Closed danilobenozzo closed 1 day ago
The tests also show that somehow you're now placing transmitters with duplicate GIDs. I think more tests are the only way out of this precarious mess.
If you have any deadlines that depend on this code, perhaps you can use a fork with your previous code (which seemed to work? At least it didn't break the existing tests) and then at a later point you can finish this PR with the optimal implementation.
@filimarc @danilobenozzo could you add at least add a couple of test cases that test correct GID assigment and transmitter placement for single/multiple conn models with overlapping cell types etc?
yeah.. new config files are not being found. I added them https://github.com/dbbs-lab/bsb-test/pull/7#issue-2445315781 isn't this enough?
Hi, i will publish the bsb-test package with your new config so github can dowload it with pip. Btw i tried to run test on my PC but it fails the 4chunck test. with mpirun -n 2 python -m unittest discover -v -s ./tests
:
@filimarc ok now it should work
@filimarc please use code formatting: https://docs.github.com/en/get-started/writing-on-github/working-with-advanced-formatting/creating-and-highlighting-code-blocks
For each connectivity set, kept track of GIDs grouped by pre-synaptic cell type. If a pre-synaptic cell already has a GID for [cell, branch], it is re-used in the case of multiple synapses.