The first experiment was using the 2.12 training datasets.
We added the Family information to each selected observations.
This observation selection was done at the species level, using the species threshold.
In this second experiment, we would like to gather more taxa for each family, even if these taxa only have a few photos.
We could use the existing script with a lower "species threshold" to generate this new training dataset
The first experiment was using the 2.12 training datasets. We added the Family information to each selected observations. This observation selection was done at the species level, using the species threshold.
In this second experiment, we would like to gather more taxa for each family, even if these taxa only have a few photos.
We could use the existing script with a lower "species threshold" to generate this new training dataset