ZexinChen / AlphaTracker

AlphaTracker is a computer vision pipeline with the practical and real-time advantages , which requires minimal hardware requirements and produces reliable tracking of multiple unmarked animals. An easy-to-use user interface further enables manual inspection and curation of results.
66 stars 16 forks source link

failed to create symbolic link and import cv2 #16

Open AZAintheLab opened 2 years ago

AZAintheLab commented 2 years ago

When using the google colab version I get the following error at Step 8

ln: failed to create symbolic link '/gdrive/My Drive': Input/output error /content/drive/My Drive/AlphaTracker/Tracking/AlphaTracker Traceback (most recent call last): File "trainCOLAB.py", line 7, in import cv2 ModuleNotFoundError: No module named 'cv2'

I've checked to ensure that the folder was correctly added to My Drive

laurelrr commented 2 years ago

Hi AZAintheLab,

I was able to recreate this issue. I think there are 2 separate errors happening in this one step. We'll try these modifications to the notebook before running any cells and then run all cells at the end.

  1. Please enter a new code cell above step 8 with this information:

    !ln -s "/content/drive/My Drive" /gdrive
    %cd /gdrive/AlphaTracker/Tracking/AlphaTracker
    !ls

    That should correctly set up the symbolic link to the AlphaTracker directory. It should not have an error the first time you run it. (If you run it a second time, however, you should get the message "ln: failed to create symbolic link '/gdrive/My Drive': Input/output error".) The !ls step is there because you should see it print out the contents of the AlphaTracker directory.

  2. Add another new code cell with this information:

    !pip install tqdm==4.31.1
    !pip install h5py==2.9.0 
    !pip install opencv-python==4.0.0.21
    !python -c "print('Starting cv2 test'); import cv2; print(cv2.__version__);"

    It should install packages and return some comments ending with: "Successfully installed opencv-python-4.0.0.21 Starting test 4.0.0" "

  3. Next, comment out the first line in Step 8 like so:

    #!ln -s "/content/drive/My Drive" /gdrive
  4. Select Runtime -> Change Runtime type -> GPU

  5. Select Runtime -> Run all

Let me know if this fixes the issue. If it works, you can save a copy of the notebook to your local drive so you don't have to make changes again.

laurelrr commented 2 years ago

Actually, it looks like some of our requirements for colab are out of date with the current repo. I ran into an issue when I went to track. I will need to update our colab notebook to reflect the changes.

AZAintheLab commented 2 years ago

OK! So I'll wait till its updated. Then I'll try again 🙏🏿

mcarone001 commented 2 years ago

Actually, it looks like some of our requirements for colab are out of date with the current repo. I ran into an issue when I went to track. I will need to update our colab notebook to reflect the changes.

Hi, did you manage to update the colab notebook?

laurelrr commented 2 years ago

Thanks for your patience. The colab notebook has been updated!

I tested it with the sample data and it is working. The one hitch you might run into is automatically downloading the data files and our pre-trained models. If you have a problem, please verify that the data has downloaded correctly. If not, please follow the instructions here to manually download each file and place in the appropriate place in your google drive.

The updated colab notebook is here: https://colab.research.google.com/drive/1qzl-lQC59di5v2jrSARtZfR2JlOIdCgc?usp=sharing