djimenezsanchez / NaroNet

Trained only with subject-level labels, NaroNet discovers phenotypes, neighborhoods, and areas with the highest influence when classifying subject types.
GNU Affero General Public License v3.0
15 stars 6 forks source link

how to define the interaction(attraction/repulsion) rules btw. phnotypes or btw. neighbors? #3

Closed inyeopjang closed 1 year ago

inyeopjang commented 1 year ago

To train the model, how the labels (phenotypes, neighbors, interactions) could be defined in the raw dataset?

djimenezsanchez commented 1 year ago

Hi,

In this model, phenotypes, neighborhoods, and their interactions (i.e., areas) are automatically learned from the images and the patient-level labels. However, the number of phenotypes, neighborhoods, and areas could be changed. To this end, change the parameters in DatasetParameters.py file, where the lines that must be changed are the following:

Best, Daniel