Training multiple objects for same class

[swastiksomani]

We want to detect multiple object of same class from a single given image. We have the following queries:

Q1: Can we implement this problem statement using the current framework?

Q2: How do we train the model for this problem statement. Do we need images/labels.txt with multiple same class objects?

Q3: What parameters we need to change in cfg files?

Q4: Do we need to make some changes in valid.ipynb file?

[TrinhNC]

Swastik Somani (@swastiksomani) any update on this? I'm wondering if we have to use the multi training code for this task or use the single object one?