Examples of medical image segmentation using publically available annoymized data from Kaggle and other online data sources
- I suspect some of my issues around
TypeError: new() missing 1 required positional argument: 'x'
that I get when I run my micro-CT segmentation notebook arise from improper creation of my dataset, including how to convert floats to ints and save as a file and then reload retaining value. - This is likely the case because when I use premade datasets like mnist I can train on my cloud machine, but when I use a homemade dataset, I get the error (even though my custom dataset works on my local PC!)
- Hopefully if I follow other examples of making datasets I can overcome these challenges.
- Cites this notebook: dicom jpg