Since commit b78ff782c4e92a7a77a3598bfe1d0e859d31149b, samples_idx parameter apparently have no effect.
Reverting this commit recover the expected behavior that a single sample out of many samples can be processed, allowing
parallel computing over many nodes on a cluster. With current state, all the parallel process executes from index 1 and thus disrupting the file each other to make the abundance estimation process fail.
The reason of removal here is not very clear to me; there might have been some necessity for debugging.
I hope the function recovered or removed from the help to make them coherent with the documentation.
Since commit b78ff782c4e92a7a77a3598bfe1d0e859d31149b, samples_idx parameter apparently have no effect. Reverting this commit recover the expected behavior that a single sample out of many samples can be processed, allowing parallel computing over many nodes on a cluster. With current state, all the parallel process executes from index 1 and thus disrupting the file each other to make the abundance estimation process fail. The reason of removal here is not very clear to me; there might have been some necessity for debugging. I hope the function recovered or removed from the help to make them coherent with the documentation.