI-Camera.git
$ cd ~/projects/CSI-Camera
$ rm -rf env
$ python3 -m virtualenv -p python3 env --system-site-packages
created virtual environment CPython3.6.9.final.0-64 in 1417ms
  creator CPython3Posix(dest=/home/ed/projects/CSI-Camera/env, clear=False, no_vcs_ignore=False, global=True)
  seeder FromAppData(download=False, pip=bundle, setuptools=bundle, wheel=bundle, via=copy, app_data_dir=/home/ed/.local/share/virtualenv)
    added seed packages: pip==21.3.1, setuptools==59.2.0, wheel==0.37.0
  activators BashActivator,CShellActivator,FishActivator,NushellActivator,PowerShellActivator,PythonActivator
$ . env/bin/activate
$ python --version
Python 3.6.9

$ python simple_camera.py 
nvarguscamerasrc ! video/x-raw(memory:NVMM), width=(int)1280, height=(int)720, format=(string)NV12, framerate=(fraction)60/1 ! nvvidconv flip-method=0 ! video/x-raw, width=(int)1280, height=(int)720, format=(string)BGRx ! videoconvert ! video/x-raw, format=(string)BGR ! appsink
GST_ARGUS: Creating output stream
CONSUMER: Waiting until producer is connected...
GST_ARGUS: Available Sensor modes :
GST_ARGUS: 3264 x 2464 FR = 21.000000 fps Duration = 47619048 ; Analog Gain range min 1.000000, max 10.625000; Exposure Range min 13000, max 683709000;

GST_ARGUS: 3264 x 1848 FR = 28.000001 fps Duration = 35714284 ; Analog Gain range min 1.000000, max 10.625000; Exposure Range min 13000, max 683709000;

GST_ARGUS: 1920 x 1080 FR = 29.999999 fps Duration = 33333334 ; Analog Gain range min 1.000000, max 10.625000; Exposure Range min 13000, max 683709000;

GST_ARGUS: 1640 x 1232 FR = 29.999999 fps Duration = 33333334 ; Analog Gain range min 1.000000, max 10.625000; Exposure Range min 13000, max 683709000;

GST_ARGUS: 1280 x 720 FR = 59.999999 fps Duration = 16666667 ; Analog Gain range min 1.000000, max 10.625000; Exposure Range min 13000, max 683709000;

GST_ARGUS: 1280 x 720 FR = 120.000005 fps Duration = 8333333 ; Analog Gain range min 1.000000, max 10.625000; Exposure Range min 13000, max 683709000;

GST_ARGUS: Running with following settings:
   Camera index = 0 
   Camera mode  = 5 
   Output Stream W = 1280 H = 720 
   seconds to Run    = 0 
   Frame Rate = 120.000005 
GST_ARGUS: Setup Complete, Starting captures for 0 seconds
GST_ARGUS: Starting repeat capture requests.
CONSUMER: Producer has connected; continuing.
[ WARN:0] global /home/nvidia/host/build_opencv/nv_opencv/modules/videoio/src/cap_gstreamer.cpp (933) open OpenCV | GStreamer warning: Cannot query video position: status=0, value=-1, duration=-1
^CTraceback (most recent call last):
  File "simple_camera.py", line 66, in <module>
    show_camera()
  File "simple_camera.py", line 55, in show_camera
    keyCode = cv2.waitKey(30) & 0xFF
KeyboardInterrupt
GST_ARGUS: Cleaning up
CONSUMER: Done Success
GST_ARGUS: Done Success

$ pip install imgaug
Collecting imgaug
  Using cached imgaug-0.4.0-py2.py3-none-any.whl (948 kB)
Requirement already satisfied: six in /usr/local/lib/python3.6/dist-packages (from imgaug) (1.15.0)
Collecting opencv-python
  Using cached opencv_python-4.5.4.60-cp36-cp36m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl (39.0 MB)
Requirement already satisfied: scipy in /usr/lib/python3/dist-packages (from imgaug) (0.19.1)
Collecting imageio
  Using cached imageio-2.13.3-py3-none-any.whl (3.3 MB)
Collecting scikit-image>=0.14.2
  Using cached scikit_image-0.17.2-cp36-cp36m-linux_aarch64.whl
Collecting Shapely
  Using cached Shapely-1.8.0-cp36-cp36m-linux_aarch64.whl
Requirement already satisfied: numpy>=1.15 in /usr/local/lib/python3.6/dist-packages (from imgaug) (1.19.4)
Requirement already satisfied: matplotlib in /usr/lib/python3/dist-packages (from imgaug) (2.1.1)
Collecting Pillow
  Using cached Pillow-8.4.0-cp36-cp36m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl (3.0 MB)
Collecting tifffile>=2019.7.26
  Using cached tifffile-2020.9.3-py3-none-any.whl (148 kB)
Collecting PyWavelets>=1.1.1
  Using cached PyWavelets-1.1.1-cp36-cp36m-manylinux2014_aarch64.whl (4.3 MB)
Collecting networkx>=2.0
  Using cached networkx-2.5.1-py3-none-any.whl (1.6 MB)
Collecting scipy
  Using cached scipy-1.5.4-cp36-cp36m-manylinux2014_aarch64.whl (24.5 MB)
Collecting decorator<5,>=4.3
  Using cached decorator-4.4.2-py2.py3-none-any.whl (9.2 kB)
Installing collected packages: Pillow, decorator, tifffile, scipy, PyWavelets, networkx, imageio, Shapely, scikit-image, opencv-python, imgaug
  Attempting uninstall: decorator
    Found existing installation: decorator 4.1.2
    Not uninstalling decorator at /usr/lib/python3/dist-packages, outside environment /home/ed/projects/CSI-Camera/env
    Can't uninstall 'decorator'. No files were found to uninstall.
  Attempting uninstall: scipy
    Found existing installation: scipy 0.19.1
    Not uninstalling scipy at /usr/lib/python3/dist-packages, outside environment /home/ed/projects/CSI-Camera/env
    Can't uninstall 'scipy'. No files were found to uninstall.
Successfully installed Pillow-8.4.0 PyWavelets-1.1.1 Shapely-1.8.0 decorator-4.4.2 imageio-2.13.3 imgaug-0.4.0 networkx-2.5.1 opencv-python-4.5.4.60 scikit-image-0.17.2 scipy-1.5.4 tifffile-2020.9.3
(env) ed@diydonkeybot:~/projects/CSI-Camera$ python simple_camera.py 
nvarguscamerasrc ! video/x-raw(memory:NVMM), width=(int)1280, height=(int)720, format=(string)NV12, framerate=(fraction)60/1 ! nvvidconv flip-method=0 ! video/x-raw, width=(int)1280, height=(int)720, format=(string)BGRx ! videoconvert ! video/x-raw, format=(string)BGR ! appsink
Unable to open camera

Ezward · 2021-12-20T19:35:42Z

Here's a twist; I can reproduce the bug without installing imgaug. Just install opencv-python:

$ cd ~/projects/CSI-Camera
$ rm -rf env
$ python3 -m virtualenv -p python3 env --system-site-packages
created virtual environment CPython3.6.9.final.0-64 in 1169ms
  creator CPython3Posix(dest=/home/ed/projects/CSI-Camera/env, clear=False, no_vcs_ignore=False, global=True)
  seeder FromAppData(download=False, pip=bundle, setuptools=bundle, wheel=bundle, via=copy, app_data_dir=/home/ed/.local/share/virtualenv)
    added seed packages: pip==21.3.1, setuptools==59.2.0, wheel==0.37.0
  activators BashActivator,CShellActivator,FishActivator,NushellActivator,PowerShellActivator,PythonActivator
$ . env/bin/activate
$ python simple_camera.py 
nvarguscamerasrc ! video/x-raw(memory:NVMM), width=(int)1280, height=(int)720, format=(string)NV12, framerate=(fraction)60/1 ! nvvidconv flip-method=0 ! video/x-raw, width=(int)1280, height=(int)720, format=(string)BGRx ! videoconvert ! video/x-raw, format=(string)BGR ! appsink
GST_ARGUS: Creating output stream
CONSUMER: Waiting until producer is connected...
GST_ARGUS: Available Sensor modes :
GST_ARGUS: 3264 x 2464 FR = 21.000000 fps Duration = 47619048 ; Analog Gain range min 1.000000, max 10.625000; Exposure Range min 13000, max 683709000;

GST_ARGUS: 3264 x 1848 FR = 28.000001 fps Duration = 35714284 ; Analog Gain range min 1.000000, max 10.625000; Exposure Range min 13000, max 683709000;

GST_ARGUS: 1920 x 1080 FR = 29.999999 fps Duration = 33333334 ; Analog Gain range min 1.000000, max 10.625000; Exposure Range min 13000, max 683709000;

GST_ARGUS: 1640 x 1232 FR = 29.999999 fps Duration = 33333334 ; Analog Gain range min 1.000000, max 10.625000; Exposure Range min 13000, max 683709000;

GST_ARGUS: 1280 x 720 FR = 59.999999 fps Duration = 16666667 ; Analog Gain range min 1.000000, max 10.625000; Exposure Range min 13000, max 683709000;

GST_ARGUS: 1280 x 720 FR = 120.000005 fps Duration = 8333333 ; Analog Gain range min 1.000000, max 10.625000; Exposure Range min 13000, max 683709000;

GST_ARGUS: Running with following settings:
   Camera index = 0 
   Camera mode  = 5 
   Output Stream W = 1280 H = 720 
   seconds to Run    = 0 
   Frame Rate = 120.000005 
GST_ARGUS: Setup Complete, Starting captures for 0 seconds
GST_ARGUS: Starting repeat capture requests.
CONSUMER: Producer has connected; continuing.
[ WARN:0] global /home/nvidia/host/build_opencv/nv_opencv/modules/videoio/src/cap_gstreamer.cpp (933) open OpenCV | GStreamer warning: Cannot query video position: status=0, value=-1, duration=-1
^CTraceback (most recent call last):
  File "simple_camera.py", line 66, in <module>
    show_camera()
  File "simple_camera.py", line 55, in show_camera
    keyCode = cv2.waitKey(30) & 0xFF
KeyboardInterrupt
GST_ARGUS: Cleaning up
CONSUMER: Done Success
GST_ARGUS: Done Success

$ pip install opencv-python
Collecting opencv-python
  Using cached opencv_python-4.5.4.60-cp36-cp36m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl (39.0 MB)
Requirement already satisfied: numpy>=1.19.3 in /usr/local/lib/python3.6/dist-packages (from opencv-python) (1.19.4)
Installing collected packages: opencv-python
Successfully installed opencv-python-4.5.4.60

$ python simple_camera.py 
nvarguscamerasrc ! video/x-raw(memory:NVMM), width=(int)1280, height=(int)720, format=(string)NV12, framerate=(fraction)60/1 ! nvvidconv flip-method=0 ! video/x-raw, width=(int)1280, height=(int)720, format=(string)BGRx ! videoconvert ! video/x-raw, format=(string)BGR ! appsink
Unable to open camera

So it seems the issue is with the camera driver, since it is the one that needs opencv. It might even by one of it's dependencies (like argus_camera or V4L).

Ezward · 2021-12-20T20:05:33Z

I've been able to repeat this bug with the base Jetpack 4.6 python which is python3.6.9, and with python3.7.5 and python3.8.0

Ezward · 2021-12-20T20:29:02Z

It works 'out-of-the-box' with no virtual environment, so this is what it is using for opencv:

ed@diydonkeybot:~/projects/CSI-Camera$ python3
Python 3.6.9 (default, Dec  8 2021, 21:08:43) 
[GCC 8.4.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import cv2
>>> print(cv2)
<module 'cv2' from '/usr/lib/python3.6/dist-packages/cv2/python-3.6/cv2.cpython-36m-aarch64-linux-gnu.so'>
>>> print(cv2.__version__)
4.1.1

Ezward · 2021-12-20T20:29:36Z

this works in the virtual environment is we don't try to install opencv-python

ed@diydonkeybot:~/projects/CSI-Camera$ rm -rf env
ed@diydonkeybot:~/projects/CSI-Camera$ python3 -m virtualenv -p python3 env --system-site-packages
created virtual environment CPython3.6.9.final.0-64 in 1291ms
  creator CPython3Posix(dest=/home/ed/projects/CSI-Camera/env, clear=False, no_vcs_ignore=False, global=True)
  seeder FromAppData(download=False, pip=bundle, setuptools=bundle, wheel=bundle, via=copy, app_data_dir=/home/ed/.local/share/virtualenv)
    added seed packages: pip==21.3.1, setuptools==59.2.0, wheel==0.37.0
  activators BashActivator,CShellActivator,FishActivator,NushellActivator,PowerShellActivator,PythonActivator
ed@diydonkeybot:~/projects/CSI-Camera$ . env/bin/activate
(env) ed@diydonkeybot:~/projects/CSI-Camera$ python --version
Python 3.6.9
(env) ed@diydonkeybot:~/projects/CSI-Camera$ python simple_camera.py 
nvarguscamerasrc ! video/x-raw(memory:NVMM), width=(int)1280, height=(int)720, format=(string)NV12, framerate=(fraction)60/1 ! nvvidconv flip-method=0 ! video/x-raw, width=(int)1280, height=(int)720, format=(string)BGRx ! videoconvert ! video/x-raw, format=(string)BGR ! appsink
GST_ARGUS: Creating output stream
CONSUMER: Waiting until producer is connected...
GST_ARGUS: Available Sensor modes :
GST_ARGUS: 3264 x 2464 FR = 21.000000 fps Duration = 47619048 ; Analog Gain range min 1.000000, max 10.625000; Exposure Range min 13000, max 683709000;

GST_ARGUS: 3264 x 1848 FR = 28.000001 fps Duration = 35714284 ; Analog Gain range min 1.000000, max 10.625000; Exposure Range min 13000, max 683709000;

GST_ARGUS: 1920 x 1080 FR = 29.999999 fps Duration = 33333334 ; Analog Gain range min 1.000000, max 10.625000; Exposure Range min 13000, max 683709000;

GST_ARGUS: 1640 x 1232 FR = 29.999999 fps Duration = 33333334 ; Analog Gain range min 1.000000, max 10.625000; Exposure Range min 13000, max 683709000;

GST_ARGUS: 1280 x 720 FR = 59.999999 fps Duration = 16666667 ; Analog Gain range min 1.000000, max 10.625000; Exposure Range min 13000, max 683709000;

GST_ARGUS: 1280 x 720 FR = 120.000005 fps Duration = 8333333 ; Analog Gain range min 1.000000, max 10.625000; Exposure Range min 13000, max 683709000;

GST_ARGUS: Running with following settings:
   Camera index = 0 
   Camera mode  = 5 
   Output Stream W = 1280 H = 720 
   seconds to Run    = 0 
   Frame Rate = 120.000005 
GST_ARGUS: Setup Complete, Starting captures for 0 seconds
GST_ARGUS: Starting repeat capture requests.
CONSUMER: Producer has connected; continuing.
[ WARN:0] global /home/nvidia/host/build_opencv/nv_opencv/modules/videoio/src/cap_gstreamer.cpp (933) open OpenCV | GStreamer warning: Cannot query video position: status=0, value=-1, duration=-1
^CTraceback (most recent call last):
  File "simple_camera.py", line 66, in <module>
    show_camera()
  File "simple_camera.py", line 52, in show_camera
    ret_val, img = cap.read()
KeyboardInterrupt
GST_ARGUS: Cleaning up
CONSUMER: Done Success
GST_ARGUS: Done Success
(env) ed@diydonkeybot:~/projects/CSI-Camera$ python
Python 3.6.9 (default, Dec  8 2021, 21:08:43) 
[GCC 8.4.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import cv
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
ModuleNotFoundError: No module named 'cv'
>>> import cv2
>>> print(cv2.__version__)
4.1.1

Ezward · 2021-12-20T22:22:05Z

Thanks @TCIII for finding this issue in the imgaug github, aleju/imgaug#737 and aleju/imgaug#473

That indicated that the pip wheel install of imgaug will install opencv-python. My own testing above shows this will conflict with the Jetpack opencv and will cause the camera error.

So I'm going to trying installing from source as indicated in that imgaug issue.

pip install imgaug --no-binary imgaug

That did not work either; it still installed opencv-python-headless

Ezward · 2022-01-09T20:04:13Z

The maintainer of the imgaug library has said they will not fix their installer. So we should implement our augmentations another way. There is an alternative library that used to use imgaug but has removed it because of this bug. It may be a good replacement for us https://github.com/albumentations-team/albumentations#list-of-augmentations

- this is a workaround for issue 970 - #970 - it is meant to be temporary until we refactor the image augmentations to not use imgaug

* Mock ImageAugmentation if imgaug not installed - this is a workaround for issue 970 - #970 - it is meant to be temporary until we refactor the image augmentations to not use imgaug * Cleanup linting issues * version="4.3.4"

Ezward · 2022-01-15T05:49:12Z

PR #978 is a temporary workaround (read HACK) that mocks out the augmentations if imgaug dependency is not installed. It is not installed by default on the Nano. So this will prevent the camera issue. It warns the user at runtime if they use a configuration that utilized augmentations.

This temporary hack should be removed when we replace the imaug library with another implemenation.

naisy · 2022-01-25T12:58:33Z

EDIT: 2022/06/14 Corrected a wrong option.

@Ezward ~~You can uninstall opencv-python after installing imgaug 0.4.0~~
use --no-deps option.

In my Dockerfile:

########################################
# imgaug 0.4.0
########################################
RUN . /virtualenv/$VENV/bin/activate \
    && pip install -U \
      six \
      scipy \
      scikit-image \
      imageio \
      Shapely

RUN . /virtualenv/$VENV/bin/activate \
    && pip install --no-deps \
      imgaug \
      imagecorruptions

Ezward · 2023-03-08T21:21:51Z

see PR #1111

Ezward · 2023-03-22T03:39:48Z

Replace imgaug with augmentations. PR #1111

Ezward changed the title ~~Installing imgaug on Jetson Nano break CSIC camera~~ Installing imgaug on Jetson Nano breaks CSIC camera Dec 17, 2021

Ezward mentioned this issue Dec 18, 2021

Protect LineFollower from None cam_img #961

Closed

12 tasks

Ezward added a commit that referenced this issue Jan 10, 2022

Mock ImageAugmentation if imgaug not installed

462932c

- this is a workaround for issue 970 - #970 - it is meant to be temporary until we refactor the image augmentations to not use imgaug

Ezward mentioned this issue Jan 10, 2022

970 imgaug workaround mock augmentations #978

Merged

This was referenced Mar 4, 2023

Refactor Image Augmentations/Transformation to use parts #1109

Closed

How can I train my donkeycar on jetson nano #783

Open

1109 refactor augmentations/transformations #1111

Merged

Ezward self-assigned this Mar 8, 2023

Ezward closed this as completed Mar 22, 2023

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Installing imgaug on Jetson Nano breaks CSIC camera #970

Installing imgaug on Jetson Nano breaks CSIC camera #970

Ezward commented Dec 17, 2021 •

edited

Loading

Ezward commented Dec 17, 2021

Ezward commented Dec 20, 2021

Ezward commented Dec 20, 2021

Ezward commented Dec 20, 2021 •

edited

Loading

Ezward commented Dec 20, 2021

Ezward commented Dec 20, 2021

Ezward commented Dec 20, 2021

Ezward commented Dec 20, 2021 •

edited

Loading

Ezward commented Jan 9, 2022

Ezward commented Jan 15, 2022

naisy commented Jan 25, 2022 •

edited

Loading

Ezward commented Mar 8, 2023

Ezward commented Mar 22, 2023

Installing imgaug on Jetson Nano breaks CSIC camera #970

Installing imgaug on Jetson Nano breaks CSIC camera #970

Comments

Ezward commented Dec 17, 2021 • edited Loading

Ezward commented Dec 17, 2021

Ezward commented Dec 20, 2021

Ezward commented Dec 20, 2021

Ezward commented Dec 20, 2021 • edited Loading

Ezward commented Dec 20, 2021

Ezward commented Dec 20, 2021

Ezward commented Dec 20, 2021

Ezward commented Dec 20, 2021 • edited Loading

Ezward commented Jan 9, 2022

Ezward commented Jan 15, 2022

This temporary hack should be removed when we replace the imaug library with another implemenation.

naisy commented Jan 25, 2022 • edited Loading

Ezward commented Mar 8, 2023

Ezward commented Mar 22, 2023

Ezward commented Dec 17, 2021 •

edited

Loading

Ezward commented Dec 20, 2021 •

edited

Loading

Ezward commented Dec 20, 2021 •

edited

Loading

naisy commented Jan 25, 2022 •

edited

Loading