In this repository, we've implemented the code resulting in the paper "Deep regression model of the distance transform for the alpaca fiber diameter measurement". the code in this repository is made with PyTorch.
The data was generated with https://github.com/alainno/fibergen.git and is located at folder: ./data_dm_overlapping
.
Execute the following command: python train_all.py
. This command will train U-Net regression and SkeletonNet regression.
Test are run on notebooks:
- Fiber diameter Mean and Std: test_mean_std.ipynb
- Measurement error: test_measurement_error.ipyb