1) Using only CPUIndexer.index succeeds and adding new texts via IndexUpdater.add also succeeds (expected)
2) Using GPUIndexer.index succeeds but adding new texts via IndexUpdater.add fails in the ninja call reporting Error building extension 'segmented_lookup_cpp' coming from the load call here
Is there perhaps a specific version of ninja we need in order for the IndexUpdater to initialize correctly? Or are there different versions of the CUDA libraries needed for the IndexUpdater to work correctly?
We are working with ColBERT at https://github.com/stanford-futuredata/ColBERT/commit/89ab526a08db475e7843487bb87259630b5e4d7d, and we are seeing errors when attempting to use IndexUpdater on GPU.
1) Using only CPU
Indexer.index
succeeds and adding new texts viaIndexUpdater.add
also succeeds (expected) 2) Using GPUIndexer.index
succeeds but adding new texts viaIndexUpdater.add
fails in theninja
call reportingError building extension 'segmented_lookup_cpp'
coming from theload
call hereWe see this using the libraries defined in https://github.com/c3-e/ColBERT/blob/89ab526a08db475e7843487bb87259630b5e4d7d/conda_env.yml
Is there perhaps a specific version of
ninja
we need in order for the IndexUpdater to initialize correctly? Or are there different versions of the CUDA libraries needed for the IndexUpdater to work correctly?