I don't understand how the batching work in the Mask Generator or the Predictor

[bely66]

As I can see in the Blog Post:

1. You're using different batch sizes of 1, 8, 32
2. After checking the code: it seems that the batching isn't on the image and on the points

Which I don't comprehend why would the points need batching?

[cpuhrsch]

Hello @bely66 - Thank you for your interest in the project! There's a different number of points per image. See https://github.com/pytorch-labs/segment-anything-fast/blob/main/experiments/README.md#data .

[bely66]

Hello @bely66 - Thank you for your interest in the project! There's a different number of points per image. See https://github.com/pytorch-labs/segment-anything-fast/blob/main/experiments/README.md#data .

@cpuhrsch thank you for the timely reply

The documentation doesn't mention anything about batching or how it works