juglab / n2v

This is the implementation of Noise2Void training.
Other
385 stars 107 forks source link

How to run through the code #142

Closed tianshukai99 closed 10 months ago

tianshukai99 commented 11 months ago

Hello, I have some questions for you.

  1. i have 384*288 infrared images now, what should i use to train with my data? Should I use trainN2V.py or which one? 2.Which folder do I put my data in when training, and what is the exact process to run through the code?
jdeschamps commented 11 months ago

Hi,

  1. I would suggest you use one of the jupyter notebook, e.g. this one. You will need to adapt the loading data part to your own images. If you are not comfortable with writing python code, then you can use the napari plugin.
  2. In the notebooks, the data is loaded in memory so you need to simply point to the right folder and then use a function to load the data. If you have tiff images, then you could use the tifffile package to load the images. Then you run each cell of the notebook one by one and make sure that you adapt them to your data.

If you don't know how to use jupyter notebooks, I suggest this tutorial. Note that it still requires being able to write python code. If you are not comfortable with programming in python, then you can use the napari plugin. It requires setting up the conda environment properly, the rest is through a user interface.

jdeschamps commented 10 months ago

I will close this issue due to lack of answer, don't hesitate to comment on it if you want it reopen or to open a new one.